Let's Talk About Jockeys

[Segwin]

Presently reading the book " The New Pace Makes the Race". Only into it a short ways so far but the the author goes on to say that the jockey really doesn't make much of a difference and for the first part of the race they're just trying to hang on. I would imagine that he feels the same way about trainer stats but only speculation so far.

How does a jockey and horse get paired (I know, noob question)? If the jockey doesn't make any difference then why is there such a buzz regarding the top earners?

Same goes for trainer although I would think that the trainer and stable has a lot more affect on the horse than the jockey.

I always look at the stats for the jockey and trainer. I'd be more willing to bet on a so-so horse with a great jockey than a better horse and a jockey with poor stats.

Interested in the thoughts of others.

[WeirdWilly]

A good jockey on a bad horse vs a so so jockey on a great horse is probably still not going to win. But with roughly equal horses, how can it NOT be at least somewhat of a factor?

The jockey is the one who steers the horse into a particular path. They have to know when to hold back and when to cut loose. And, as living, breathing, aware animals, I am sure horses have people they like and people they don't.

Personally, I always look at a jockey's past experience on a particular horse more than their overall record.

[rubicon55]

A good jockey on a bad horse vs a so so jockey on a great horse is probably still not going to win. But with roughly equal horses, how can it NOT be at least somewhat of a factor?

The jockey is the one who steers the horse into a particular path. They have to know when to hold back and when to cut loose. And, as living, breathing, aware animals, I am sure horses have people they like and people they don't.

Personally, I always look at a jockey's past experience on a particular horse more than their overall record.

I agree with WW, also when looking at jocks I mandate they have at 15% winning percentage, not to say the low percentage jocks do well on good mounts but as a rule this works. Start with the trainer win pecentage first and see what trainer/jockey combos have the best win percentage together. But as always start with the horse first.

[DeltaLover]

I think that it is fairly accurate to assume, that jockeys indeed can make a difference since their success depends to several skills, abilities such as athetism, natural talent, experience and of course life style is part of the equation too.

Despite this though, the crowd is easily misled by primitive stats published by DRF and several others. Merely reporting a jock as a 5% winner does not necessary means that he is inferior to a 15% or at least the magnitude of their ability difference is certainly not reflected by this figures.

A better metric of a jockey's ability that I have used in the past is its impact value only with favorites, a concept that can be generalized by comparing the chi squares in various odd categories. Following this approach for long time, I have concluded that very rare are the cases where the ability of a certain jockey is so important to create over or underlays.

[proximity]

dave,

what kind of jockey information does your percentages and probabilities book give??

proximity

[speculus]

IMO, if you classify factors* contributing to a horse's most optimum performance in two groups, ESSENTIAL & DESIRABLE, then I would list them in the following manner:

ESSENTIAL:
1) Suitability for the trip (distance)
2) CLASS
3) CONDITION (racing fitness)

DESIRABLE:
1) Jockey SKILL (& his/her current) FORM
2) Underfoot condition (track status)
3) Weight carried

If a horse even moderately qualifies on all the essential factors, then the jockey becomes the MOST DECISIVE factor, IMO.

*The impact of wind (or what they call 'drag' in physics) is discounted.

[speed]

Jockey can make a huge difference the first part of a race. From getting out of the gate to judging the early pace. Plenty of races are lost early in the race.

[Robert Fischer]

How does a jockey and horse get paired (I know, noob question)? I

Jockey's have agents, and top trainers develop a relationship with the best jockey they can, who will ride first-call for that trainer. After top trainers and owners, Agents may handicap the race and find the best business decision.

Overall and in general, the jockey is more of an indication of the trainer and connections, than any kind of factor.

Specific circumstances (which are sometimes very hard to predict) call for a certain type of ride, or the versatility to read the pace on the fly. The best excel in these instances, the worst are incompetent.

[Stillriledup]

Presently reading the book " The New Pace Makes the Race". Only into it a short ways so far but the the author goes on to say that the jockey really doesn't make much of a difference and for the first part of the race they're just trying to hang on. I would imagine that he feels the same way about trainer stats but only speculation so far.

How does a jockey and horse get paired (I know, noob question)? If the jockey doesn't make any difference then why is there such a buzz regarding the top earners?

Same goes for trainer although I would think that the trainer and stable has a lot more affect on the horse than the jockey.

I always look at the stats for the jockey and trainer. I'd be more willing to bet on a so-so horse with a great jockey than a better horse and a jockey with poor stats.

Interested in the thoughts of others.

I don't want to go as far as saying Jockey means everything, but its a much bigger part of the equasion than most people think. Many races and board spots are decided by 1/100th of a second, and, in a 6 furlong race that goes in 110, that's 70 seconds times 100 that equals 7,000 'moments' during a race and each one of them can matter when you're sweating out a 10 minute print photo. Jocks are athletes and as we all know, some athletes are better than others, the difference between Joel Rosario and the worst jock at the lowest rung track is a monumental margin.

[burnsy]

I don't want to go as far as saying Jockey means everything, but its a much bigger part of the equasion than most people think. Many races and board spots are decided by 1/100th of a second, and, in a 6 furlong race that goes in 110, that's 70 seconds times 100 that equals 7,000 'moments' during a race and each one of them can matter when you're sweating out a 10 minute print photo. Jocks are athletes and as we all know, some athletes are better than others, the difference between Joel Rosario and the worst jock at the lowest rung track is a monumental margin.

True, but the difference between a slower horse and a faster horse....means so much more, and the difference in any given jockey colony really is not as big as people make it. This is also a game of politics, agents and who knows who. He who gets the fastest horses wins the most races. Thats why the author writes he doesn't care about jocks that much. The horse itself means so much more and so does the trainer. Besides everyone knows the top jocks and people bet the jocks whether they can handicap or not. Sometimes the best plays are the under the radar jocks on capable horses, this works really well at saratoga because there are so many people there betting the jockeys. They are on vacation, can't really read the form....believe me theres people that think the jockey can make a horse faster...........99.9% of the time, its not true. Will it sometimes make a difference???? Of course, but sticking to horses with the POTENTIAL to run well and win is light years more important than who is riding.

[Stillriledup]

True, but the difference between a slower horse and a faster horse....means so much more, and the difference in any given jockey colony really is not as big as people make it. This is also a game of politics, agents and who knows who. He who gets the fastest horses wins the most races. Thats why the author writes he doesn't care about jocks that much. The horse itself means so much more and so does the trainer. Besides everyone knows the top jocks and people bet the jocks whether they can handicap or not. Sometimes the best plays are the under the radar jocks on capable horses, this works really well at saratoga because there are so many people there betting the jockeys. They are on vacation, can't really read the form....believe me theres people that think the jockey can make a horse faster...........99.9% of the time, its not true. Will it sometimes make a difference???? Of course, but sticking to horses with the POTENTIAL to run well and win is light years more important than who is riding.

That's what great about this game, we all can look at the same thing and see something completely different. :ThmbUp:

[Capper Al]

I like to watch jockey changes and jockeys riding for top trainers. A returning jockey who has previously won on a horse always gets my interest.

[castaway01]

To me, while you do get the occasional hot streak where a rider can do no wrong, a bad jockey (or a poor ride by a good jockey) gets the fastest horse in the race beat a lot more often than a great jockey holds an inferior horse together and somehow beats the faster horse (no matter who is riding). You'd probably be better off, financially, with a mid-level rider who can get the horse home in one piece because everyone bets the top rider at the track anyway.

[Stillriledup]

To me, while you do get the occasional hot streak where a rider can do no wrong, a bad jockey (or a poor ride by a good jockey) gets the fastest horse in the race beat a lot more often than a great jockey holds an inferior horse together and somehow beats the faster horse (no matter who is riding). You'd probably be better off, financially, with a mid-level rider who can get the horse home in one piece because everyone bets the top rider at the track anyway.

This is a great post 01, i agree with this general premise. I love digging out a competent rider who doesnt get many chances, but can really ride when he does, you get great value that way, if you have a good horse to play, you don't need the leading rider, they pay off the same amount whether you win by a nose or win by 5 so if you're much the best, its better to take the higher price and hope that your calculations are correct on the horse.

[Tom]

I break my jockey stats into catagories - dirt sprints, routes, turf sprints, routes, poly sprints, routes, then by early speed. Then even more by specifics, like SA 6.5F races, Saratoga 5.5 Turf sprints, etc.

[Dave Schwartz]

question for dave schwartz????
dave,

what kind of jockey information does your percentages and probabilities book give??

proximity

Oh, that is perfect! I am going to do an article on that for next week's show!

BTW, that show is Wednesday @5:30pm (pacific) instead of Thursday because of July 4th.

Sign up here (https://www2.gotomeeting.com/register/225284506)

Dave

[MJC922]

Rider makes a big difference an awful lot of the time. A good example last night in the 8th at PEN. 2-turn flat mile, heavy chalk is breaking from post 7 with just enough early foot drawn inside him to expect a wide chasing trip. The two horse Sixsixteen showed he was back in form last time after a couple of mini-layoffs, now back in just two weeks figured to fire another good effort, importantly he's tied with the 7 for best early pace. My big fear was the rail horse though, Socio d'Oro, had back class, also had essentially the same early pace figure as Sixsixteen did so the way I saw it, this was shaping up to be a 'riders race'. Now I noticed Badamo is riding Socio d'Oro, hmm, well I don't follow PEN at all so I didn't know much about Belmonte's plans on Sixsixteen, however I have about a decade of close observation under my belt at FL during the 1990s and I know Badamo is far more likely to rate a horse early than he his to send hard out of there. So my money goes on Sixsixteen at 4-1 as they load up, but this is every bit as much a bet that Badamo will rate and let Sixsixteen go on with it. Gate pops and both of them break sharp, but Badamo rates, Belmonte more aggressive, exited the first turn on a clear lead, perfectly executed, game over, go cash. Flip the riders and maybe you have a different result. Many races are decided in the first quarter mile on tactics, it can pay to know rider tendencies.

[Capper Al]

If the jockey doesn't matter then the trainers wouldn't be so concerned about getting good jockeys for their prize stake horses.

[aaron]

Presently reading the book " The New Pace Makes the Race". Only into it a short ways so far but the the author goes on to say that the jockey really doesn't make much of a difference and for the first part of the race they're just trying to hang on. I would imagine that he feels the same way about trainer stats but only speculation so far.

How does a jockey and horse get paired (I know, noob question)? If the jockey doesn't make any difference then why is there such a buzz regarding the top earners?

Same goes for trainer although I would think that the trainer and stable has a lot more affect on the horse than the jockey.

I always look at the stats for the jockey and trainer. I'd be more willing to bet on a so-so horse with a great jockey than a better horse and a jockey with poor stats.

Interested in the thoughts of others.
is the book you are referring to by Tom Hambelton ? In my opinion,jockey changes from a bad rider to a top rider can be the most important factor in making a bet in certain circumstances.

[burnsy]

This is IMO a mistake people make.......you have to think "big" picture or,.....will you be betting next week, next month, next year? Most times i could care less who the jock is. Its all about the chance the HORSE has relative to its odds. Everyone loses tough beats by a nose here or a kneck there....if you plan on betting long term. It comes with the game. The jockey matters, but not to the point people make it. The Pen example MCJ922 uses is good, i would never bet that track, but if i were i'd most likely be betting the 4-1 shot anyway, simply because its better than 6-5. Thats the point this author and many handicappers are trying to make. The "name" jockeys get the most chalk horses, they win 1 out of 3....are you going to leave all that money out there??? Just to bet a short horse with the better jock? When the race is about 95% how fast the horse can run, THAT DAY? In the big races every jock there is "world class" with the exception of the derby because there are at times 20 horses. You don't even get a sniff around NY in the summer unless you are a pretty damn good rider. You have to realize that i only bet the major tracks anyway. Anyone of these guys or gals are good enough to win at ANY TIME with a capable horse. Are there some i favor over others? Of course. But on my list of how i pick a horse, jocks is near the bottom of that list. I want the horse thats in form, has the class and distance capability and most IMPORTANTLY IS NOT OVERBET. In a race i plan on betting its usually down to 3 or so and if i'm betting next week, next year, next decade......it usually means the horse i think is the BEST PRICE. Some trainers don't just go for the name jock....Jerkens....the "Giant Killer"....ever heard of him? Won alot of races and the jock was usually someone he could tell what to do....he trusts his training ability and his horse. Are some at the top a little better? Yeah. But what are you going to do? Bet chalk and overbet horses all day? Not me. I have to be able to afford to come back tommorrow.

[Robert Goren]

Jockey are very important in one case, when the jockey doesn't fit the horse's running style. For instance you can toss a deep closer in a turf route ridden by David Cohen. Put him on a front runner in a dirt race on a speed favoring track, he become a real threat with even the most confirmed quitters.

[Tom]

Jockeys......not just passengers anymore!

http://sports.yahoo.com/blogs/the-turnstile/horse-throws-rider-goes-win-race-belmont-130603903.html

[MJC922]

This is IMO a mistake people make.......you have to think "big" picture or,.....will you be betting next week, next month, next year? Most times i could care less who the jock is.

Agree it's a very tough factor to deal with, because it can make the difference quite often IMO but is difficult for me to take advantage of. The PEN case was an isolated one, rarely does it come down to that one factor. I like to ignore who's up until very late in the process, but at the lesser tracks I think you can't always afford to ignore it. My lack of sexism for example has cost me dearly this year. Lately when I land on a horse and find it ridden by one of the females I would get more run out of the horse if a wet diaper bag was tied onto it.

[DeltaLover]

Its all about the chance the HORSE has relative to its odds.



:ThmbUp: :ThmbUp: :ThmbUp:

The good thing is that people seem to not get it!

[Tom]

Doesn't the jockey affect the horse's chances?

[acorn54]

If the jockey doesn't matter then the trainers wouldn't be so concerned about getting good jockeys for their prize stake horses.

i don't know his modus operandi anymore because i haven't been following the horses connections since i started using a computer 9 years ago but at that time allen jerkens often used riders that werent the leading riders at the belmont-aqu-sar- circuit.

[Robert Fischer]

Doesn't the jockey affect the horse's chances?

yes.

Pace (and an individual horse's position within the pace dynamic) plays a role in a fair number of races.

Tactics (gun from the gate? rate?, move up to the front mid-race with a slow pace?,...) plays a role in a minority of races.

Individual skills (relaxing while rating at an even pace, prompting footwork, stickwork with both hands, etc...) plays a role in a small percentage of races.


All of these things gain value with a greater number of competitive horses in a given race.

[therussmeister]

Rider makes a big difference an awful lot of the time. A good example last night in the 8th at PEN. 2-turn flat mile, heavy chalk is breaking from post 7 with just enough early foot drawn inside him to expect a wide chasing trip. The two horse Sixsixteen showed he was back in form last time after a couple of mini-layoffs, now back in just two weeks figured to fire another good effort, importantly he's tied with the 7 for best early pace. My big fear was the rail horse though, Socio d'Oro, had back class, also had essentially the same early pace figure as Sixsixteen did so the way I saw it, this was shaping up to be a 'riders race'. Now I noticed Badamo is riding Socio d'Oro, hmm, well I don't follow PEN at all so I didn't know much about Belmonte's plans on Sixsixteen, however I have about a decade of close observation under my belt at FL during the 1990s and I know Badamo is far more likely to rate a horse early than he his to send hard out of there. So my money goes on Sixsixteen at 4-1 as they load up, but this is every bit as much a bet that Badamo will rate and let Sixsixteen go on with it. Gate pops and both of them break sharp, but Badamo rates, Belmonte more aggressive, exited the first turn on a clear lead, perfectly executed, game over, go cash. Flip the riders and maybe you have a different result. Many races are decided in the first quarter mile on tactics, it can pay to know rider tendencies.

The jockey made a big difference to you, because he gave you more confidence in your selection, but that does not mean a different jockey would have used different tactics with worse results.

[Stillriledup]

Doesn't the jockey affect the horse's chances?

Horses run just as fast for Joel Rosario, Johnny V and Rafael Bejarano as they do for Joe Blow, John Doe and Mary Smith.

Pay no attention to the little man on the back of the horse, they're all interchangable. ;)

[Dave Schwartz]

Its all about the chance the HORSE has relative to its odds.

Of course it is. But those chances are not created in a vacuum. EVERYTHING plays a part, albeit some things are small and many others are not what we (the horse playing public) might think.

I get Delta's agreement, though. He (I think) is saying that it always boils down to probabilities and payoff. And, as is usually the case, he and I are in agreement on that.



Regards,
Dave Schwartz

[proximity]

Oh, that is perfect! I am going to do an article on that for next week's show!

BTW, that show is Wednesday @5:30pm (pacific) instead of Thursday because of July 4th.

Sign up here (https://www2.gotomeeting.com/register/225284506)

Dave

i'm asking about the book. a lot of books on amazon, barnes and noble... have tables of contents listed. for example would there be an overall table of jockey impact values by specific overall jockey win percentages.... for example, 19% jockeys have an impact value of 1.xx, 18 percent jockeys have an impact value of 1.xy, .... 0 percent jockeys have an impact value of 0.xx??

[thaskalos]

Horses run just as fast for Joel Rosario, Johnny V and Rafael Bejarano as they do for Joe Blow, John Doe and Mary Smith.

Pay no attention to the little man on the back of the horse, they're all interchangable. ;)

This used to be my opinion too...but I eventually changed my mind.

I now have a list of jockeys that I wouldn't even risk a mind bet on.

[cj]

This used to be my opinion too...but I eventually changed my mind.

I now have a list of jockeys that I wouldn't even risk a mind bet on.

I have had similar experience. I formerly ignored jockeys, but it just costs you too many bets in my opinion. It is amazing how often a low percentage rider will screw something up.

[Stillriledup]

I have had similar experience. I formerly ignored jockeys, but it just costs you too many bets in my opinion. It is amazing how often a low percentage rider will screw something up.

Not only do they screw it up positionally, but they just can't make them run as fast, they don't look aerodynimaclly 'pleasing' in the saddle, their form is suspect, they just dont make as much 'speed' so you get the worst of both worlds.

[Longshot6977]

Personally, I like to have the horses I bet to have a mid-high % jockey on him. I realize the odds may be lower, but I figure his agent is getting good horses and a good jock will exploit any track bias. A good jock will understand the track and any existing bias and reads the PPs between races and knows where to position his/her horse to exploit any rail or speed bias.

Of course, I make many bets also where I don't really put too much emphesis on who the rider is since the horse figures to be a good bet no matter who is riding it.

[thaskalos]

...a good jock will exploit any track bias. A good jock will understand the track and any existing bias and reads the PPs between races and knows where to position his/her horse to exploit any rail or speed bias.



You have more faith in these jockeys than I do...

[Longshot6977]

You have more faith in these jockeys than I do...

That's why I emphasized a good jockey.:)

[Dave Schwartz]

i'm asking about the book. a lot of books on amazon, barnes and noble... have tables of contents listed. for example would there be an overall table of jockey impact values by specific overall jockey win percentages.... for example, 19% jockeys have an impact value of 1.xx, 18 percent jockeys have an impact value of 1.xy, .... 0 percent jockeys have an impact value of 0.xx??

I understood that.

I do a segment on every show from the book.

Yes, jockeys are covered.

[Edward DeVere]

I now have a list of jockeys that I wouldn't even risk a mind bet on.

Quote Of The Week.

[Johnny V]

I am probably in the minority on this but I pay very little attention to who the jockey is. IMO if the horse has a competent trainer then I just go with that in regard to the human connections of the horse. If the jock has such a dismal winning % or routinely gets horses in trouble I will not bet the horse, otherwise I think any competent jock can get the job done.

[acorn54]

I am probably in the minority on this but I pay very little attention to who the jockey is. IMO if the horse has a competent trainer then I just go with that in regard to the human connections of the horse. If the jock has such a dismal winning % or routinely gets horses in trouble I will not bet the horse, otherwise I think any competent jock can get the job done.


i agree, a competent trainer knows who the competent jockeys are i'm sure, anyway more so than i do as i am not a horseman.
i am sure there are incompetent jockeys just like any other profession, but when they show time and time again their incompetentcy riding a horse i would be suspect that they get good horses to ride.

[Capper Al]

I am probably in the minority on this but I pay very little attention to who the jockey is. IMO if the horse has a competent trainer then I just go with that in regard to the human connections of the horse. If the jock has such a dismal winning % or routinely gets horses in trouble I will not bet the horse, otherwise I think any competent jock can get the job done.

I mix it up here. I agree that the trainer is more important than the jockey, but jockey moves can still be important.

[jasperson]

All jockeys make mistakes. The good jockeys just make fewer of them. That's is why I want a good jockey on my horse.

[burnsy]

I look at it like this, i am not a trainer, a jockey, an agent or an owner. I am a gambler. Their job is TO NOT GAMBLE, big diff. The motives of a trainer are different than mine. Of course most of them look for the best jock. In most cases it is their DUTY to do so. They work for owners they have to give their horse every LEGAL edge to be doing their job for the owner. Notice the term LEGAL......when they pick a jockey or a race, the owner is trusting the trainers judgement especially if the horse appears to have a legit chance. But i am betting, anything can happen, its still a horse race. I am on the other side of the rail and so are my motives. I don't care who wins as long as I picked the right one to bet. I HAVE TO LOOK AT EVERY HORSE. Stats are fine but EACH race is an opportunity or entity. NO one knows whose going to win. On the NYRA circuit almost every jockey could win any race, reasonably in the top 12 (jocks) or so. There will be people at Saratoga that only win 5 to 10 races (the whole meet)but if i have one of them, on the right ticket, i could give a shit who rode it! The grass courses i will look longer at the jocks because there is always a segment that tends to dominate. BUT IF SEE A HORSE I LIKE AND ITS A SQUARE PRICE, i could give a crap, its summer here. If you can't ride you won't be around. People don't even account for the horses that win, quite frankly, anyone could of rode. Sometimes the "hungry" jocks on the "sleeper" catch the cocky ones sleeping. It happens every day. There are people on top but the difference in a jockey colony like this one.....i would bet on almost every one of them if i like the horses chance relative to its odds.

[Viruss]

IMO I want to see that the trainer has won with this jockey before. Trainers that have a medium to high win % and don't show that they have used the jockey before may be a negative factor. I think its a positive sign the trainers trying if he uses a rider he normally puts up on his winning horses.

Earl J

[NY BRED]

often times a top jockey will take a mount on a horse as he is looking
to ride another "star" in the trainer's barn .

Sure, top jockeys are always sought by trainers especially if they
fit the ability to ride the horse per the trainer's demands.

That said, when the gate opens the drama begins and the
game plan can go into plan b etc...

[Johnny V]

i'm asking about the book. a lot of books on amazon, barnes and noble... have tables of contents listed. for example would there be an overall table of jockey impact values by specific overall jockey win percentages.... for example, 19% jockeys have an impact value of 1.xx, 18 percent jockeys have an impact value of 1.xy, .... 0 percent jockeys have an impact value of 0.xx??
The tables that I have seen measuring jockey values including the one in Dave's Percentages and Probabilities all look at the rankings based on the top 5 or 10 or rear half etc. We all know that the favorites get the better jocks and the jocks get them because they supposedly have proved they are the better riders so it is kind of go around perpetuating thing. I think maybe way to measure the ability of a rider is to see how they perform when they are on the favorite. The favorite is theoretically supposed to win. Measure it that way and I think the so called second tier jockey colony would do quite well in comparison or maybe at the least be very competitive.
Some of these young riders are on hopeless long shots so their win % is going to suffer in comparison but they are good enough riders IMO when given the chance.

[DeltaLover]

The favorite is theoretically supposed to win.

Think again about this statement..

[chadk66]

anybody that thinks the jockey means little doesn't understand a thing about the business.

[Johnny V]

Think again about this statement..
What I mean is the % of winning favorites is pretty much in a sense an overall constant number. The favorite just by being the favorite alone is maybe the highest factor in winning % that is easily available and verifiable. I was thinking along the lines of seeing how jockeys do when compared to that when they are theoretically on the horse with the best chance to win. Then you are comparing them when they are on a live mount. Does that make any sense? Just thinking out loud ova' here.

[DeltaLover]

anybody that thinks the jockey means little doesn't understand a thing about the business.


I would say that it depends about what business we are talking about.

Although it is true that a top level jockey increases significantly the chances of his starter proving it significance for the business of the owner and trainer the same does not apply for betting purposes.

Since this concept is well known to the public it is reflected with high accuracy to the prices neutralizing it as a profit making factor.



F1: top_jockey_on_fav_or_cofav
F2: bottom_jockey_on_fav_or_cofav

F1 is significant better than F2 (chi2 = 126.72) considering wins.

F1 is NOT significant better than F2 for betting purposes.

Although F2 wins only 28% compared to 32% of F1 it is easy to conclude

from the ROIs that this makes no difference for betting purposes

factor winners losers roi win% iv starters races
F1 1420 3074 0.85 32% 2.38 4494 4279
F2 8066 20724 0.84 28% 1.23 28790 15505

[so.cal.fan]

I always check the trainer/jockey combo stats.
If you get to know the riders on your circuit, and watch plenty of races, you can see certain jockeys are superior with certain types of races/horses.
I like jockeys who ride well on turf courses. Or, ride well on bullrings, or on very off tracks (not too common in So. Cal.).
Some jockeys are better with distances.
Trainers are just as important with the above factors.....and you can add, some are much better training fillies than others. This is information that I don't think is very well known. BRIS, years ago, used to publish a trainer book with this filly and mare stat in it. I don't think they do anymore.
It is very important, at least here in So. California.
Same with turf, some trainers are better turf horse trainers than others.
So, Jockey and Trainer stats are important, but then........it really does depend on the ability of the horse.

[Johnny V]

I would say that it depends about what business we are talking about.

Although it is true that a top level jockey increases significantly the chances of his starter proving it significance for the business of the owner and trainer the same does not apply for betting purposes.

Since this concept is well known to the public it is reflected with high accuracy to the prices neutralizing it as a profit making factor.



F1: top_jockey_on_fav_or_cofav
F2: bottom_jockey_on_fav_or_cofav

F1 is significant better than F2 (chi2 = 126.72) considering wins.

F1 is NOT significant better than F2 for betting purposes.

Although F2 wins only 28% compared to 32% of F1 it is easy to conclude

from the ROIs that this makes no difference for betting purposes

factor winners losers roi win% iv starters races
F1 1420 3074 0.85 32% 2.38 4494 4279
F2 8066 20724 0.84 28% 1.23 28790 15505

Delta,
Thanks for posting that table. That is exactly what I was looking for. I would venture that the stats on the individual jockeys on how they ride when on those live horses would surprise me on how some of the so called bottom jocks would compare to some of the top name jocks on an individual basis both in % and ROI. An interesting table of stats there.

[DeltaLover]

Data research proves that the common approach for seeking for high profile connections does not constitute a winning approach.

[PhantomOnTour]

Jocks are a factor, but far less important than the ability of the horse.

Some jocks are just better than others at getting out of the gate quickly, rating/relaxing lone E horses to save energy, finishing strong in the lane etc...

I don't like Leparoux or Cornelio with speed horses - Leparoux likes to fall out of the gate and (from my observations) Cornelio just doesn't get horses to relax up front like others can.

[Segwin]

Overall and in general, the jockey is more of an indication of the trainer and connections, than any kind of factor.



To drill down a bit would a top notch jockey team up with a dog of a horse? I would think (but have been known to be wrong more than once) that a top jockey would steer clear of a dog. No money in it. So having a top jockey would mean that the horse, perhaps despite the numbers, has a pretty fair chance - no? Just thinking out loud.

[Robert Fischer]

Data research proves that the common approach for seeking for high profile connections does not constitute a winning approach.

I don't know how comprehensive we can be here. We can say that this data does show that jockeys are not a factor when betting on favorites.

Being on the public favorite has some advantages. The need for the jockey to use his skill set in an adverse situation is minimized, if not nearly absent. The public has also by definition, already given it's seal of approval to the betting entry of horse+jockey for that given situation.

These two 'filters' can and should logically level the playing field.

[Segwin]

is the book you are referring to by Tom Hambelton ?

Yes sir.

[Johnny V]

To drill down a bit would a top notch jockey team up with a dog of a horse? I would think (but have been known to be wrong more than once) that a top jockey would steer clear of a dog. No money in it. So having a top jockey would mean that the horse, perhaps despite the numbers, has a pretty fair chance - no? Just thinking out loud.
The top guys are on long shots. You see it at times. There is an old horsey saying at the track. "Never let the leading jockey go off at 10-1 without a bet". Or something to that effect. I don't know how that one works out but those riders are on some shots.

[Robert Fischer]

To drill down a bit would a top notch jockey team up with a dog of a horse? I would think (but have been known to be wrong more than once) that a top jockey would steer clear of a dog. No money in it. So having a top jockey would mean that the horse, perhaps despite the numbers, has a pretty fair chance - no? Just thinking out loud.

If a top jockey rides 'first-call' for a trainer, he will ride some slow horses along with that trainer's best stock. He'll ride pretty much everything that trainer has.
Occasionally you will see a horse off a layoff, maybe dropping in class, and the trainer's usual jockey is replaced with a low% jockey. That usually is not a good sign.

[raybo]

Jockeys/jockey agents, by and large, are in this game to make money, it's a business. Jockeys must ride races in order to make money. The agent's job is to keep the jockey busy, making money. One way of doing that is to stay on a trainer's good side, so that trainer keeps using that jockey. So, yes, sometimes a top jockey will ride a poor horse for a trainer, in order to keep getting the good mounts from that trainer. Cost of doing business.

[traynor]

I would say that it depends about what business we are talking about.

Although it is true that a top level jockey increases significantly the chances of his starter proving it significance for the business of the owner and trainer the same does not apply for betting purposes.

Since this concept is well known to the public it is reflected with high accuracy to the prices neutralizing it as a profit making factor.



F1: top_jockey_on_fav_or_cofav
F2: bottom_jockey_on_fav_or_cofav

F1 is significant better than F2 (chi2 = 126.72) considering wins.

F1 is NOT significant better than F2 for betting purposes.

Although F2 wins only 28% compared to 32% of F1 it is easy to conclude

from the ROIs that this makes no difference for betting purposes

factor winners losers roi win% iv starters races
F1 1420 3074 0.85 32% 2.38 4494 4279
F2 8066 20724 0.84 28% 1.23 28790 15505


This is a good example of too much data diluting its effectiveness. If you are really interested in meaningful statistics in regard to jockeys, you may want to layer it into class levels. Specifically, there is a direct correlation between individual jockey success and class level. Stated simply, it might be useful to consider that jockeys--much like everyone else--have specialties. Locate those specialties, and jockey statistics can be quite useful.

Additionally, because it is not information readily available for the cost of a DRF or data download, it is also more profitable.

[jpren37]

Perhaps a useful study would be to track who are the jocks that out ride their odds.

[traynor]

Perhaps a useful study would be to track who are the jocks that out ride their odds.

Yes, but the real questions would be, "On which horses? At what class level? At what distances? On what surfaces?" The whole trick is that people invariably seek (and more than willing to accept) "simple" answers to very complex questions. Understanding that "generic" data is almost worthless (or less than worthless if you bet on it) because it blurs distinctions that should be discrete (separate) is a very big step for most people.

There are many areas of profit in pari-mutuel wagering. One of the easiest to get into is an area in which there is LOTS of generic information, but not much specific. If one can strike a balance between "too much data" (that makes everything look the same) and "too small sample size" (that requires data be cleaned and corrected to avoid semantic distortion in the interpretation of that data), life gets much easier.

[flatstats]

Pareto Principle
This holds true with jockeys. 80% of all races are won by the same 20% of jockeys.

Ratings - Jockey Element
The jockey influences the ratings (of our UK races) anything from 9% to 23% depending on the type of race.

Jockey Biases
The biggest mistake (UK) punters make is that they think a jockey is a machine just there to steer the horse. They assume that jockeys all have the same ability but this is not the case.

Some jockeys are way better in routes than sprints, some jockeys are way better in mud rather than firm going.

[raybo]

Pareto Principle
This holds true with jockeys. 80% of all races are won by the same 20% of jockeys.

Ratings - Jockey Element
The jockey influences the ratings (of our UK races) anything from 9% to 23% depending on the type of race.

Jockey Biases
The biggest mistake (UK) punters make is that they think a jockey is a machine just there to steer the horse. They assume that jockeys all have the same ability but this is not the case.

Some jockeys are way better in routes than sprints, some jockeys are way better in mud rather than firm going.

Absolutely! :ThmbUp:

[jpren37]

Yes, but the real questions would be, "On which horses? At what class level? At what distances? On what surfaces?" The whole trick is that people invariably seek (and more than willing to accept) "simple" answers to very complex questions. Understanding that "generic" data is almost worthless (or less than worthless if you bet on it) because it blurs distinctions that should be discrete (separate) is a very big step for most people.

There are many areas of profit in pari-mutuel wagering. One of the easiest to get into is an area in which there is LOTS of generic information, but not much specific. If one can strike a balance between "too much data" (that makes everything look the same) and "too small sample size" (that requires data be cleaned and corrected to avoid semantic distortion in the interpretation of that data), life gets much easier.

Traynor,

I don't post much here because I'm a relatively new player and don't have a whole lot to contribute yet. Perhaps you can help me with some of my nagging ideas/questions.

If I were to track jock results using the final odds ( e.g. going 5 positions deep---public's Favorite choice, 2nd choice, 3rd choice etc. and awarding a jock who finishes 1st when he was third public choice two points and subtracting 2 points when he was the fav. and finished 3rd. and then used more filters such as purse and distance...would I be reducing the sample size too much with these filters? Can I assume that in the long run those factors in those additional filters are already incorporated in the final odds and therefore not necessary?

And at the risk of taking this thread in too different a direction, can you or anyone who would like to chime in, explain to me what might be examples or the difference between Causal factors vs Correlated factors vs winning characteristics that one looks for when trying to predict the outcome of a horse race.

and!!!

when evaluating the usefulness of a factor, should your only interest be the roi of that factor vs the percentage hit rate. (assuming your sample is large enough...whatever that might mean)

Hope this post is at least somewhat clear. I know, I know...I should probable go out and buy a couple of good stat books!!

[traynor]

Pareto Principle
This holds true with jockeys. 80% of all races are won by the same 20% of jockeys.

Ratings - Jockey Element
The jockey influences the ratings (of our UK races) anything from 9% to 23% depending on the type of race.

Jockey Biases
The biggest mistake (UK) punters make is that they think a jockey is a machine just there to steer the horse. They assume that jockeys all have the same ability but this is not the case.

Some jockeys are way better in routes than sprints, some jockeys are way better in mud rather than firm going.

Understanding that one basic fact can do wonders for one's bottom line. The question is not "how good is this jockey?" It is "how good is this jockey on this surface, at this distance, on this (grade/class/whatever) horse?" Broad categorizations on not especially useful where jockeys are concerned.

[so.cal.fan]

One big advantage of playing one circuit and attending the races live often is that you get to know the strengths and weaknesses of jockeys and trainers.
Hang around the paddock and get to know all the regular players...most are worth talking to. There are so few people who attend live races these days, you'll get to know lots of horsemen and women.

[traynor]

Traynor,

I don't post much here because I'm a relatively new player and don't have a whole lot to contribute yet. Perhaps you can help me with some of my nagging ideas/questions.

If I were to track jock results using the final odds ( e.g. going 5 positions deep---public's Favorite choice, 2nd choice, 3rd choice etc. and awarding a jock who finishes 1st when he was third public choice two points and subtracting 2 points when he was the fav. and finished 3rd. and then used more filters such as purse and distance...would I be reducing the sample size too much with these filters? Can I assume that in the long run those factors in those additional filters are already incorporated in the final odds and therefore not necessary?

And at the risk of taking this thread in too different a direction, can you or anyone who would like to chime in, explain to me what might be examples or the difference between Causal factors vs Correlated factors vs winning characteristics that one looks for when trying to predict the outcome of a horse race.

and!!!

when evaluating the usefulness of a factor, should your only interest be the roi of that factor vs the percentage hit rate. (assuming your sample is large enough...whatever that might mean)

Hope this post is at least somewhat clear. I know, I know...I should probable go out and buy a couple of good stat books!!

Get the chart of a race. Six furlongs on the dirt works well to start. Look at the PPs of the winner (ignoring all the other horses for the moment). Mark or highlight or circle or whatever what you consider to be "indicators" that were visible in the PPs that you could have used to select that horse. No deep, agonizing, soul-searching, pie-in-the-sky, mind-bending, arcane mysteries. Look for class moves, positions and beaten lengths at various calls, odds, jockey records, trainer records--initially, almost anything that seems it might be predictive.

It shouldn't take more than a handful of races to give you a generic understanding of what the PPs of a potential winner (or close finisher) look like. They form a pattern, and what you will learn is how to recognize that pattern (those patterns). If you initially concentrate on one distance/surface/general class you will spare yourself the endless confusion that many seem to afflict themselves with. Mid-level claimers at six furlongs on the dirt are a good place to start.

Initially, ignore ROI. The objective is to pick winners. Once you recognize the patterns of winners (and close up finishers) in mid-level claiming sprints on the dirt, there is plenty of time to start wandering down the lane of "value wagers." Two-thirds of the favorites lose. That can be money in your pocket (and a healthy ROI) if you learn to recognize the false favorites, and learn to recognize whch entries will be the winners in those races.

[traynor]

One big advantage of playing one circuit and attending the races live often is that you get to know the strengths and weaknesses of jockeys and trainers.
Hang around the paddock and get to know all the regular players...most are worth talking to. There are so few people who attend live races these days, you'll get to know lots of horsemen and women.

Great advice! That also gives one the opportunity to learn a LOT about the physical appearance and behavior of horses that is missed completely by the ivory tower crowd that only sees horses on their computers.

[jpren37]

Pareto Principle
This holds true with jockeys. 80% of all races are won by the same 20% of jockeys.

Ratings - Jockey Element
The jockey influences the ratings (of our UK races) anything from 9% to 23% depending on the type of race.

Jockey Biases
The biggest mistake (UK) punters make is that they think a jockey is a machine just there to steer the horse. They assume that jockeys all have the same ability but this is not the case.

Some jockeys are way better in routes than sprints, some jockeys are way better in mud rather than firm going.
Flatstats,

Do you find that jocks who have shown in the past to be weak at routes can improve to become very good at sprints, can go from bad to good on mud etc. If so, do you make allowances for improvement/change when you consider jockey element

[Boris]

Jockeys/jockey agents, by and large, are in this game to make money, it's a business. Jockeys must ride races in order to make money. The agent's job is to keep the jockey busy, making money. One way of doing that is to stay on a trainer's good side, so that trainer keeps using that jockey. So, yes, sometimes a top jockey will ride a poor horse for a trainer, in order to keep getting the good mounts from that trainer. Cost of doing business.
Tagging on to this thought, and be advised I don't handicap much these days, jockey moves like what I see at Gulfstream today catch my eye. The card is very "standard" today, yet Rosario catches a plane (assuming he shows up) from Belmont to ride some horses. Same with Prado and J Velazquez. JR & JV then head to Parx tomorrow for a $250K race while EP has a beach day before they all get back to Belmont for Wednesday's card. The Tuesday behavior makes sense, but the trip to FL is not to bump their frequent flyer miles. Like Raybo, I believe there is a business decision for the mounts at GP today. The challenge is to determine which mount is "the reason". Another piece of the Jockey Puzzle that can be profitable.

[so.cal.fan]

Good points, Raybo and Boris.
Horseracing is a business.
It pays to always handicap keeping that in mind.
Again, know your circuit.

[DeltaLover]

Yes, but the real questions would be, "On which horses? At what class level? At what distances? On what surfaces?" The whole trick is that people invariably seek (and more than willing to accept) "simple" answers to very complex questions. Understanding that "generic" data is almost worthless (or less than worthless if you bet on it) because it blurs distinctions that should be discrete (separate) is a very big step for most people.



This is a good example of too much data diluting its effectiveness. If you are really interested in meaningful statistics in regard to jockeys, you may want to layer it into class levels.

You are touching one of the most fundamental and misunderstood topic when it comes to horse betting and handicapping.

Although exceptions to the generic is exactly what we as bettors are looking for, still this does not mean that we necessary need to refine our conditions to the more specific versus the more general.

As far a condition can be clearly expressed and processed using a correct methodology it does not really matter how restrictive or relaxed are the selection conditions.

The specific example of top vs bottom jockeys riding the favorite of the co favorite of the race serves as an indicator to the fact, that although the ability of the jockey really matters for the outcome of a race it is of little betting value since the crowd seems to adjust with great accuracy.

I do not want to say that this necessary proves that it is impossible to find some cases where the crowd might get confused based in class level, or any other condition but I still believe that something like this is very difficult to happen given that the large sample proves the opposite.

Anyway, the topic is deep and probably represents the most important handicapping concept there is and we can talk endlessly about it.

[pondman]

Jockeys can be rated by the caliber of horses people allow them to sit on. This includes morning riding. I'm constantly searching for those ringers from S. Cal or NY being ridden by their exercise riders in other venues. I'm happy when the geezers are asking "who is that ?" Those are the horses that send my wife to Hawaii. There are a number of rider like this who you don't see very often, but they'll take a horse and win it's maiden and then come right back and win another. And then you don't see the jockey for 3 weeks. They make their money from morning work, but they'll knock big home runs, when they ride.

I've also had the personally experience of watching in-laws and relative ship to the basement, and get a girl to hang on, and the horse wins by a huge margin at a huge price. So it's really going to depend on the scenario.

[traynor]

You are touching one of the most fundamental and misunderstood topic when it comes to horse betting and handicapping.

Although exceptions to the generic is exactly what we as bettors are looking for, still this does not mean that we necessary need to refine our conditions to the more specific versus the more general.

As far a condition can be clearly expressed and processed using a correct methodology it does not really matter how restrictive or relaxed are the selection conditions.

The specific example of top vs bottom jockeys riding the favorite of the co favorite of the race serves as an indicator to the fact, that although the ability of the jockey really matters for the outcome of a race it is of little betting value since the crowd seems to adjust with great accuracy.

I do not want to say that this necessary proves that it is impossible to find some cases where the crowd might get confused based in class level, or any other condition but I still believe that something like this is very difficult to happen given that the large sample proves the opposite.

Anyway, the topic is deep and probably represents the most important handicapping concept there is and we can talk endlessly about it.

Once more, the most basic question business analysts ask (or should ask) when viewing data is, "Does this stuff really mean what I/we/they think it means?" In the case of most "broad brush" studies of jockeys and trainers, the answer is a simple, "no." Layering data is one of the most fundamental approaches in data mining. It tends to expose that which is obscured by assumptions that large masses of data are essentially similar. In many cases they are not, and only appear to be so because of the way they are studied.

If all the data elements are essentially similar, extracting random samples (as in bootstrapping) will essentially replicate the distribution in the main population. With jockeys and trainers (and a number of other areas well worth more focused research), that is rarely the case.

[traynor]

I think I have a fundamentally different way of looking at data than most. Or at least so some of the more outspoken professors in graduate school seemed to think. I understand the underlying logic. I think in many cases, that logic is wrong. Most of all, I think there is as pervasive a tendency when looking at large collections of data to be misled as there is to be misled by small collections. What is missing is that "outside observer" point-of-view, to analyze the process--not just the data--and to objectively evaluate that process to assure that the conclusions reached are in fact accurate representations of the reality studied, rather than reflections of the researchers preconceived beliefs.

It is not easy. Profitable, but not easy. Especially when studying jockeys and trainers. Thoughtful (and above all else, appropriate) layering is prerequisite. You know as well as I do that the answers are derivatives of the questions. When the wrong questions are asked, the answers will invariably be wrong.

[rubicon55]

I think I have a fundamentally different way of looking at data than most. Or at least so some of the more outspoken professors in graduate school seemed to think. I understand the underlying logic. I think in many cases, that logic is wrong. Most of all, I think there is as pervasive a tendency when looking at large collections of data to be misled as there is to be misled by small collections. What is missing is that "outside observer" point-of-view, to analyze the process--not just the data--and to objectively evaluate that process to assure that the conclusions reached are in fact accurate representations of the reality studied, rather than reflections of the researchers preconceived beliefs.

It is not easy. Profitable, but not easy. Especially when studying jockeys and trainers. Thoughtful (and above all else, appropriate) layering is prerequisite. You know as well as I do that the answers are derivatives of the questions. When the wrong questions are asked, the answers will invariably be wrong.

Per the OP I would ask that by breaking down jockeys into sub categories as suggested by surface, distance, trainer relationships, purse values, etc. that by the time you render a reasonable sample which is quantifiable that the general public does not have then the meet is over. Keeping track of these for years may end up being not so useful such as when Rosario went from So Cal to NY. Yes certain traits would carry over to the east coast but its back to square one as far as surface, trainers, etc. Good, high percentage jockey agents are always trying to find a track to get better rides. Help me out since Iam trying to see how this effort is being worth the time and energy taken to that level where micro-managing jockey colonies takes a certain amount of time and effort while still having time to look at other race factors that may be more or just as important. Just curious.

[Robert Fischer]

Here is a recent ride that I liked.

AREPEATEE :7:#7 ($6.00-1) 6/23 race 6 prx
Watch the replay
I thought the jockey fought hard from the gate to gain initiative, then slowed down the pace mid-way. He benefited from some bad rides by others, but it was nice handling.
chart: http://www.equibase.com/static/chart/pdf/PRX062313USA6.pdf


This is a jockey who has flashed some individual talent, but has had a rough time of it. I had somewhat lost patience in his situation, but he is showing signs of returning to form lately.

In his last 10 mounts
3 Wins (1.80-1*, 6-1, 3.3-1)
1 Place (12.9-1)
1 Shw (21.3-1)

He has a mount today, but it is a bad horse who he brought in 3rd last time and will be overbet (Whirling Dervish). [his last race is a contrast to the Arepeatee race, with much different tactics]

These kind of jocks can take one quality in a horse (such as early speed) and give him somewhat of an outside chance to win or contend. The rides may look nothing like earlier past performances with previous jockeys. Usually these guys show some consistent success, they get better mounts, and the public catches on to their general ability level.

[traynor]

Per the OP I would ask that by breaking down jockeys into sub categories as suggested by surface, distance, trainer relationships, purse values, etc. that by the time you render a reasonable sample which is quantifiable that the general public does not have then the meet is over. Keeping track of these for years may end up being not so useful such as when Rosario went from So Cal to NY. Yes certain traits would carry over to the east coast but its back to square one as far as surface, trainers, etc. Good, high percentage jockey agents are always trying to find a track to get better rides. Help me out since Iam trying to see how this effort is being worth the time and energy taken to that level where micro-managing jockey colonies takes a certain amount of time and effort while still having time to look at other race factors that may be more or just as important. Just curious.

Jockeys have individual strengths and weaknesses. Those strengths and weaknesses can be discovered with a reasonable amount of effort. Once discovered, that jockey can be placed in a category, with further monitoring by means of a performance envelope to determine if the jockey should remain in that category, or be moved to another. Each jockey may be in more than one category.

The process involves structuring the categories, not the individual jockeys. Once the categories are established, the performance of the jockey determines which category he or she fits in, and to what degree (a range of values describing a performance envelope).

Initially, it is no more complex that coding an "extractor"--a few lines of code that place the jockey's name in a "dump file" for various categories. A jockey wins a mile and an eighth on the dirt, that jockey's name goes in a dump file along with the details of the race. Probably a huge amount of work to do by hand, but fairly simple to do with a computer. A simple parsing function can then be applied to the dump files to determine frequency--how many times each jockey wins, places, whatever, in the specific category of race. Mutuel data, etc., is all attached to the name.

Rather than an endless process of micro-management, it is a design problem. Once the meaningful categories are established, the rest of it becomes pretty much automatic--the data is extracted automatically, and each category is continually updated to reflect current reality. Lots of work to start, but well worth the effort.

Specifically, the focus is shifted from the individual jockey's performance to the individual jockey's inclusion in a category. It is not so much a matter of accumulating records on the individual jockey as it is a determination of whether or not that jockey fits the profile for jockeys in that category. The accuracy and layering of the categories and profiles is more imprtant than the individual jockey's performance.

[traynor]

When serious amounts of money are wagered on races, the bettors want the most useful information available on which to base wagering decisions. I think it may be less than useful to assume those serious bettors are using the same set of values as everyone else, massaged in a slightly different way.

[traynor]

I assume that most who are serious about betting have noticed the proliferation of software applications "analyzing" pace and speed (because those "values" are easily quantified and massaged using existing data sources--with little or no proof of their effectiveness) with an associated lack of software applications that generate meaningful information about jockeys and trainers (because the design and implementation involve more than Computer Programming 101 design and data analysis skills and are therefore considered to be "too complex")?

Consider the lonely whale, earning his or her "seven-figure income from churning the rebate shops." Do you actually believe that she or he is going to make his or her wagers based on simplistic number crunching of an app available to anyone with a few dollars and an Internet connection?

The value of information is in direct proportion to the number of people who have that information. Rather than endlessly mulling over the same pace and speed figures everyone else is looking at, those intent on earning a profit from wagering might do well to consider doing a bit more real work and a bit less mulling.

Astute analysis of jockey and trainer factors may prove to be considerably more valuable than those unwilling to exert the effort necessary declare them to be.

[Robert Fischer]

oops, human error

In his last 10 mounts
4 Wins (1.80-1*, 15.40, 6-1, 3.3-1)
1 Place (12.9-1)
1 Shw (21.3-1)


In his last 10 mounts
3 Wins (1.80-1*, 6-1, 3.3-1)
1 Place (12.9-1)
1 Shw (21.3-1)

[Robert Fischer]

Jockeys have individual strengths and weaknesses. Those strengths and weaknesses can be discovered with a reasonable amount of effort. Once discovered, that jockey can be placed in a category, with further monitoring by means of a performance envelope to determine if the jockey should remain in that category, or be moved to another. Each jockey may be in more than one category.

The process involves structuring the categories, not the individual jockeys. Once the categories are established, the performance of the jockey determines which category he or she fits in, and to what degree (a range of values describing a performance envelope).

Initially, it is no more complex that coding an "extractor"--a few lines of code that place the jockey's name in a "dump file" for various categories. A jockey wins a mile and an eighth on the dirt, that jockey's name goes in a dump file along with the details of the race. Probably a huge amount of work to do by hand, but fairly simple to do with a computer. A simple parsing function can then be applied to the dump files to determine frequency--how many times each jockey wins, places, whatever, in the specific category of race. Mutuel data, etc., is all attached to the name.

Rather than an endless process of micro-management, it is a design problem. Once the meaningful categories are established, the rest of it becomes pretty much automatic--the data is extracted automatically, and each category is continually updated to reflect current reality. Lots of work to start, but well worth the effort.

Specifically, the focus is shifted from the individual jockey's performance to the individual jockey's inclusion in a category. It is not so much a matter of accumulating records on the individual jockey as it is a determination of whether or not that jockey fits the profile for jockeys in that category. The accuracy and layering of the categories and profiles is more important than the individual jockey's performance.

I agree with this. The proper categories are going to be what makes the jockey data useful.
You don't have to be a computer whiz or an elite handicapper to look at the traditional jockey stats, and see that someone like Rosario at the current Belmont meet is very good. The traditional statistical categories do a decent job of expressing the general success that a jockey has had. They just don't provide an edge in betting a given race.

[Segwin]

I assume that most who are serious about betting have noticed the proliferation of software applications "analyzing" pace and speed (because those "values" are easily quantified and massaged using existing data sources--with little or no proof of their effectiveness) with an associated lack of software applications that generate meaningful information about jockeys and trainers (because the design and implementation involve more than Computer Programming 101 design and data analysis skills and are therefore considered to be "too complex")?

Consider the lonely whale, earning his or her "seven-figure income from churning the rebate shops." Do you actually believe that she or he is going to make his or her wagers based on simplistic number crunching of an app available to anyone with a few dollars and an Internet connection?

The value of information is in direct proportion to the number of people who have that information. Rather than endlessly mulling over the same pace and speed figures everyone else is looking at, those intent on earning a profit from wagering might do well to consider doing a bit more real work and a bit less mulling.

Astute analysis of jockey and trainer factors may prove to be considerably more valuable than those unwilling to exert the effort necessary declare them to be.

Some really good and thoughtful posts - thanks. :ThmbUp:

[Robert Fischer]

He has a mount today, but it is a bad horse who he brought in 3rd last time and will be overbet (Whirling Dervish). [his last race is a contrast to the Arepeatee race, with much different tactics]

Here we are race 3 prx.
the :8: Whirling Dervish is an outrageous 5/2 on the early tote.

This is a really slow horse who got a great ride last time to get 3rd @ 21.3-1.

I'll still be rooting for him but in the next 10min I will be looking around frantically elsewhere for value.

----
update 2min to post now 4-1. More realistic but still a big underlay.
7 is 3 lengths best and is bet like it. not a ton of value here

[flatstats]

Flatstats,

Do you find that jocks who have shown in the past to be weak at routes can improve to become very good at sprints, can go from bad to good on mud etc. If so, do you make allowances for improvement/change when you consider jockey element

Jockeys tend to be conditioned. Few of them break the mould.

Sometimes apprentices will show an improvement in stats when they find a groove but overall I would say jockeys keep on doing what they have always done.

Why are they conditioned?

Because of the yards they have worked for mostly.

A Nicholls rarely wins out of sprint distances because he mostly rides for his father and his father is a pure sprint trainer.

Trainer D Nicholls Distance Stats
Distance I.V. A/E ROI%
--------------------------------
Sprints 1.04 0.95 -8%
Middle 0.87 0.79 -33%
Long 0.50 0.55 -40%


Now the stats for his jockey son:

Jockey A Nicholls Distance Stats (Any Trainer)
Distance I.V. A/E ROI%
--------------------------------
Sprints 1.13 0.87 -21%
Middle 0.80 0.72 -46%
Long 0.30 0.34 -79%


There's a correlation there with the trainer stats in that A Nicholls will have more negative stats the longer the distance in the same way with the trainer. So now look at the stats for when A Nicholls is not riding for D Nicholls:

Jockey A Nicholls Distance Stats (Not riding for D Nicholls)
Distance I.V. A/E ROI%
--------------------------------
Sprints 1.13 0.69 -46%
Middle 0.90 0.57 -74%
Long 0.73 0.41 -72%


This indicates two things. The first is that when he rides for his father his performance improves. The second is that he still has a distance bias when riding for other trainers.

Champion Jockey
Paul Hanagan was a northern jockey riding mostly at the lower class races and lower grade courses up north. He hardly rode down south and when he did his usual 10-15% northern course strikes would plummet to 5%.

Then he became Champion Jockey, winning more races than the other top jockeys. This change in fortune opened up new opportunities for him in that he became the number one retained jockey for Hamdan Al Maktoum.

He now had access to some of the best bred horses and the facilities of the top arab yards.

Unfortunately this meant that he now had to ride more often at the southern courses and due to his lack of experience at these courses he is not doing so well despite now riding for far wealthier owners.

Hanagan still struggles at southern Grade 1 courses such as Epsom (The Derby course), Ascot (Royal), Goodwood as well as all southern courses in general.

In the past 2 years his northern courses strike rate has improved from around 15%-19% whilst his southern course strike rate is still only 7.7%.

Why is this? It has nothing to do with leylines, latitude or the position of the sun! It is purely because he has never had the winning experience at these southern courses. It's not the fact that they are in the south but the fact that they have unique properties (shape, direction, undulations) and this is what he has not yet mastered despite being a Champion Jockey.

Of course the masses of punters do not understand this because they seem him as a Champion Jockey riding for wealthy owners. They never understand that he is as ineffective as greenhorn apprentice at Epsom and Ascot.

[Dave Schwartz]

Good post.

:ThmbUp:

[pondman]

Most of all, I think there is as pervasive a tendency when looking at large collections of data to be misled as there is to be misled by small collections. What is missing is that "outside observer" point-of-view, to analyze the process--not just the data--and to objectively evaluate that process to assure that the conclusions reached are in fact accurate representations of the reality studied, rather than reflections of the researchers preconceived beliefs.

This game requires you to position your money for the surprise. The surprises won't be found in the average. You want the anomalies to win. Unless you have a "when all hell breaks loose" method, I don't think your long term profit will be positive.

I began to play well when I stopped the nonsense with the super jockeys, and looked for other riders to win.
.

[jpren37]

Jockeys tend to be conditioned. Few of them break the mould.

Sometimes apprentices will show an improvement in stats when they find a groove but overall I would say jockeys keep on doing what they have always done.

Why are they conditioned?

Because of the yards they have worked for mostly.

A Nicholls rarely wins out of sprint distances because he mostly rides for his father and his father is a pure sprint trainer.

Trainer D Nicholls Distance Stats
Distance I.V. A/E ROI%
--------------------------------
Sprints 1.04 0.95 -8%
Middle 0.87 0.79 -33%
Long 0.50 0.55 -40%


Now the stats for his jockey son:

Jockey A Nicholls Distance Stats (Any Trainer)
Distance I.V. A/E ROI%
--------------------------------
Sprints 1.13 0.87 -21%
Middle 0.80 0.72 -46%
Long 0.30 0.34 -79%


There's a correlation there with the trainer stats in that A Nicholls will have more negative stats the longer the distance in the same way with the trainer. So now look at the stats for when A Nicholls is not riding for D Nicholls:

Jockey A Nicholls Distance Stats (Not riding for D Nicholls)
Distance I.V. A/E ROI%
--------------------------------
Sprints 1.13 0.69 -46%
Middle 0.90 0.57 -74%
Long 0.73 0.41 -72%


This indicates two things. The first is that when he rides for his father his performance improves. The second is that he still has a distance bias when riding for other trainers.

Champion Jockey
Paul Hanagan was a northern jockey riding mostly at the lower class races and lower grade courses up north. He hardly rode down south and when he did his usual 10-15% northern course strikes would plummet to 5%.

Then he became Champion Jockey, winning more races than the other top jockeys. This change in fortune opened up new opportunities for him in that he became the number one retained jockey for Hamdan Al Maktoum.

He now had access to some of the best bred horses and the facilities of the top arab yards.

Unfortunately this meant that he now had to ride more often at the southern courses and due to his lack of experience at these courses he is not doing so well despite now riding for far wealthier owners.

Hanagan still struggles at southern Grade 1 courses such as Epsom (The Derby course), Ascot (Royal), Goodwood as well as all southern courses in general.

In the past 2 years his northern courses strike rate has improved from around 15%-19% whilst his southern course strike rate is still only 7.7%.

Why is this? It has nothing to do with leylines, latitude or the position of the sun! It is purely because he has never had the winning experience at these southern courses. It's not the fact that they are in the south but the fact that they have unique properties (shape, direction, undulations) and this is what he has not yet mastered despite being a Champion Jockey.

Of course the masses of punters do not understand this because they seem him as a Champion Jockey riding for wealthy owners. They never understand that he is as ineffective as greenhorn apprentice at Epsom and Ascot.

Thanks much for responding to my question with such a clear response backed by examples.

In my earlier post in this thread I asked about the worth of a technique used to evaluate jocks. It was presented by Lee Russo (at the time, a very successful professional handicapper) at a handicapping expo which he claimed added a lot to his success. Briefly, he recorded a jocks performance in relation to their odds by adding/subtracting points based on race outcome. For example, if the jock was on the fav. and finished third, he subtracted two points from the jocks running total. If he was on the third choice and won, he added two points to the jocks running total. This was done without regard to further filters such as class, distance etc. I guess he felt that the additional filters had already been discounted in the odds.

In your opinion, do you think such a system has merit?

[traynor]

Jockeys tend to be conditioned. Few of them break the mould.

Sometimes apprentices will show an improvement in stats when they find a groove but overall I would say jockeys keep on doing what they have always done.

Why are they conditioned?

Because of the yards they have worked for mostly.

A Nicholls rarely wins out of sprint distances because he mostly rides for his father and his father is a pure sprint trainer.

Trainer D Nicholls Distance Stats
Distance I.V. A/E ROI%
--------------------------------
Sprints 1.04 0.95 -8%
Middle 0.87 0.79 -33%
Long 0.50 0.55 -40%


Now the stats for his jockey son:

Jockey A Nicholls Distance Stats (Any Trainer)
Distance I.V. A/E ROI%
--------------------------------
Sprints 1.13 0.87 -21%
Middle 0.80 0.72 -46%
Long 0.30 0.34 -79%


There's a correlation there with the trainer stats in that A Nicholls will have more negative stats the longer the distance in the same way with the trainer. So now look at the stats for when A Nicholls is not riding for D Nicholls:

Jockey A Nicholls Distance Stats (Not riding for D Nicholls)
Distance I.V. A/E ROI%
--------------------------------
Sprints 1.13 0.69 -46%
Middle 0.90 0.57 -74%
Long 0.73 0.41 -72%


This indicates two things. The first is that when he rides for his father his performance improves. The second is that he still has a distance bias when riding for other trainers.

Champion Jockey
Paul Hanagan was a northern jockey riding mostly at the lower class races and lower grade courses up north. He hardly rode down south and when he did his usual 10-15% northern course strikes would plummet to 5%.

Then he became Champion Jockey, winning more races than the other top jockeys. This change in fortune opened up new opportunities for him in that he became the number one retained jockey for Hamdan Al Maktoum.

He now had access to some of the best bred horses and the facilities of the top arab yards.

Unfortunately this meant that he now had to ride more often at the southern courses and due to his lack of experience at these courses he is not doing so well despite now riding for far wealthier owners.

Hanagan still struggles at southern Grade 1 courses such as Epsom (The Derby course), Ascot (Royal), Goodwood as well as all southern courses in general.

In the past 2 years his northern courses strike rate has improved from around 15%-19% whilst his southern course strike rate is still only 7.7%.

Why is this? It has nothing to do with leylines, latitude or the position of the sun! It is purely because he has never had the winning experience at these southern courses. It's not the fact that they are in the south but the fact that they have unique properties (shape, direction, undulations) and this is what he has not yet mastered despite being a Champion Jockey.

Of course the masses of punters do not understand this because they seem him as a Champion Jockey riding for wealthy owners. They never understand that he is as ineffective as greenhorn apprentice at Epsom and Ascot.

I think anyone who is serious about his or her wagering should have (and use) such information on a daily basis. Anyone who does not is taking a serious hit on her or his potential profits before the first wager is made. Scrambling after nickels and dimes is perfectly fine, if that is what one wants to do, or if one is wagering as a way to fill an otherwise empty life. My problem is the cost of holidays in Madagascar. Or Cape Town. Nickels and dimes just don't get it for me.

[Dave Schwartz]

In your opinion, do you think such a system has merit?

Why not something more like a true weighting system?

Why not create a metric called "Pool Impact Value" that measures the jockey's performance precisely relative to the odds on his horses?

[jpren37]

Why not something more like a true weighting system?

Why not create a metric called "Pool Impact Value" that measures the jockey's performance precisely relative to the odds on his horses?

Dave, Since you've been so generous in the past with your handicapping insights,I was hoping you might present just such a metric on your next handicapping show!!! :)

[Robert Fischer]

may as well update.


Villanueva In his last 10 mounts
4 Wins (1.80-1*, 15.4-1, 6.0-1, 3.3-1)
2 Place (4.2-1, 12.9-1)
1 Shw (21.3-1)

Such a small sample size, that these placings at these odds could be pure luck, and the fact that I happened to look at this guy when thinking about the thread could have been a lucky misconception that happened to seem good.

Still interesting to see a jockey bring these type of odds horses in.
If such a sample was in fact representative, it would represent something much different from the task a jockey has when bringing in public favorites. Just interesting to look at, and ponder.


:confused: ? Did anyone happen to catch Joel Rosario's ride on the :6: Somali Lemonade @ 6-1, in the $250K turf feature?
It was an evenly matched field where the trainer of the 5 was getting bet, and one or two others were underlaid based on current form, but he was 4th choice out of 6 horses.
I don't know if it was an all-star ride or not, but there was a part during the first half where he let the 6/5 public favorite go on the lead and opted for an even pace that seemed to be a wise choice, and then late in the race he did show strong handling. The announcer pumped up the ride in his call for whatever that is worth.

[Dave Schwartz]

Dave, Since you've been so generous in the past with your handicapping insights,I was hoping you might present just such a metric on your next handicapping show!!!


Count on it.

[Phantombridgejumpe]

rider.

I thought he was an underlay and not even Joel could help, I paid.

I think he will show a flat bet profit at Saratoga this summer.

[DeltaLover]

If you are really interested in meaningful statistics in regard to jockeys, you may want to layer it into class levels. Specifically, there is a direct correlation between individual jockey success and class level.


The data I posted above contained a misconception in the definition of the bottom jockey.

I am reposting, using the following definitions:

top jockey: 20%+ this+last year
bottom jockey: 5%- this+last year


Note that class levels are insignificant



Main Factor F1 = top_jockey_on_fav_or_cofav
Ransack Factor F2 = bottom_jockey_on_fav_or_cofav

F2 adds significant performance change to F1 chi2 = 8.50

factor winners losers roi win% iv starters races
top_jockey_on_fav_or_cofav 1420 3074 0.85 32% 2.38 4494 4279
bottom_jockey_on_fav_or_cofav 198 555 0.87 26% 2.03 753 701



Main Factor F1 = high odds + topJock
Ransack Factor F2 = high odds + bottomJock

F2 adds significant performance change to F1 chi2 = 16.72

factor winners losers roi win% iv starters races
high odds + topJock 47 1202 0.78 4% 0.32 1249 1187
high odds + bottomJock 133 6798 0.58 2% 0.12 6931 4918



Main Factor F1 = bottomJock+stakes
Ransack Factor F2 = bottomJock+claimer

Not Significant chi2 = 1.94

factor winners losers roi win% iv starters races
bottomJock+stakes 83 1485 0.59 5% 0.31 1568 1045
bottomJock+claimer 236 3517 0.70 6% 0.34 3753 2379



Main Factor F1 = bottomJock+ALW
Ransack Factor F2 = bottomJock+CLM

Not Significant chi2 = 0.12

factor winners losers roi win% iv starters races
bottomJock+ALW 49 690 0.66 7% 0.41 739 548
bottomJock+CLM 236 3517 0.70 6% 0.34 3753 2379



Main Factor F1 = topJock+STKS
Ransack Factor F2 = topJock+CLM

Not Significant chi2 = 1.44

factor winners losers roi win% iv starters races
topJock+STKS 751 2895 0.84 21% 1.22 3646 2748
topJock+CLM 519 1852 0.80 22% 1.51 2371 1983

[traynor]

The data I posted above contained a misconception in the definition of the bottom jockey.

I am reposting, using the following definitions:

top jockey: 20%+ this+last year
bottom jockey: 5%- this+last year


Note that class levels are insignificant



Main Factor F1 = top_jockey_on_fav_or_cofav
Ransack Factor F2 = bottom_jockey_on_fav_or_cofav

F2 adds significant performance change to F1 chi2 = 8.50

factor winners losers roi win% iv starters races
top_jockey_on_fav_or_cofav 1420 3074 0.85 32% 2.38 4494 4279
bottom_jockey_on_fav_or_cofav 198 555 0.87 26% 2.03 753 701



Main Factor F1 = high odds + topJock
Ransack Factor F2 = high odds + bottomJock

F2 adds significant performance change to F1 chi2 = 16.72

factor winners losers roi win% iv starters races
high odds + topJock 47 1202 0.78 4% 0.32 1249 1187
high odds + bottomJock 133 6798 0.58 2% 0.12 6931 4918



Main Factor F1 = bottomJock+stakes
Ransack Factor F2 = bottomJock+claimer

Not Significant chi2 = 1.94

factor winners losers roi win% iv starters races
bottomJock+stakes 83 1485 0.59 5% 0.31 1568 1045
bottomJock+claimer 236 3517 0.70 6% 0.34 3753 2379



Main Factor F1 = bottomJock+ALW
Ransack Factor F2 = bottomJock+CLM

Not Significant chi2 = 0.12

factor winners losers roi win% iv starters races
bottomJock+ALW 49 690 0.66 7% 0.41 739 548
bottomJock+CLM 236 3517 0.70 6% 0.34 3753 2379



Main Factor F1 = topJock+STKS
Ransack Factor F2 = topJock+CLM

Not Significant chi2 = 1.44

factor winners losers roi win% iv starters races
topJock+STKS 751 2895 0.84 21% 1.22 3646 2748
topJock+CLM 519 1852 0.80 22% 1.51 2371 1983


That is really interesting, but I must be a bit slow this morning. Where is the part where it is broken into class levels by individual jockey. Specifically, where the performance of jockeys as individuals--not all jockeys generically--is categorized as to success (a euphemism for winning)?

When the wrong questions are asked, the answers are invariably wrong. Or misleading.

Given the number (and success) of serious bettors using such factors as a significant component of their decision-making process, it is surprising that you would conclude it is insignificant.

[Dave Schwartz]

The value of information is in direct proportion to the number of people who have that information. Rather than endlessly mulling over the same pace and speed figures everyone else is looking at, those intent on earning a profit from wagering might do well to consider doing a bit more real work and a bit less mulling.

Traynor, this is true in the sense that they probably do not look at the numbers in the same format as we do.

However, we must remember that while the whales may use a factor with rigorous math behind it, the general public en mass uses so many factors that they (effectively) use it as well.

Speed is speed, whether you use the Beyer number or you make your own. Pace numbers are still pace numbers, those some clearly better than others.

In other words, the whales do not win because of the UNIQUENESS of their factors but because of their PROPER application of PRINCIPLES.

[traynor]

Traynor, this is true in the sense that they probably do not look at the numbers in the same format as we do.

However, we must remember that while the whales may use a factor with rigorous math behind it, the general public en mass uses so many factors that they (effectively) use it as well.

Speed is speed, whether you use the Beyer number or you make your own. Pace numbers are still pace numbers, those some clearly better than others.

In other words, the whales do not win because of the UNIQUENESS of their factors but because of their PROPER application of PRINCIPLES.

That is true to the degree that the whales use exactly the same data, from exactly the same source, in exactly the same way, as the majority of bettors. Add new (proprietary or restricted access) data to the mix, and it is no longer true. Add information derived in a manner inconsistent with the manner in which it is derived by the majority of bettors, and it is no longer true.

[raybo]

You know, all this gibberish about statistics is really becoming painful. Why don't one of you stats guys, that have done all this great research and already have cracked this game, just go ahead and post which factor or factors are the most important regarding hit rate, ROI, and profit. Stop beating around the bush and post it. After all, if all it takes is a little statistical research then hundreds or thousands should have already figured it out by now, so it can't be some big secret that will change the face of racing.

[DeltaLover]

Why don't one of you stats guys, that have done all this great research and already have cracked this game, just go ahead and post which factor or factors are the most important regarding hit rate, ROI, and profit.

Not yet Raybo, not yet...

After all, if all it takes is a little statistical research then hundreds or thousands should have already figured it out by now, so it can't be some big secret that will change the face of racing.

I would not call it exactly little and as it turns out there are not hundreds of thousands those who have figured it out by now, I am not sure even if they are even hundreds!

[DeltaLover]

That is really interesting, but I must be a bit slow this morning. Where is the part where it is broken into class levels by individual jockey. Specifically, where the performance of jockeys as individuals--not all jockeys generically--is categorized as to success (a euphemism for winning)?

When the wrong questions are asked, the answers are invariably wrong. Or misleading.

Given the number (and success) of serious bettors using such factors as a significant component of their decision-making process, it is surprising that you would conclude it is insignificant.

What the research shows is that bad jocks win with the same rate both in claimers and stakes or allowances. Despite this their ROI appears to be higher in claimers (for example for stakes is 0.59 while for claimers is 0.70). The specific report does not clarify if this is significant or not.

As far as top jocks are going there is not difference based in class levels (both in winning rate and ROIwise)..

I am not sure that the correct approach is to refine data down to individual jocks. I can see many cases where the available data will not be sufficient. There is nothing wrong by applied a research using high level filtering, what is important is to know how to interpret the outcome.

To summarize, based in my own approach jockeys make no significant difference to the bottom line of my betting strategy. Besides that, there is no doubt that top jocks are increasing the chances of a horse to win. The catch is that this represents a well known concept that is very well reflected to the odds, neutralizing its impact as I said before.

[Robert Fischer]

With jockeys, like most aspects, the public has a mediocre perception of reality. This perception is believed to be reality and is reflected in the wagering.

While specific races can occasionally be complex, the game is not all that cryptic in terms of general results, and the public generally catches on to their misperceptions relatively quickly.

[Tom]

You know, all this gibberish about statistics is really becoming painful. Why don't one of you stats guys, that have done all this great research and already have cracked this game, just go ahead and post which factor or factors are the most important regarding hit rate, ROI, and profit. Stop beating around the bush and post it. After all, if all it takes is a little statistical research then hundreds or thousands should have already figured it out by now, so it can't be some big secret that will change the face of racing.


Dave has graciously done that - many times.
In fact, again tonight, on this very topic!

[Magister Ludi]

You know, all this gibberish about statistics is really becoming painful. Why don't one of you stats guys, that have done all this great research and already have cracked this game, just go ahead and post which factor or factors are the most important regarding hit rate, ROI, and profit. Stop beating around the bush and post it. After all, if all it takes is a little statistical research then hundreds or thousands should have already figured it out by now, so it can't be some big secret that will change the face of racing.

Factors:

Odds
Overlays
Optimal Bet Sizing


Prerequisites:

PhD in a quantitative discipline
Minimum five years experience building automated trading platforms
The mindset of a quant. Mental math - you have five seconds – what is 15% of 165?

No more information forthcoming.

[DeltaLover]

Minimum five years experience building automated trading platforms

Speaking from personal experience, I really do not see how knowing how to build an automated trading platform helps in horse betting. I have never seen anything equivalent to FIX or Geneva or Watchtower or SFC or Bridge or Bloomberg or any kind of a netting engine or anything similar when betting horses.

[raybo]

Not yet Raybo, not yet...



I would not call it exactly little and as it turns out there are not hundreds of thousands those who have figured it out by now, I am not sure even if they are even hundreds!

I was being facetious, by saying "little", the fact is that people have been researching with statistics for many decades (lots and lots of research with statistics), and yet, there are still less than 2% of players worldwide making long term profit.

[raybo]

Dave has graciously done that - many times.
In fact, again tonight, on this very topic!

Yes, I am quite familiar with Dave's use of statistics, but Dave doesn't come on here and try to pound statistics into oblivion either, like a few here do. The people I am referring to state all this "junk" about what one should be doing, and that proper statistical research might prove this, or will disprove that, etc., without ever posting proof that their own "junk" has any value at all.

Research is research, statistical or otherwise, but the results of that research is only meaningful if you apply it properly, something that most do not, or cannot, do.

[raybo]

Factors:

Odds
Overlays
Optimal Bet Sizing


Prerequisites:

PhD in a quantitative discipline
Minimum five years experience building automated trading platforms
The mindset of a quant. Mental math - you have five seconds – what is 15% of 165?

No more information forthcoming.

I agree with your factors, but not your prerequisites, I have none of them, and don't need them to be profitable. But thanks for posting the factors, they are, really, all that matters. All the other stuff is just that, stuff. How you arrive at the 2nd and 3rd factors is, of course, variable and individually inconsistent, because we're dealing with animals and humans in racing, but if you get them right often enough, and at high enough prices, that's all that counts.

By the way, no other information is needed.

[Dave Schwartz]

That is true to the degree that the whales use exactly the same data, from exactly the same source, in exactly the same way, as the majority of bettors

Boy, do I disagree with that!

If one player uses the last race Beyer number and another player uses the last race speed rating from the DRF, there will be a high degree of correlation between their picks, despite the fact that the Beyer number will undoubtably generate more winners.

If a thousand players use the last race speed rating, from (say) 6 different sources, and a big player uses the last race speed rating from the best source (whatever that may be), the "voting" of the many (in the pool) will likely correlate even more so with the big player.

A large number of analyzers will, via their betting, normalize the impact of a factor.

I believe that source hardly matters across a large sample of races.

I do not believe that there is a speed number in existence that is good enough to beat the game. Now, perhaps there is a method of speed handicapping that might do that, but I do not believe that the success lies in the number itself.

I still believe, as I have for many years that in order to win you must do one of two things:

1. Do what the public is doing, but do it much better.
or
2. Do something different.

In this age, #1 is tough because of the whales. They are just so good at what they do (and what we are trying to do). I think racing is much like any other investment medium: The biggest and most successful have an edge and work to keep it.

#2 is difficult because coming up with "something different" is not so easy. I mean, there was a time when "pace analysis" was very different. Now, doing pace analysis has changed from being a #2 to being a #1.

For that matter, there was a time (way back) when speed handicapping was considered different.

However, using a different set of speed ratings... or even making your own... does not really qualify as "different." Perhaps better, but not different.

Being "better enough" (LOL - now there is a fractured phrase) is very difficult.

One of the things we try to do with HSH is a #2-type approach. That is, we look for "races like this one" to analyze because very few people - including the whales - have the ability to handicap with that level of local optimum, as their systems are naturally geared towards a more global optimum.

In other words, in the words of Wee Willie Keeler, we try to "hit it where they ain't."


Regards,
Dave Schwartz

[raybo]

Boy, do I disagree with that!

If one player uses the last race Beyer number and another player uses the last race speed rating from the DRF, there will be a high degree of correlation between their picks, despite the fact that the Beyer number will undoubtably generate more winners.

If a thousand players use the last race speed rating, from (say) 6 different sources, and a big player uses the last race speed rating from the best source (whatever that may be), the "voting" of the many (in the pool) will likely correlate even more so with the big player.

A large number of analyzers will, via their betting, normalize the impact of a factor.

I believe that source hardly matters across a large sample of races.

I do not believe that there is a speed number in existence that is good enough to beat the game. Now, perhaps there is a method of speed handicapping that might do that, but I do not believe that the success lies in the number itself.

I still believe, as I have for many years that in order to win you must do one of two things:

1. Do what the public is doing, but do it much better.
or
2. Do something different.

In this age, #1 is tough because of the whales. They are just so good at what they do (and what we are trying to do). I think racing is much like any other investment medium: The biggest and most successful have an edge and work to keep it.

#2 is difficult because coming up with "something different" is not so easy. I mean, there was a time when "pace analysis" was very different. Now, doing pace analysis has changed from being a #2 to being a #1.

For that matter, there was a time (way back) when speed handicapping was considered different.

However, using a different set of speed ratings... or even making your own... does not really qualify as "different." Perhaps better, but not different.

Being "better enough" (LOL - now there is a fractured phrase) is very difficult.

One of the things we try to do with HSH is a #2-type approach. That is, we look for "races like this one" to analyze because very few people - including the whales - have the ability to handicap with that level of local optimum, as their systems are naturally geared towards a more global optimum.

In other words, in the words of Wee Willie Keeler, we try to "hit it where they ain't."


Regards,
Dave Schwartz

Agree, it's the "application" of the available individual factors, not the individual factors themselves, that is important to profitability. And, "value" is the trump card.

[traynor]

Boy, do I disagree with that!

If one player uses the last race Beyer number and another player uses the last race speed rating from the DRF, there will be a high degree of correlation between their picks, despite the fact that the Beyer number will undoubtably generate more winners.

If a thousand players use the last race speed rating, from (say) 6 different sources, and a big player uses the last race speed rating from the best source (whatever that may be), the "voting" of the many (in the pool) will likely correlate even more so with the big player.

A large number of analyzers will, via their betting, normalize the impact of a factor.

I believe that source hardly matters across a large sample of races.

I do not believe that there is a speed number in existence that is good enough to beat the game. Now, perhaps there is a method of speed handicapping that might do that, but I do not believe that the success lies in the number itself.

I still believe, as I have for many years that in order to win you must do one of two things:

1. Do what the public is doing, but do it much better.
or
2. Do something different.

In this age, #1 is tough because of the whales. They are just so good at what they do (and what we are trying to do). I think racing is much like any other investment medium: The biggest and most successful have an edge and work to keep it.

#2 is difficult because coming up with "something different" is not so easy. I mean, there was a time when "pace analysis" was very different. Now, doing pace analysis has changed from being a #2 to being a #1.

For that matter, there was a time (way back) when speed handicapping was considered different.

However, using a different set of speed ratings... or even making your own... does not really qualify as "different." Perhaps better, but not different.

Being "better enough" (LOL - now there is a fractured phrase) is very difficult.

One of the things we try to do with HSH is a #2-type approach. That is, we look for "races like this one" to analyze because very few people - including the whales - have the ability to handicap with that level of local optimum, as their systems are naturally geared towards a more global optimum.

In other words, in the words of Wee Willie Keeler, we try to "hit it where they ain't."


Regards,
Dave Schwartz

I was referring specifically to data silos. The "differences" in interpretation of that data are relatively cosmetic and trivial. Minor tweaking and massaging aside, when everyone is looking at the same (basic, limited) set of data, similar conclusions are the norm.

[MJC922]

Agree, it's the "application" of the available individual factors, not the individual factors themselves, that is important to profitability. And, "value" is the trump card.

I'm all about quality of the data, true you have to apply it properly but for a lot of people out there it's the same old flawed stuff -- the GIGO principle applies. If I have to second guess the data that I'm using then I need to get out of the game and look into ways of generating and/or obtaining more reliable info to work from. I think people using weak data spread out into a lot of other handicapping factors primarily as a safety mechanism because they quickly lose trust in the data they have available from the usual sources. The topic of jockeys is a good example of what the majority have to work with data-wise. Let's start with the raw win percentages, that's the basic metric everyone including grandma uses to start with, I'm sure it's used by countless software and spreadsheets, and yeah break it into categories that's fine, small samples are easy to interpret, I suspect they make good talking points on pre-race shows. Much like the NFL team is 0-12 against the spread when there's a full moon on natural grass.

So the problem with jockey stats is really what about the horses? If you ride bad horses you're going to lose, the win% is flawed. Or racing luck, i.e. maybe the rider has 1 win in 15 on grass but 6 places on longshots, he can't ride? Then favorites are brought into this as an option because surely everyone should win 33% when they ride a favorite, the trouble is how much higher are the odds when a bad rider is up? It's built into the public's estimate.

When we really look to obtain more quality data, more interesting and more valuable data then we need to look into things like how much better do the horses as a group run when they switch to this specific jockey. Now you're starting to look at some electrifying (ahem) data. So I don't buy into it's all about how the factors are applied because there's plenty of weak data out there people accept at face value. The research to bring even partial answers to these questions is not easy to carry out. I see profitable handicappers in the future being able to manage quality data they obtain from other sources, largely, as the effort and time required to get these answers is increasingly impossible to carry out as a one man show.

[Stillriledup]

I'm all about quality of the data, true you have to apply it properly but for a lot of people out there it's the same old flawed stuff -- the GIGO principle applies. If I have to second guess the data that I'm using then I need to get out of the game and look into ways of generating and/or obtaining more reliable info to work from. I think people using weak data spread out into a lot of other handicapping factors primarily as a safety mechanism because they quickly lose trust in the data they have available from the usual sources. The topic of jockeys is a good example of what the majority have to work with data-wise. Let's start with the raw win percentages, that's the basic metric everyone including grandma uses to start with, I'm sure it's used by countless software and spreadsheets, and yeah break it into categories that's fine, small samples are easy to interpret, I suspect they make good talking points on pre-race shows. Much like the NFL team is 0-12 against the spread when there's a full moon on natural grass.

So the problem with jockey stats is really what about the horses? If you ride bad horses you're going to lose, the win% is flawed. Or racing luck, i.e. maybe the rider has 1 win in 15 on grass but 6 places on longshots, he can't ride? Then favorites are brought into this as an option because surely everyone should win 33% when they ride a favorite, the trouble is how much higher are the odds when a bad rider is up? It's built into the public's estimate.

When we really look to obtain more quality data, more interesting and more valuable data then we need to look into things like how much better do the horses as a group run when they switch to this specific jockey. Now you're starting to look at some electrifying (ahem) data. So I don't buy into it's all about how the factors are applied because there's plenty of weak data out there people accept at face value. The research to bring even partial answers to these questions is not easy to carry out. I see profitable handicappers in the future being able to manage quality data they obtain from other sources, largely, as the effort and time required to get these answers is increasingly impossible to carry out as a one man show.

I agree with you on your NFL example, a team is 0-12 under certain conditions, but the game that was "12 games ago" might have taken place 10 years ago, which means it has zero affect on today's game and yet, its part of the 0 for 12 that people are using to make decisions, which is obviously silly.

To me, watching the jocks ride, its more important than any 'statistic' i can come up with, the visuals tell a much more complete tale than the jocks record as it tells you how that jock looks on a horse TODAY. His record is a combination of his skill and the horsepower he has had.....and that really means nothing to us as far as his next mount goes. His "record" factors in the talent of the other horse's he's ridden, so its a mixture of his own talent and the horse talent, but part of the equation which is the talent of the horse's he's ridden at the meet, has nothing to do with this particular horse on this particular day...so, really, a jocks record is stale information as far as todays mount goes imo and can certainly be misleading.

[traynor]

I'm all about quality of the data, true you have to apply it properly but for a lot of people out there it's the same old flawed stuff -- the GIGO principle applies. If I have to second guess the data that I'm using then I need to get out of the game and look into ways of generating and/or obtaining more reliable info to work from. I think people using weak data spread out into a lot of other handicapping factors primarily as a safety mechanism because they quickly lose trust in the data they have available from the usual sources. The topic of jockeys is a good example of what the majority have to work with data-wise. Let's start with the raw win percentages, that's the basic metric everyone including grandma uses to start with, I'm sure it's used by countless software and spreadsheets, and yeah break it into categories that's fine, small samples are easy to interpret, I suspect they make good talking points on pre-race shows. Much like the NFL team is 0-12 against the spread when there's a full moon on natural grass.

So the problem with jockey stats is really what about the horses? If you ride bad horses you're going to lose, the win% is flawed. Or racing luck, i.e. maybe the rider has 1 win in 15 on grass but 6 places on longshots, he can't ride? Then favorites are brought into this as an option because surely everyone should win 33% when they ride a favorite, the trouble is how much higher are the odds when a bad rider is up? It's built into the public's estimate.

When we really look to obtain more quality data, more interesting and more valuable data then we need to look into things like how much better do the horses as a group run when they switch to this specific jockey. Now you're starting to look at some electrifying (ahem) data. So I don't buy into it's all about how the factors are applied because there's plenty of weak data out there people accept at face value. The research to bring even partial answers to these questions is not easy to carry out. I see profitable handicappers in the future being able to manage quality data they obtain from other sources, largely, as the effort and time required to get these answers is increasingly impossible to carry out as a one man show.

What people accept at face value is the old-time, "tried-and-true" interpretations (a jockey's "win percentage" for example, with zero critical thought about what that number is, what it indicates, and what it most definitely does NOT indicate). The uncritical acceptance of specific data points as containing important and significant information--to the point of completely ignoring any alternatives--is not especially useful. The uncritical acceptance of the simplistic representaions of "speed" and "pace" as the linchpins of prediction is comical. Readily available, easy to massage, and easy to present in formats that make them seem "new and innovative," yes. The best predictors? No.

Anyone who believes that "whales" are agonizing over "recent speed figures" or "pace matchups" really, really needs to get out of the house more. Why would they want to chase nickels and dimes and pathetic little margins when there are so many dollars out there waiting to be plucked?

[traynor]

I agree with you on your NFL example, a team is 0-12 under certain conditions, but the game that was "12 games ago" might have taken place 10 years ago, which means it has zero affect on today's game and yet, its part of the 0 for 12 that people are using to make decisions, which is obviously silly.

To me, watching the jocks ride, its more important than any 'statistic' i can come up with, the visuals tell a much more complete tale than the jocks record as it tells you how that jock looks on a horse TODAY. His record is a combination of his skill and the horsepower he has had.....and that really means nothing to us as far as his next mount goes. His "record" factors in the talent of the other horse's he's ridden, so its a mixture of his own talent and the horse talent, but part of the equation which is the talent of the horse's he's ridden at the meet, has nothing to do with this particular horse on this particular day...so, really, a jocks record is stale information as far as todays mount goes imo and can certainly be misleading.

Exactly. The significant point to be discerned is how this jockey will do on this specific horse, on this specific track, in this specific race--today. Not some simplistic evaluation based on history that is used for no apparent reason other than that it is easy to calculate and it is easy to pretend that it is important. Not quite as worthless as backfitting from last month's races to decide what to bet on today based on what happened last month, but real close.

Massaging numbers with a computer is simple to understand. Especially when the numbers can be downloaded at the click of a mouse. That does NOT make those numbers of primary importance in predicting race outcomes. Just simple to understand and easy to get.

[MJC922]

I agree with you on your NFL example, a team is 0-12 under certain conditions, but the game that was "12 games ago" might have taken place 10 years ago, which means it has zero affect on today's game and yet, its part of the 0 for 12 that people are using to make decisions, which is obviously silly.

To me, watching the jocks ride, its more important than any 'statistic' i can come up with, the visuals tell a much more complete tale than the jocks record as it tells you how that jock looks on a horse TODAY. His record is a combination of his skill and the horsepower he has had.....and that really means nothing to us as far as his next mount goes. His "record" factors in the talent of the other horse's he's ridden, so its a mixture of his own talent and the horse talent, but part of the equation which is the talent of the horse's he's ridden at the meet, has nothing to do with this particular horse on this particular day...so, really, a jocks record is stale information as far as todays mount goes imo and can certainly be misleading.

No disagreement about watching. When I know less than nothing about a 5% rider I get to watching a race before I trust any stats. I don't put much stock in stats for Jockeys, even with the best of intentions we just as well end up rating agents, so yes I trust my eyes far more than stats. I've seen some interesting stats supporting prior suspicions that I'd long held about a certain Jock but that's as far as it goes. There are riders like Theriot for instance I don't care about stats of any kind, IMO the guy is the hands of stone and I convince myself to play him when I think the horse measures up however it rarely leads to anything but deep regret. :)

[rubicon55]

No disagreement about watching. When I know less than nothing about a 5% rider I get to watching a race before I trust any stats. I don't put much stock in stats for Jockeys, even with the best of intentions we just as well end up rating agents, so yes I trust my eyes far more than stats. I've seen some interesting stats supporting prior suspicions that I'd long held about a certain Jock but that's as far as it goes. There are riders like Theriot for instance I don't care about stats of any kind, IMO the guy is the hands of stone and I convince myself to play him when I think the horse measures up however it rarely leads to anything but deep regret. :)

A contrarian viewpoint, possibly consider types of races that high percentage jockeys do not perform well at, could be used as a tie breaker between two good value horses.

[rubicon55]

Agree, it's the "application" of the available individual factors, not the individual factors themselves, that is important to profitability. And, "value" is the trump card.

Ray I have agreed with you on this point more than once and that this subject has been beat to death on this board but could be the most important factor IMO, with massive amounts of info available and all things considered equal IMO it really comes down to value since many players have the data but do not know how to use if effectively in wagering properly.

[raybo]

You know, you guys that feel that if one doesn't have an advanced degree in math and/or statistics has no chance of being profitable in racing, really need to "get out of the house more". I still have seen no examples of how any of you came to these conclusions, other than that you've got the degree so you must be better than those that don't have them. Sorry to deflate your balloon, but it's simply not true.

[castaway01]

I break my jockey stats into catagories - dirt sprints, routes, turf sprints, routes, poly sprints, routes, then by early speed. Then even more by specifics, like SA 6.5F races, Saratoga 5.5 Turf sprints, etc.

Honestly, I don't have the patience to do that because jockey percentages on stats like this are so much lower than trainer percentages, but if you can do it and it works for you, then go for it.

[castaway01]

You know, you guys that feel that if one doesn't have an advanced degree in math and/or statistics has no chance of being profitable in racing, really need to "get out of the house more". I still have seen no examples of how any of you came to these conclusions, other than that you've got the degree so you must be better than those that don't have them. Sorry to deflate your balloon, but it's simply not true.

Of course you're right. No one can prove anything unless they want to show their betting records, but from what we know of the most successful bettors, they weren't stats majors in grad school. I think the stats majors like to try to sound smart and post vaguely, but it has nothing to do with real-life wagering.

[castaway01]

This is a great post 01, i agree with this general premise. I love digging out a competent rider who doesnt get many chances, but can really ride when he does, you get great value that way, if you have a good horse to play, you don't need the leading rider, they pay off the same amount whether you win by a nose or win by 5 so if you're much the best, its better to take the higher price and hope that your calculations are correct on the horse.

You know, I have to say one thing. All the times I've ripped you for being a guy who just posts and never says anything useful, and you still say I made a useful post. I respect that. You may be smarter than I originally perceived.

Probably not. But maybe. ;)

[Dave Schwartz]

You know, you guys that feel that if one doesn't have an advanced degree in math and/or statistics has no chance of being profitable in racing, really need to "get out of the house more".

I don't think he means have no chance to win. I think he means you have no chance to WIN!!!

And he is probably right, to a great degree.

I think it not coincidental that the whales all employ quants. I would hire one myself if I could find one that would work for less that 20 dimes a month.

I can tell you that I do not possess enough of THAT horsepower to compete on THAT level.

[Tom]

Honestly, I don't have the patience to do that because jockey percentages on stats like this are so much lower than trainer percentages, but if you can do it and it works for you, then go for it.

Actually, you would be surprised - I have found many good situations.
Using HTR export, I have a very easy task if querying lots of filters quickly. I had one last year that a rider was very good at turf sprints, but when I looked at only his Saratoga stats, he was dismal over three seasons. Having the info at your fingertips during a race card is helpful. Most of the time, I admit, I find nothing really good or really bad, but that is important as well.

[traynor]

I don't think he means have no chance to win. I think he means you have no chance to WIN!!!

And he is probably right, to a great degree.

I think it not coincidental that the whales all employ quants. I would hire one myself if I could find one that would work for less that 20 dimes a month.

I can tell you that I do not possess enough of THAT horsepower to compete on THAT level.

Do what the dotcoms did (very successfully). Offer them a piece of the pie. It worked for Bill Gates, and quite a few others. They have the skills, and you have the smarts to know how to apply those skills. It is a great combination.

Most of the

[johnhannibalsmith]

You know, I have to say one thing. All the times I've ripped you for being a guy who just posts and never says anything useful, and you still say I made a useful post. I respect that. You may be smarter than I originally perceived.

Probably not. But maybe. ;)

You've discovered why he's one of my favorites even if I shake my head at 85% of what he posts.

[Dave Schwartz]

Do what the dotcoms did (very successfully). Offer them a piece of the pie. It worked for Bill Gates, and quite a few others. They have the skills, and you have the smarts to know how to apply those skills. It is a great combination.

Been trying that for YEARS.

To date, I have interviewed 26 people with the theoretical doctoral experience. Only three actually knew what I was talking when I told them what we were trying to accomplish. Never have I met so many people who live in a theoretical world, with absolutely no intention of ever doing the real thing.

Then there were the three. One said that he would work for $30k per month. The cheapest said that he would work for $20k per month and 20% of "the pie."

One of the three told me "Honestly, I don't think it can be done, so I am not willing to work on just the chance of getting paid."

[Robert Fischer]

may as well update.


Villanueva In his last 10 mounts
4 Wins (1.80-1*, 15.4-1, 6.0-1, 3.3-1)
2 Place (4.2-1, 12.9-1)
1 Shw (21.3-1)

no rides today with 17 races taking place at either Parx or Penn National.
Who is this kid's agent? What are these trainers watching when evaluating jocks?

Guess he must be making a living riding some of the workouts :confused:

[Stillriledup]

no rides today with 17 races taking place at either Parx or Penn National.
Who is this kid's agent? What are these trainers watching when evaluating jocks?

Guess he must be making a living riding some of the workouts :confused:

Very few are really watching and evaluating. Every circuit seems to have either overrated or washed up veteran jocks who continue to get and butcher live mounts, the toughest part of handicapping is when you see a live runner with an old, overrated, washed up jock and the dilemma is that you know the jock won't always ride bad, just half the time he'll screw something up, but the other half of the time, the live horse will drag him kicking and screaming to the wire first...do you toss out a live runner because the guy *might* act washed up or do you bet on the guy knowing there's a fairly decent change he will screw something up?

[flatstats]

Thanks much for responding to my question with such a clear response backed by examples.

In my earlier post in this thread I asked about the worth of a technique used to evaluate jocks. It was presented by Lee Russo (at the time, a very successful professional handicapper) at a handicapping expo which he claimed added a lot to his success. Briefly, he recorded a jocks performance in relation to their odds by adding/subtracting points based on race outcome. For example, if the jock was on the fav. and finished third, he subtracted two points from the jocks running total. If he was on the third choice and won, he added two points to the jocks running total. This was done without regard to further filters such as class, distance etc. I guess he felt that the additional filters had already been discounted in the odds.

In your opinion, do you think such a system has merit?

Why bother with points when the A/E is a perfectly fine key indicator?

A/E = actual wins / expected wins so can tell you if a jockey tends to be an overlay or underlay.

Popular jockeys may be overbet by the masses but may just have that extra ability to counteract. Similarly a journeyman jockey may be underbet but be useless in general. Thus how the jockey beats or loses to the odds is the key performance indicator.

Look at the Top 10 Jockeys based on A/E (value) at any time in the UK and you will see the table dominated by:

a) Apprentices
b) Female Jockeys

The A List jockeys rarely break the top 20.

[traynor]

Been trying that for YEARS.

To date, I have interviewed 26 people with the theoretical doctoral experience. Only three actually knew what I was talking when I told them what we were trying to accomplish. Never have I met so many people who live in a theoretical world, with absolutely no intention of ever doing the real thing.

Then there were the three. One said that he would work for $30k per month. The cheapest said that he would work for $20k per month and 20% of "the pie."

One of the three told me "Honestly, I don't think it can be done, so I am not willing to work on just the chance of getting paid."

You might consider contacting some of your serious wagering clients. Most wagering groups have access to (or are in contact with) contract developers who specialize in "sophisticated data analysis." That is, you may find what you are looking for in someone already involved in the field.

"Hiring" someone may not work so well. The people I know who do such work only do it on a contract basis with explicit objectives, scope, and deliverables. Katherine Jung is one of the best, but very pricey. Last I hear she was doing work on the Sha Tin/Happy Valley circuit. Her rates are high, but so is the quality of her work.

[traynor]

Been trying that for YEARS.

To date, I have interviewed 26 people with the theoretical doctoral experience. Only three actually knew what I was talking when I told them what we were trying to accomplish. Never have I met so many people who live in a theoretical world, with absolutely no intention of ever doing the real thing.

Then there were the three. One said that he would work for $30k per month. The cheapest said that he would work for $20k per month and 20% of "the pie."

One of the three told me "Honestly, I don't think it can be done, so I am not willing to work on just the chance of getting paid."

You also might try something like the Swedish forums (most Swedes speak English at least as well as most native English speakers) or even Craigslist. Some of the best computer science people around come out of the tech institute in Stockholm, or are currently attending there.

You might also consider doing what a LOT of developers do--join Monster and Dice and post a closed job description. PhD's don't impress me that much, so if it were me, I would tend to look more for "real world data analysts"--MBAs (MMIS) or MBA (MMIS) wannabes specializing in finance are really good places to start looking. So are current students in, or graduates of, the dual MBA/MMIS major at the University of Arizona.

[TrifectaMike]

I don't think he means have no chance to win. I think he means you have no chance to WIN!!!

And he is probably right, to a great degree.

I think it not coincidental that the whales all employ quants. I would hire one myself if I could find one that would work for less that 20 dimes a month.

I can tell you that I do not possess enough of THAT horsepower to compete on THAT level.

Bayesian Decision Model

[traynor]

Bayesian Decision Model

That looks a lot like Weick's sensemaking. Thanks for posting it. The diagram makes a lot more sense (to me) than (most attempts at) verbal explanations.

[Magister Ludi]

Bayesian Decision Model
Welcome back, my friend!

[TrifectaMike]

Welcome back, my friend!

With you (The master) here and some others, I'm in good company.

Mike

[Robert Fischer]

Domingo Chacaltana coming up on the super-solid Castle's Corner @ mountaineer r3.


I'm a little bit gun-shy with this jockey. I haven't scouted him but he's had some failed races

[Robert Fischer]

^4th @ even money



his wps probables were: 4.20 2.40 2.20

had good position early, failed to approach leader on critical backstretch area, appeared to "come up empty"



edit - wow... he sure was quiet on that horse, rode it like he was loaded for bear, a good replay to watch. Wasn't even asking late backstretch and the 4 came up and took his position under strong urging. I lucked out by passing that one.

[flatstats]

You know, all this gibberish about statistics is really becoming painful. Why don't one of you stats guys, that have done all this great research and already have cracked this game, just go ahead and post which factor or factors are the most important regarding hit rate, ROI, and profit. Stop beating around the bush and post it. After all, if all it takes is a little statistical research then hundreds or thousands should have already figured it out by now, so it can't be some big secret that will change the face of racing.

It will never change the face of racing. Because every year a new mug is introduced to racing.

Mugs are what make this game profitable. They are the ones that provide the bookmakers / tote operators with a reason to want to play the game. The smart punters are then able to feed off a small percentage off those bookmakers / tote operators.

Mugs come and go but they are always there. They are just replaced with new mugs and this is why you can profit from it.

Certainly here in the UK they continue to use old, outdated methods because that is what they read and hear from the racing pundits: i.e. Form. They will thus always be part of the 95% who lose in the long run.

---

The most important factor for determining the winner is the EPS. Next is the sire stats, followed by the trainer stats and then the jockey stats, which account for, say 10% of the input but it does depend on the type of race.

[raybo]

It will never change the face of racing. Because every year a new mug is introduced to racing.

Mugs are what make this game profitable. They are the ones that provide the bookmakers / tote operators with a reason to want to play the game. The smart punters are then able to feed off a small percentage off those bookmakers / tote operators.

Mugs come and go but they are always there. They are just replaced with new mugs and this is why you can profit from it.

Certainly here in the UK they continue to use old, outdated methods because that is what they read and hear from the racing pundits: i.e. Form. They will thus always be part of the 95% who lose in the long run.

---

The most important factor for determining the winner is the EPS. Next is the sire stats, followed by the trainer stats and then the jockey stats, which account for, say 10% of the input but it does depend on the type of race.

Maybe you guys in the UK can make a profit with just class and connections, but in the US, you better bring some more game to the game.

[Capper Al]

How would you prefer to follow jockeys, by win percent overall, by win percent for today's race niche, or by jockey's move from one horse to another. I like jockey moves.

[traynor]

How would you prefer to follow jockeys, by win percent overall, by win percent for today's race niche, or by jockey's move from one horse to another. I like jockey moves.

All the above and much more. As with most other things in racing, if it can be simply coded into an algorithm and automatically monitored--it is likely to be of very little value. What makes jockey and trainer studies useful (for wagering) is that the simplistic percentages are "incomplete representations of the reality they supposedly represent."

[flatstats]

I would forget win percentages. They are flawed due to field sizes. Forget ROI too as that is flawed by a rogue 100/1 winner.

Use the I.V. at the very least but better still is the A/E.

From what I have read over the years only Gordon Pine advocates the A/E for US racing.

[pondman]

How would you prefer to follow jockeys, by win percent overall, by win percent for today's race niche, or by jockey's move from one horse to another. I like jockey moves.

Early in my gambling escapades, I got involved with a group, which keyed on Pincay. If he jumped on a horse, especially a maiden, we would follow the walker and the connections to the windows. If they stood in line for the $50 dollar window, we knew something was up. We would try and strike up a conversation, "what do you like in the race." Most of the time the employees would give you some input. And then we'd watch to see if the connection were going to stand in front of the winner circle before the race.

In those days you could get information from going to morning works and milling around, as if you were a novice, and people would talk to you about their horses. I learned to pay attention to who was exercising a horse, because sometime those riders show up and hit home runs.

[Overlay]

I would forget win percentages. They are flawed due to field sizes. Forget ROI too as that is flawed by a rogue 100/1 winner.

Use the I.V. at the very least but better still is the A/E.

From what I have read over the years only Gordon Pine advocates the A/E for US racing.
I haven't re-looked A/E for a while. When I first ran across it, it seemed to me that data to calculate it would be hard to find (as compared to IV's), since (if I recall) you had to know the field size for each of the races in the sample -- not just the cumulative number of starters and winners associated with a given factor. Plus, it didn't seem to me that the added value of the information (as compared to IV's) justified the extra time and effort required. Is my recollection mistaken? Am I interpreting A/E incorrectly?

[pondman]

You know, you guys that feel that if one doesn't have an advanced degree in math and/or statistics has no chance of being profitable in racing, really need to "get out of the house more".

+++++

I agree with you on this one. Anyone who thinks there are no shenanigans in this game is a fool. Unless a player knows how to reasonably estimate these shenanigans, they will have no possibility of winning. However, don't underestimate some of the high end players. They are good with or without a high end system. With the right data, a quant system could keep a player more consistent, makes them more efficient. But their method needs to be sound. And they've got to be take the right position. They're not going to just collect data and have a black box spit out a bet every race.

In context with the thread, try and find yourself in a room full of drunk jockeys, preferably trying to eat in a rat hole such as the Ohio Valley. You'll find out-- most of them have sandbagged horses. Don't see it at the A level, because those people have more at stack, and would run this type of activity out of town. That's why I stick with the tracks with racing commissioners. Anyone who thinks they are going to run a quant system on a place such as mountaineer is a fool. That I know to be a fact. Been there. Done that. You can make money at that track, but you had better be on the grounds, drinking with the trainers and riders.

[raybo]

I would forget win percentages. They are flawed due to field sizes. Forget ROI too as that is flawed by a rogue 100/1 winner.

Use the I.V. at the very least but better still is the A/E.

From what I have read over the years only Gordon Pine advocates the A/E for US racing.

What is "A/E"?

[Overlay]

What is "A/E"?
www.netcapper.com/TrackTractsArchive/TT010223.htm

[raybo]

www.netcapper.com/TrackTractsArchive/TT010223.htm

Thanks Overlay, makes sense. :ThmbUp:

[traynor]

Quote from the reference cited above:

"What if you knew that horses that won their last start went off at odds lower than average? In fact, they do. That would mean that the horses in this group might be winning a lot because of their low odds. Their winning the last race might not have anything to do with it."

We all need more people like this pumping their money into the mutuel pools.

"When I lose, I console myself with the knowledge that it was probably because the horse I bet on failed to read the tote board."

[Capper Al]

www.netcapper.com/TrackTractsArchive/TT010223.htm

It is always good to re-read netcappers archives. That's one of my of my many favorites there.

Thanks

[raybo]

Quote from the reference cited above:

"What if you knew that horses that won their last start went off at odds lower than average? In fact, they do. That would mean that the horses in this group might be winning a lot because of their low odds. Their winning the last race might not have anything to do with it."

We all need more people like this pumping their money into the mutuel pools.

"When I lose, I console myself with the knowledge that it was probably because the horse I bet on failed to read the tote board."

Get a grip! That's not what he meant and if you don't know that then you're not nearly as smart, about racing, as some here think you are. But, thanks for the confirmation!

[Overlay]

I haven't re-looked A/E for a while. When I first ran across it, it seemed to me that data to calculate it would be hard to find (as compared to IV's), since (if I recall) you had to know the field size for each of the races in the sample -- not just the cumulative number of starters and winners associated with a given factor. Plus, it didn't seem to me that the added value of the information (as compared to IV's) justified the extra time and effort required. Is my recollection mistaken? Am I interpreting A/E incorrectly?
Accessing Pine's article itself again (which I couldn't do earlier this afternoon), I was reminded that my previous question concerning the usefulness of A/E in comparison to IV's had been availability of prior odds for each horse in the sample, rather than any consideration related to field size. (Sorry for the memory lapse.)

[traynor]

Something everyone who bets on races should read, and especially those who (implicitly or explicitly) accept the writing of others as "expert commentary." Bear in mind that those writing about jockeys, trainers, and other important aspects of race analysis may be essentially clueless, and far more interested in being considered "experts" than they are in providing useful ideas or information.

Unskilled and Unaware of It: How Difficulties in Recognizing One's Own Incompetence Lead to Inflated Self-Assessments


https://www.math.ucdavis.edu/~suh/metacognition.pdf

[Magister Ludi]

Something everyone who bets on races should read, and especially those who (implicitly or explicitly) accept the writing of others as "expert commentary." Bear in mind that those writing about jockeys, trainers, and other important aspects of race analysis may be essentially clueless, and far more interested in being considered "experts" than they are in providing useful ideas or information.

Unskilled and Unaware of It: How Difficulties in Recognizing One's Own Incompetence Lead to Inflated Self-Assessments


https://www.math.ucdavis.edu/~suh/metacognition.pdf (https://www.math.ucdavis.edu/%7Esuh/metacognition.pdf)

I wonder how many of the gods of handicapping have mumbled the parimutuel equivalent of "But I wore the juice".

[DeltaLover]

Quote from the reference cited above:

"What if you knew that horses that won their last start went off at odds lower than average? In fact, they do. That would mean that the horses in this group might be winning a lot because of their low odds. Their winning the last race might not have anything to do with it."

We all need more people like this pumping their money into the mutuel pools.

"When I lose, I console myself with the knowledge that it was probably because the horse I bet on failed to read the tote board."

Favorites who won their last race are significantly underbet compared to all the other favorites.

If I ever had to design a model focusing in high winning frequency I would certainly use this angle as one of the primary signal agents.

Of course this is not the best approach to the game since by relaxing your striking rate you can find much better betting angles that although are cashing far fewer tickets they still win disproportionate more money by the end of the year.

As I said before when it comes to favorites you should not be looking for positives (that tend to be overbet) but for negatives (that tend to be ignored).

[traynor]

Favorites who won their last race are significantly underbet compared to all the other favorites.

If I ever had to design a model focusing in high winning frequency I would certainly use this angle as one of the primary signal agents.

Of course this is not the best approach to the game since by relaxing your striking rate you can find much better betting angles that although are cashing far fewer tickets they still win disproportionate more money by the end of the year.

As I said before when it comes to favorites you should not be looking for positives (that tend to be overbet) but for negatives (that tend to be ignored).

This is an area in which the data can be (very productively) layered by class level. Specifically, there is a direct correlation between the probability of repeated (and multiple repeated) wins and the class level of the race. Few would expect a low-level claimer or maiden claimer to repeat a win, while many more would expect a "higher class" allowance or stakes entry to repeat a win. (That is a gross oversimplification for illustrative purposes.)

I think a lot of bettors realize the above, and bet (or refrain from betting) accordingly. Layering the data would enable a more insightful conclusion about a specific race or races, based on the class level. It could also be argued that the win percentage of a given horse is a key indicator of the probability of a repeat win. A horse with 4 wins in 7 races is more likely to be bet to repeat than a horse with 1 win (the last race) in 19 starts.

Layering data can be provide a lot of useful information, if it is done skillfully and thoughtfully.

[Robert Fischer]

Something everyone who bets on races should read, and especially those who (implicitly or explicitly) accept the writing of others as "expert commentary." Bear in mind that those writing about jockeys, trainers, and other important aspects of race analysis may be essentially clueless, and far more interested in being considered "experts" than they are in providing useful ideas or information.

Unskilled and Unaware of It: How Difficulties in Recognizing One's Own Incompetence Lead to Inflated Self-Assessments


https://www.math.ucdavis.edu/~suh/metacognition.pdf

the great thing about horse racing is that you get paid for being right.

the great thing about being an expert is that you get paid for broadcasting a perception.


2 different things entirely.

[traynor]

the great thing about horse racing is that you get paid for being right.

the great thing about being an expert is that you get paid for broadcasting a perception.


2 different things entirely.

The bad thing about horse racing is that most of the material disseminated on the topic (including that by self-proclaimed "experts") seems aimed at an audience unable to distinguish their subjective view of lala land from the real world. It is a neverending source of wonder to me that those who seem the most disconnected from reality are usually the first (and loudest) to proclaim "how things really are."

The bad thing about being an expert is that you have to continually seek out an audience naive and uncritical enough to accept your statements--regardless of veracity or value--as objective, insightful descriptions of a reality that exists outside your own head.

[raybo]

The bad thing about horse racing is that most of the material disseminated on the topic (including that by self-proclaimed "experts") seems aimed at an audience unable to distinguish their subjective view of lala land from the real world. It is a neverending source of wonder to me that those who seem the most disconnected from reality are usually the first (and loudest) to proclaim "how things really are."

The bad thing about being an expert is that you have to continually seek out an audience naive and uncritical enough to accept your statements--regardless of veracity or value--as objective, insightful descriptions of a reality that exists outside your own head.

Do you not consider yourself an "expert"? That seems to be the underlying thought process in every one of your posts. :rolleyes:

[Robert Fischer]

The bad thing about horse racing is that most of the material disseminated on the topic (including that by self-proclaimed "experts") seems aimed at an audience unable to distinguish their subjective view of lala land from the real world. It is a neverending source of wonder to me that those who seem the most disconnected from reality are usually the first (and loudest) to proclaim "how things really are."

I haven't had much luck finding reality based discussion on any complex topic, let alone horse racing. Health willing, I hope to one day have greater access to these rare people. Some of the specialists seem to be competent in their area of expertise, and much can be learned from that at least. You just can't expect much. With capital-based things like horse racing, at least you get monetarily rewarded for being right. If you want acknowledgement or mutual agreement, it is the wrong place to look. The crowd goes with the popular perception, whether it is true or not. You can explicitly outline an unpopular outcome before the race, and you will simply raise disagreement. And if that outcome which you foresaw does play out, do not expect acknowledgement or even discussion. The crowd will simply explain the unpopular outcome away, even if a fable is needed. After two or three such outcomes, the public generally adjusts their popular perception, and value is lost anyway. The reward for being right about horses is the money. If you want public approval, pick a favorite that everyone loves, and if he wins, you can have a social reward.

The bad thing about being an expert is that you have to continually seek out an audience naive and uncritical enough to accept your statements--regardless of veracity or value--as objective, insightful descriptions of a reality that exists outside your own head.
TV helps a great deal with this.
Next best is brand association.
Promote the popular choices as your top plays, and use several horses.
Helps to be attractive, and able to speak.

[Magister Ludi]

I haven't had much luck finding reality based discussion on any complex topic, let alone horse racing. Health willing, I hope to one day have greater access to these rare people.

Those who don't know, speak. Those who know, don't speak.

[traynor]

I haven't had much luck finding reality based discussion on any complex topic, let alone horse racing. Health willing, I hope to one day have greater access to these rare people. Some of the specialists seem to be competent in their area of expertise, and much can be learned from that at least. You just can't expect much. With capital-based things like horse racing, at least you get monetarily rewarded for being right. If you want acknowledgement or mutual agreement, it is the wrong place to look. The crowd goes with the popular perception, whether it is true or not. You can explicitly outline an unpopular outcome before the race, and you will simply raise disagreement. And if that outcome which you foresaw does play out, do not expect acknowledgement or even discussion. The crowd will simply explain the unpopular outcome away, even if a fable is needed. After two or three such outcomes, the public generally adjusts their popular perception, and value is lost anyway. The reward for being right about horses is the money. If you want public approval, pick a favorite that everyone loves, and if he wins, you can have a social reward.


TV helps a great deal with this.
Next best is brand association.
Promote the popular choices as your top plays, and use several horses.
Helps to be attractive, and able to speak.

"Reality" is an incredibly complex issue. Descriptions often wind up being little more than subjective opinions--that the person believes (often earnestly and sincerely) represent something other than his or her own internal perceptions. It is getting worse all the time--people seem less and less able to distinguish their perceptions from external reality--often to their detriment.

I think it got off to a big surge back in the old "positive thinking" days. So many people lied to themselves, "I am so wonderful! I am a WINNER!" that they actually started believing their own propaganda. Then cognitive dissonance kicks in, and they go to great lengths to maintain the illusion and to avoid any "disconfirming evidence to the contrary, no matter how compelling."

I know quite a few people who are "alive and aware in the here-and-now" rather than drifting aimlessly in lala land. Many of them play blackjack or handicap horse races for a living. I don't spend much time with the others.

Relying on your blackjack or handicapping skills for income (not cookie jar money) will quickly weed out the pretenders. Not as an indulgence, with all normal needs covered by other income sources, but as a vocation.

[traynor]

Those who don't know, speak. Those who know, don't speak.

I used to think that. Not much anymore. I think those who know are more than willing to speak, but no one asks the right questions, or no one asks in a manner that indicates both a desire to learn and a willingness to listen.

What most people want is validation of their preexising beliefs, not new information or new knowledge. Given that most people have heads filled with nonsense and illusions, it is small wonder that it seems as you stated above. Those doing the most talking are the ones pandering to the nonsense and illusions. They are also the most "popular."

As in, "He's brilliant! He agrees with me!"

[Segwin]

Then cognitive dissonance kicks in, and they go to great lengths to maintain the illusion and to avoid any "disconfirming evidence to the contrary, no matter how compelling."


Sounds like politics. ;)

[DeltaLover]

Then cognitive dissonance kicks in, and they go to great lengths to maintain the illusion and to avoid any "disconfirming evidence to the contrary, no matter how compelling."


Taynor nailed it here ! :ThmbUp: :ThmbUp:

[Robert Fischer]

"Reality" is an incredibly complex issue. Descriptions often wind up being little more than subjective opinions--that the person believes (often earnestly and sincerely) represent something other than his or her own internal perceptions. It is getting worse all the time--people seem less and less able to distinguish their perceptions from external reality--often to their detriment.

I think it got off to a big surge back in the old "positive thinking" days. So many people lied to themselves, "I am so wonderful! I am a WINNER!" that they actually started believing their own propaganda. Then cognitive dissonance kicks in, and they go to great lengths to maintain the illusion and to avoid any "disconfirming evidence to the contrary, no matter how compelling."

I know quite a few people who are "alive and aware in the here-and-now" rather than drifting aimlessly in lala land. Many of them play blackjack or handicap horse races for a living. I don't spend much time with the others.

Relying on your blackjack or handicapping skills for income (not cookie jar money) will quickly weed out the pretenders. Not as an indulgence, with all normal needs covered by other income sources, but as a vocation.

Players capable of doing things like horseplaying or blackjack for a living are extremely rare.
It seems that it should follow a power law, with a small percentage of winners and a long tail of non-winners.
I'm not sure whether the takeout in horseplaying shapes the power law even more steeply than the 20-80 kind of distribution. Haven't thought about it a ton, but superficially, it's a possibility.

Distributions aside, the nature of the parimutuel system with takeout, ensures that there will be few capable of making a living playing the races.

These winners are not only hard to come by, but if they value their livelihood, they aren't going to want to give others the food off their table. If there are such players among us online, they probably view this as more of a social aspect. Not only would it be to their detriment to give away the farm, these players would have to be so advanced in comparison to the average horseplayer, as to view many of these debates as childish.

[raybo]

Players capable of doing things like horseplaying or blackjack for a living are extremely rare.
It seems that it should follow a power law, with a small percentage of winners and a long tail of non-winners.
I'm not sure whether the takeout in horseplaying shapes the power law even more steeply than the 20-80 kind of distribution. Haven't thought about it a ton, but superficially, it's a possibility.

Distributions aside, the nature of the parimutuel system with takeout, ensures that there will be few capable of making a living playing the races.

These winners are not only hard to come by, but if they value their livelihood, they aren't going to want to give others the food off their table. If there are such players among us online, they probably view this as more of a social aspect. Not only would it be to their detriment to give away the farm, these players would have to be so advanced in comparison to the average horseplayer, as to view many of these debates as childish.

Many of these debates ARE childish.

Of course, nobody who has a "farm" is going to just give it away, but they can offer ways for others to get their own farm. There are so many ways to interpret data, one would have to post selections to give away the farm. Because, the method of selecting those picks would have to be interpreted and reproduced exactly the same way as the guy that owns the method. Very few would be able, or be willing, to do that.

There's very little danger of those who are successful to become unsuccessful, because they have that "something" that allows them to be successful, and that isn't likely to change.

[traynor]

Players capable of doing things like horseplaying or blackjack for a living are extremely rare.
It seems that it should follow a power law, with a small percentage of winners and a long tail of non-winners.
I'm not sure whether the takeout in horseplaying shapes the power law even more steeply than the 20-80 kind of distribution. Haven't thought about it a ton, but superficially, it's a possibility.

Distributions aside, the nature of the parimutuel system with takeout, ensures that there will be few capable of making a living playing the races.

These winners are not only hard to come by, but if they value their livelihood, they aren't going to want to give others the food off their table. If there are such players among us online, they probably view this as more of a social aspect. Not only would it be to their detriment to give away the farm, these players would have to be so advanced in comparison to the average horseplayer, as to view many of these debates as childish.

Well, yeah. The trick is like watching television or a movie--"suspension of judgement" (at least temporarily). That is, at the moment, one must act as if it is all incredibly important and meaningful. Probably not on a par with he-say, she-say or tales from the old days, but close. At least as important as politics, religion, literature, popular culture, and monster tractors pulling each other through the mud.

For the most part, it doesn't matter--most would not be able to use the information even if it were given. Especially if it involves work or thinking.

[traynor]

I think it is more a matter of concentration and effort than "secrets." Almost everyone believes that all he or she needs to do (to play professional level blackjack or wagering on horse races at a professional-level) is the same old same old they have always done(or that they are doing now) with a few "insider secrets" woven in. In short, lala land--with a lot more money. It ain't gonna happen.

To cash some tickets, to make a few bucks, even to take in a supplemental income--easy squeezy. To do it full time, without the "safety net" of external income and resources to fall back on, not very likely. It is not so much that people are incapable, as that they are way too comfortable doing things the same old same old way. That is the price of living in lala land.

Some of the sharpest handicappers (and blackjack counters) I have ever encountered were foreign students in business school (who needed the money and could not work because it would violate the provisions of their student visa status).

[traynor]

When Ken Uston wrote Million Dollar Blackjack, he "gave away" the strategy that his teams had used to rake in a LOT of money. Almost no one used it, because it was "too hard." They wanted to "play" blackjack, not work at it. The exact same situation exists in handicapping horse races. Everyone wants to learn a few quick and easy strategies so they can keep doing what they have always done--with a few more winners--and then slide back into lala land with the others of their kind.

[pondman]

There are so many ways to interpret data, one would have to post selections to give away the farm.

I could post selections of horses that had beyer's #s that were at least 9 points below the high, because that's fairly normal play for me. And I'd do +40% on these. And you know what would happen? Most of the members at PA wouldn't be able to reverse engineer it. Why is that? It's 6th grade, probably closer to 5th grade stuff.

In the context of the thread, jockeys don't matter. If you want some value, load up when the little known jockey is on a contender, and go against the crowd. Consistently bet against the super jockeys when you can.

[Maximillion]

When Ken Uston wrote Million Dollar Blackjack, he "gave away" the strategy that his teams had used to rake in a LOT of money. Almost no one used it, because it was "too hard." They wanted to "play" blackjack, not work at it. The exact same situation exists in handicapping horse races. Everyone wants to learn a few quick and easy strategies so they can keep doing what they have always done--with a few more winners--and then slide back into lala land with the others of their kind.

My opinion is the majority of the players on this board want to improve their game....(I know I do)...I will be so bold as to say you are included in that group.(perhaps not).To say that "everyone wants to learn a few quick and easy strategies" and are not willing to do the "work" necessary may be underestimating many of the posters here.

[traynor]

My opinion is the majority of the players on this board want to improve their game....(I know I do)...I will be so bold as to say you are included in that group.(perhaps not).To say that "everyone wants to learn a few quick and easy strategies" and are not willing to do the "work" necessary may be underestimating many of the posters here.

Perhaps. It seems more an issue of supporting preexisting opinions than it is an issue of cashing more tickets. While cashing more tickets may be the stated reason, when the process involves abandoning, discarding, or substantially changing existing beliefs or "knowledge" few seem able (or willing) to make the adjustments needed.

When I started posting again a year or so ago, I was intent on developing (and posting on this board) a free app--with included methodology--that would enable a novice or recreational bettor to win some profit. It was a recycling of a pattern-recognition technology I used (and use) for a card-counting/strategy trainer for blackjack teams. I thought it was a great idea. It was not well-received. The whole notion that a newbie could learn to make a profit from wagering in less than 20 years of losing, pain, and agony seemed to ruffle way too many tailfeathers.

[raybo]

I could post selections of horses that had beyer's #s that were at least 9 points below the high, because that's fairly normal play for me. And I'd do +40% on these. And you know what would happen? Most of the members at PA wouldn't be able to reverse engineer it. Why is that? It's 6th grade, probably closer to 5th grade stuff.

In the context of the thread, jockeys don't matter. If you want some value, load up when the little known jockey is on a contender, and go against the crowd. Consistently bet against the super jockeys when you can.

I wasn't talking about reverse engineering your selections, I was talking about giving away your profit, if you have one. I agree, you could probably give away your complete process and most could not, or would not reproduce it.

[raybo]

Perhaps. It seems more an issue of supporting preexisting opinions than it is an issue of cashing more tickets. While cashing more tickets may be the stated reason, when the process involves abandoning, discarding, or substantially changing existing beliefs or "knowledge" few seem able (or willing) to make the adjustments needed.

When I started posting again a year or so ago, I was intent on developing (and posting on this board) a free app--with included methodology--that would enable a novice or recreational bettor to win some profit. It was a recycling of a pattern-recognition technology I used (and use) for a card-counting/strategy trainer for blackjack teams. I thought it was a great idea. It was not well-received. The whole notion that a newbie could learn to make a profit from wagering in less than 20 years of losing, pain, and agony seemed to ruffle way too many tailfeathers.

Really? I don't recall you ever posting anything of the sort. If you had posted such a process, maybe you would be taken much more seriously than you are. All I've heard from you is put downs and conjecture towards anyone who doesn't go along with your theoretical posings.

[thaskalos]

Perhaps. It seems more an issue of supporting preexisting opinions than it is an issue of cashing more tickets. While cashing more tickets may be the stated reason, when the process involves abandoning, discarding, or substantially changing existing beliefs or "knowledge" few seem able (or willing) to make the adjustments needed.

When I started posting again a year or so ago, I was intent on developing (and posting on this board) a free app--with included methodology--that would enable a novice or recreational bettor to win some profit. It was a recycling of a pattern-recognition technology I used (and use) for a card-counting/strategy trainer for blackjack teams. I thought it was a great idea. It was not well-received. The whole notion that a newbie could learn to make a profit from wagering in less than 20 years of losing, pain, and agony seemed to ruffle way too many tailfeathers.

That's an unfair assertion, IMO.

I think we would all love to see you take up something like that...and I am sure that no feathers whatsoever will be ruffled.

I do remember you having to endure a little flack for saying that you could convert a novice into a winner in a matter of a few weeks...but, IMO, it was all rather tame compared to some of the other arguments we have had on this site.

I, for one, hope that you will reconsider...and go ahead with the original plan.

[Tom]

"But, wait! There's more!"

:rolleyes:

[raybo]

"But, wait! There's more!"

:rolleyes:

Yeah, there's more, he'll have to come up with new reasons for not posting it. God save us!

[traynor]

That's an unfair assertion, IMO.

I think we would all love to see you take up something like that...and I am sure that no feathers whatsoever will be ruffled.

I do remember you having to endure a little flack for saying that you could convert a novice into a winner in a matter of a few weeks...but, IMO, it was all rather tame compared to some of the other arguments we have had on this site.

I, for one, hope that you will reconsider...and go ahead with the original plan.

Overall, I think the "available literature" on thoroughbred handicapping is remarkably shallow, and not very useful. If a reasonably astute researcher were to perform a literature review to determine the "existing body of knowledge"--if that review included a critique of research methodologies employed in generating the material--that researcher would have to conclude that there were serious deficiencies in the material available.

However, no one except the "whales" and their "quants" seem the least bit interested in engaging in serious research or data analysis to correct those deficiencies. The whales take the lion's share of the mutuel pools, and the rest chase after the few crumbs remaining.

What whales do is fundamentally different than what everyone else does. Not because they "all hire quants" but because they operate under a different paradigm, uncorrupted by numerous fallacies and logical errors that are routine in the race analysis of others.

What many seem unable to grasp is that what the whales do is NOT simply an extension of the same thing everyone else is doing--except with more complex computer applications and better data analysis. It is a different way of looking at races, a different way of looking at data, and a very different way of interpreting that data. It is that difference that causes the "conflict."

One person believes all they need do is "make a few adjustments" to his or her computer application and "seven figure incomes from churning the rebate shops" are available. Another believes that back-fitting to last month's results is a reasonable, rational approach to data analysis. Yet another believes that statistics and mathematics only apply in "ivory towers of academia"--in the "real world" small sample models are the stuff of which fortunes are made and success is just over the horizon. And on and on and on.

Is it possible to teach a newbie to wager profitably in a few weeks? Of course it is, because a newbie does not drag along the excess baggage of "racing experience in the real world" that exerts such harmful influence on the "experienced bettors." He or she only needs to know which races can be profitably leveraged, and given a (relatively simple) approach that will isolate the most likely winners within that segment of races. Not every race, at every track, every day. Consider it a watered-down version of the approach used by whales to drain the mutuel pools.

None of it is based on "Beyer speed numbers" or "back class" or "pace profiles" or any of the other silliness that seems to provide such endless hours of amusement for the hobbyists and "recreational bettors." Why? Because there is abundant evidence that the "conventional" approaches are inadequate. Doing the same thing everyone else is doing, even in slightly different ways, is still doing the same thing everyone else is doing--chasing the crumbs left by the whales while fantasizing about getting the whole pie. Or at least a big piece of it.

Winning is not that difficult. Winning while endlessly mulling over the results of old races and old wagers from years past and the "theories" of Quirin, Beyer, Brohamer, Sartin, Ainslie, and others--that may have worked at the time they were first written but are now pathetically outdated--is what is impossible.

Buena suerte, y tenga un buen viaje!

[magwell]

Winning while endlessly mulling over the results of old races and old wagers from years past and the "theories" of Quirin, Beyer, Brohamer, Sartin, Ainslie, and others--that may have worked at the time they were first written but are now pathetically outdated--is what is impossible...... :ThmbUp::ThmbUp:

[raybo]

Another dodge! This guy is so single minded and egotistical it's painful to read his posts. So, he will join en exclusive list of 2. Congratulations Traynor! You just made my ignore list. You might impress some here but it's obvious to me that you couldn't show a profit in this game if your life depended on it. Adios!

[TrifectaMike]

Favorites who won their last race are significantly underbet compared to all the other favorites.

If I ever had to design a model focusing in high winning frequency I would certainly use this angle as one of the primary signal agents.

Of course this is not the best approach to the game since by relaxing your striking rate you can find much better betting angles that although are cashing far fewer tickets they still win disproportionate more money by the end of the year.

As I said before when it comes to favorites you should not be looking for positives (that tend to be overbet) but for negatives (that tend to be ignored).

On the subject of last time winners, races with only one last race winner has different stats than races with two or more last time winners. Races with two or more last race winners I call "inertia" races. Often times that odds are very good.

A little something, you guys may want to study. It'll take a little work to make it "useful", but then nothing is for free.

Mike

[DeltaLover]

Of course you are right Doc.

What is more interesting, is races where all the horses look awfull. Usually cheap claimers, where none of the starters has finished better than 5th and even better none of them has ever been on the lead at any call. These type of races are among my best bets recently and are the very first to consider for betting.

[traynor]

On the topic of ease of use:

"... it will take you only a few days to become so adept with this method that you literally wiil be able to handicap any race in less than five minutes and, in that short time, cover more important factors than the ordinary player touches in a half hour."

On the topic of profitability:
" losing days were remarkably infrequent ... one series of three days without a single winner ... (T)wo losing days in succession were encountered a few times at every meeting, but results invariably improved on the third day ... (F)rom April 15 to the very end of the New York season in December, only three widely separated weeks ended in the red ... showed a substantial profit at every New York meeting from April through December ... picked winners at the rate of 29 percent ... the prices were often remarkably high ..."

Huh. Imagine that. Might be something a newbie could use to get off to a flying start, and no 20 years of practice losing necessary.

[thaskalos]

On the topic of ease of use:

"... it will take you only a few days to become so adept with this method that you literally wiil be able to handicap any race in less than five minutes and, in that short time, cover more important factors than the ordinary player touches in a half hour."

On the topic of profitability:
" losing days were remarkably infrequent ... one series of three days without a single winner ... (T)wo losing days in succession were encountered a few times at every meeting, but results invariably improved on the third day ... (F)rom April 15 to the very end of the New York season in December, only three widely separated weeks ended in the red ... showed a substantial profit at every New York meeting from April through December ... picked winners at the rate of 29 percent ... the prices were often remarkably high ..."

Huh. Imagine that. Might be something a newbie could use to get off to a flying start, and no 20 years of practice losing necessary.

Reminds me of the Dan Pope literature I used to get in the mail in the '90s...

Only the sales pitch is missing... :)

[Phantombridgejumpe]

I don't want to be a whale.

My ultimate goal is to play the game and try to win here and there and have some fun and waste some time.

For most players it is just a hobby, not a job.

Not a perfect comparison, but to me the whales feel like a good rock band that has sold out to corporate interests. You didn't get any better or worse technically, but you took the fun out of it and you sold out.

[rubicon55]

Overall, I think the "available literature" on thoroughbred handicapping is remarkably shallow, and not very useful. If a reasonably astute researcher were to perform a literature review to determine the "existing body of knowledge"--if that review included a critique of research methodologies employed in generating the material--that researcher would have to conclude that there were serious deficiencies in the material available.

However, no one except the "whales" and their "quants" seem the least bit interested in engaging in serious research or data analysis to correct those deficiencies. The whales take the lion's share of the mutuel pools, and the rest chase after the few crumbs remaining.

What whales do is fundamentally different than what everyone else does. Not because they "all hire quants" but because they operate under a different paradigm, uncorrupted by numerous fallacies and logical errors that are routine in the race analysis of others.

What many seem unable to grasp is that what the whales do is NOT simply an extension of the same thing everyone else is doing--except with more complex computer applications and better data analysis. It is a different way of looking at races, a different way of looking at data, and a very different way of interpreting that data. It is that difference that causes the "conflict."

One person believes all they need do is "make a few adjustments" to his or her computer application and "seven figure incomes from churning the rebate shops" are available. Another believes that back-fitting to last month's results is a reasonable, rational approach to data analysis. Yet another believes that statistics and mathematics only apply in "ivory towers of academia"--in the "real world" small sample models are the stuff of which fortunes are made and success is just over the horizon. And on and on and on.

Is it possible to teach a newbie to wager profitably in a few weeks? Of course it is, because a newbie does not drag along the excess baggage of "racing experience in the real world" that exerts such harmful influence on the "experienced bettors." He or she only needs to know which races can be profitably leveraged, and given a (relatively simple) approach that will isolate the most likely winners within that segment of races. Not every race, at every track, every day. Consider it a watered-down version of the approach used by whales to drain the mutuel pools.

None of it is based on "Beyer speed numbers" or "back class" or "pace profiles" or any of the other silliness that seems to provide such endless hours of amusement for the hobbyists and "recreational bettors." Why? Because there is abundant evidence that the "conventional" approaches are inadequate. Doing the same thing everyone else is doing, even in slightly different ways, is still doing the same thing everyone else is doing--chasing the crumbs left by the whales while fantasizing about getting the whole pie. Or at least a big piece of it.

Winning is not that difficult. Winning while endlessly mulling over the results of old races and old wagers from years past and the "theories" of Quirin, Beyer, Brohamer, Sartin, Ainslie, and others--that may have worked at the time they were first written but are now pathetically outdated--is what is impossible.

Buena suerte, y tenga un buen viaje!

I have heard something similar in martial arts from a former teacher in that it is difficult to train someone with former martial arts training since they sometimes have picked up bad habits along the way and automatically want to revert back. My instructor said he would rather have someone without any training. He said it was difficult for him to try and retrain someone who is entrenched in bad techniques. IMO it is a mental and also physically learned behavior. Found this interesting.

[traynor]

I don't want to be a whale.

My ultimate goal is to play the game and try to win here and there and have some fun and waste some time.

For most players it is just a hobby, not a job.

Not a perfect comparison, but to me the whales feel like a good rock band that has sold out to corporate interests. You didn't get any better or worse technically, but you took the fun out of it and you sold out.

I disagree. That assumes the whales have bean counter mentalities, drudging after the money each and every day, as a surrogate for a real life. That is also the complete opposite of the serious bettors I know and have known. Well, most of them anyway. There are a few that seem to think letting a day go by without obsessing about money is like robbing themselves, but they are anomalies.

The one thing the whales have that the little fish lack is the opportunity to go places and do things that the little fish can only fantasize about.

[Tom]

Huh. Imagine that. Might be something a newbie could use to get off to a flying start, and no 20 years of practice losing necessary.

Still waiting for the talk to stop and the walk to start. And the abundant proof that nothing works anymore. I know people making money off of figures and pace scenarios and have seen plenty of examples of it here. So far, you have done a lot of talking and not much else.

[Phantombridgejumpe]

We probably all care about both, and neither answer is wrong - but I don't think many true whales would say the sport.

[traynor]

Still waiting for the talk to stop and the walk to start. And the abundant proof that nothing works anymore. I know people making money off of figures and pace scenarios and have seen plenty of examples of it here. So far, you have done a lot of talking and not much else.

If you and the people you know are making money using figures and pace scenarios, I wish you all well and much continued success. If you figure out a way that a relative novice can turn a profit using figures and pace scenarios within a reasonable period of time--without prolonged losing streaks or substantial investment in software and data downloads (that are a complete turnoff to "new" bettors)--please do those novices a favor and post your recommendations. I don't need them, but I am sure there are more than a few readers who do, and would very much appreciate them.

[traynor]

I have heard something similar in martial arts from a former teacher in that it is difficult to train someone with former martial arts training since they sometimes have picked up bad habits along the way and automatically want to revert back. My instructor said he would rather have someone without any training. He said it was difficult for him to try and retrain someone who is entrenched in bad techniques. IMO it is a mental and also physically learned behavior. Found this interesting.

The same is true in many, many fields. People who learn a specific way of doing things (not necessarily bad, just specific) may often unconsciously "support" that way of doing things, and "resist" anything that conflicts with the previous way of doing things. It has a lot to do with cognitive dissonance.

Martial arts are especially vulnerable to "set patterns" because the actions and responses are "semi-automated" in the sense that there is little conscious choice involved.

[traynor]

Reminds me of the Dan Pope literature I used to get in the mail in the '90s...

Only the sales pitch is missing... :)

The premise is a simple one, but remarkably sophisicated at the same time. Specifically, there are attributes that indicate a potential winner. Those attributes can be combined to create a synergy. The underlying logic is sound. What was lacking was the data analysis necessary to determine which attributes are significant, and the most effective weighting of those attributes in relationships with each other. Originally, the developer believed that simply adding the values together would be sufficient. It was--for a time.

Using the basic premises (rather than uncritically copying the attributes and weighting of others) required considerable research (to determine the most effective attributes by track, distance, class, etc.) and considerable regression analysis to determine the most effective weighting to assign to each attribute and to each combination of attributes.

The attributes change. So do the weightings assigned to the attributes and combinations of attributes. It takes some pretty serious programming and data analysis to stay on top of it.

However, a simplified version can be learned fairly easily by a novice--no PhDs or quants required--and successfully applied. It was that simplified version that I had in mind when I suggested a novice could learn to be profitable in a relatively short period of time.

Igor Kusuyshyn developed a similar approach for harness races, and some still use (continually adjusted, updated, and modified versions of) Kusyshyn's basic approach successfully.

[OCF]

The premise is a simple one, but remarkably sophisicated at the same time. Specifically, there are attributes that indicate a potential winner. Those attributes can be combined to create a synergy. The underlying logic is sound. What was lacking was the data analysis necessary to determine which attributes are significant, and the most effective weighting of those attributes in relationships with each other. Originally, the developer believed that simply adding the values together would be sufficient. It was--for a time.

Using the basic premises (rather than uncritically copying the attributes and weighting of others) required considerable research (to determine the most effective attributes by track, distance, class, etc.) and considerable regression analysis to determine the most effective weighting to assign to each attribute and to each combination of attributes.

The attributes change. So do the weightings assigned to the attributes and combinations of attributes. It takes some pretty serious programming and data analysis to stay on top of it.

However, a simplified version can be learned fairly easily by a novice--no PhDs or quants required--and successfully applied. It was that simplified version that I had in mind when I suggested a novice could learn to be profitable in a relatively short period of time.

Igor Kusuyshyn developed a similar approach for harness races, and some still use (continually adjusted, updated, and modified versions of) Kusyshyn's basic approach successfully.

Perhaps you could tell us just one or two of the attributes to which you are referring, and keep the the other attributes and the weighting and combinations to yourself?

I don't necessarily disagree with your premise, but without at least some crumb it is very difficult to get any kind of handle on what you're referring to.

[Tom]

If you and the people you know are making money using figures and pace scenarios, I wish you all well and much continued success. If you figure out a way that a relative novice can turn a profit using figures and pace scenarios within a reasonable period of time--without prolonged losing streaks or substantial investment in software and data downloads (that are a complete turnoff to "new" bettors)--please do those novices a favor and post your recommendations. I don't need them, but I am sure there are more than a few readers who do, and would very much appreciate them.

Nice dodge.
I never made that claim - you did.

[traynor]

Nice dodge.
I never made that claim - you did.

I am not referring to claims of any kind (other than your assertion of profitability using figures and pace scenarios). I am referring specifically to a way to encourage new bettors to participate in thoroughbred racing by bypassing the long years of painstaking "learning," prolonged losing streaks, and substantial investment in software and data downloads that many seem to believe is essential to just break even, much less earn a profit.

The method I referred to is a simplified version of a (much) more complex approach used successfully by several wagering syndicates (that would most likely qualify as "whales").

[traynor]

Perhaps you could tell us just one or two of the attributes to which you are referring, and keep the the other attributes and the weighting and combinations to yourself?

I don't necessarily disagree with your premise, but without at least some crumb it is very difficult to get any kind of handle on what you're referring to.

The quotes are from Chapter 37, Figure Handicapping, from Theory and Practice of Handicapping. Tom Ainslie.

You may recall I mentioned that one of my first serious endeavors in handicapping was a structured version of The Compleat Horseplayer. The factors and values Ainslie recommended in that work, and in Theory and Practice of Handicapping, are seriously outdated. The underlying logic is sound.

A reasonably astute developer (or handicapper) should be able to start with the factors advocated by Ainslie, and adjust/modify/add to/subtract from those factors for current racing. And do very, very well--while pretty much ignoring figures and pace scenarios. At least until the earning potential of thoroughbred racing can be established enough for normal human greed to kick in and make the expense and effort involved in more complex approaches to race analysis both feasible and acceptable.

[OCF]

The quotes are from Chapter 37, Figure Handicapping, from Theory and Practice of Handicapping. Tom Ainslie.

You may recall I mentioned that one of my first serious endeavors in handicapping was a structured version of The Compleat Horseplayer. The factors and values Ainslie recommended in that work, and in Theory and Practice of Handicapping, are seriously outdated. The underlying logic is sound.

A reasonably astute developer (or handicapper) should be able to start with the factors advocated by Ainslie, and adjust/modify/add to/subtract from those factors for current racing. And do very, very well--while pretty much ignoring figures and pace scenarios. At least until the earning potential of thoroughbred racing can be established enough for normal human greed to kick in and make the expense and effort involved in more complex approaches to race analysis both feasible and acceptable.

OK, thanks, it may take me a while to get my hands on those books but I'll give it a shot.

[traynor]

On the topic of ease of use:

"... it will take you only a few days to become so adept with this method that you literally wiil be able to handicap any race in less than five minutes and, in that short time, cover more important factors than the ordinary player touches in a half hour."

On the topic of profitability:
" losing days were remarkably infrequent ... one series of three days without a single winner ... (T)wo losing days in succession were encountered a few times at every meeting, but results invariably improved on the third day ... (F)rom April 15 to the very end of the New York season in December, only three widely separated weeks ended in the red ... showed a substantial profit at every New York meeting from April through December ... picked winners at the rate of 29 percent ... the prices were often remarkably high ..."

Huh. Imagine that. Might be something a newbie could use to get off to a flying start, and no 20 years of practice losing necessary.

The quotes I referred to are above. Not wild-eyed marketing puffery from a sleazy system seller, but rather comments by a person widely referred to as "the dean of thoroughbred handicappers."

Take special note of the bolded quote. The "claim" that--with an adequate methodology--a bettor could very quickly learn to make a profit from wagering was not mine (originally). It was Ainslie's.

[traynor]

OK, thanks, it may take me a while to get my hands on those books but I'll give it a shot.

If you can find it, Ainslie's Private Method is also quite useful as a starting point.

[traynor]

OK, thanks, it may take me a while to get my hands on those books but I'll give it a shot.

A word of advice. Whether called factors, attributes, or indicators the data points are studied in isolation, then in combination. The combinations are factorial, and each represents a different scenario. Specifically, A + B + C is a multitude of possible variations--A when B does and does not exist within a given range, AND when C does and does not exist within a given range for each option for B.

Also, the strength of Ainslie's logic is that it was developed at a time when "normal" was betting at one track--not clicking around online among a hundred tracks on several continents. The values and weightings are (at the very least) track specific and distance-at-that-track specific. If you are using a computer, you might want to break them down further by class level.

If you focus on one track, it can be done fairly easily by hand (pen-and-paper). It is WELL worth the effort.

[OCF]

A word of advice. Whether called factors, attributes, or indicators the data points are studied in isolation, then in combination. The combinations are factorial, and each represents a different scenario. Specifically, A + B + C is a multitude of possible variations--A when B does and does not exist within a given range, AND when C does and does not exist within a given range for each option for B.

Also, the strength of Ainslie's logic is that it was developed at a time when "normal" was betting at one track--not clicking around online among a hundred tracks on several continents. The values and weightings are (at the very least) track specific and distance-at-that-track specific. If you are using a computer, you might want to break them down further by class level.

If you focus on one track, it can be done fairly easily by hand (pen-and-paper). It is WELL worth the effort.

Sounds interesting. My tendency is to bet too seldom rather than too often, so the focus on one track shouldn't be a problem.

[traynor]

Sounds interesting. My tendency is to bet too seldom rather than too often, so the focus on one track shouldn't be a problem.

That is a good tendency. A bit like hunting with a single-shot rifle. Or a bow.

[DeltaLover]

The combinations are factorial, and each represents a different scenario. Specifically, A + B + C is a multitude of possible variations--A when B does and does not exist within a given range, AND when C does and does not exist within a given range for each option for B.



I have my doubts when it comes to the effectiveness of complex factors.

The crowd bets using simple scenarios; going a single step further might reveal profitable situations the same does not hold true for very complicated filters involving many layers of indirection.

One of the reasons why this happens, has to do with the fact that a complex scenario, is almost never directly reflected to the pools making it difficult to use it to find errors in the crowd's estimation.

A very simple scenario, such as an E8 horse in a race full of speed that its class inferiority makes the crowd to underestimate it chances, presents much more value than some extreme A and B and not C and (D or E) type of factor that not only is completely transparent to most of the bettors but also is very difficult to verify due to the large data set needed to derive meaningful conclusions.

Following this board for long time, makes me to believe that the most under looked aspect of the game, is its chaotic nature, something that makes it impossible to completely rule out any starter of any race.

People love to talk about 'contenders' and 'throw outs', training their selective, memory to quickly forget races where a complete throw out won the race while emphasizing those they were proven correct. An immediate side effect of this can be found in the desire to 'pick the winner' something that I am convinced it represents the weakest approach to horse betting.

[traynor]

I have my doubts when it comes to the effectiveness of complex factors.

The crowd bets using simple scenarios; going a single step further might reveal profitable situations the same does not hold true for very complicated filters involving many layers of indirection.

One of the reasons why this happens, has to do with the fact that a complex scenario, is almost never directly reflected to the pools making it difficult to use it to find errors in the crowd's estimation.

A very simple scenario, such as an E8 horse in a race full of speed that its class inferiority makes the crowd to underestimate it chances, presents much more value than some extreme A and B and not C and (D or E) type of factor that not only is completely transparent to most of the bettors but also is very difficult to verify due to the large data set needed to derive meaningful conclusions.

Following this board for long time, makes me to believe that the most under looked aspect of the game, is its chaotic nature, something that makes it impossible to completely rule out any starter of any race.

People love to talk about 'contenders' and 'throw outs', training their selective, memory to quickly forget races where a complete throw out won the race while emphasizing those they were proven correct. An immediate side effect of this can be found in the desire to 'pick the winner' something that I am convinced it represents the weakest approach to horse betting.

I understand what you are saying, and agree with it to a degree. The digression is in the definition of "pick the winner." I think that such attempts that utilize the most commonly accepted processes of speed, pace, "class," and (especially) "tote board movement" will generally prove non-productive in the long-term.

By "non-productive" I mean that one may win occasionally, but not enough to be worth serious wagering over an extended period. More specifically, "professional-level" wagering using such criteria is flaky, and anyone who doesn't understand that probably does not know as many "serious bettors" or "whales" as they may claim to know. Dumping large amounts of money into mutuel pools chasing the "figure horse" or the "class of the field" is a fool's game.

However, when the criteria used to "pick the winner" differs considerably from the criteria most used by bettors, it doesn't fall into the same category of chasing rainbows with little more than high hopes and unrealistic expectations.

[thaskalos]

The quotes I referred to are above. Not wild-eyed marketing puffery from a sleazy system seller, but rather comments by a person widely referred to as "the dean of thoroughbred handicappers."

Take special note of the bolded quote. The "claim" that--with an adequate methodology--a bettor could very quickly learn to make a profit from wagering was not mine (originally). It was Ainslie's.

The quotes you used might have been Ainslie's...but the decision to apply them to the game as it is played today was clearly yours.

I think all would agree that the betting public as a whole is WAY more sophisticated today than it was during Ainslie's time...and comments such as these would be viewed with great skepticism today...especially if they came unaccompanied by any proof.

[OCF]

The quotes you used might have been Ainslie's...but the decision to apply them to the game as it is played today was clearly yours.

I think all would agree that the betting public as a whole is WAY more sophisticated today than it was during Ainslie's time...and comments such as these would be viewed with great skepticism today...especially if they came unaccompanied by any proof.

I can't disagree with you on any of that, but:

1. Isn't it possible that in many cases our greater sophistication just takes us further down rabbit holes?

2. Shouldn't we all view everything posted on this forum with great skepticism, both the claims and (admittedly all too infrequent) proof? ;) I suppose I'm arguing more for discernment rather than skepticism, but I think there's a lot of overlap there.

[thaskalos]

I can't disagree with you on any of that, but:

1. Isn't it possible that in many cases our greater sophistication just takes us further down rabbit holes?

2. Shouldn't we all view everything posted on this forum with great skepticism, both the claims and (admittedly all too infrequent) proof? ;) I suppose I'm arguing more for discernment rather than skepticism, but I think there's a lot of overlap there.

It is extremely easy for anyone here to stand up and say that..."I can transform a novice player into a winner in a matter of only a few weeks...simply by introducing him to a few simple techniques, which would allow him not only to win money overall...but to do so without having to suffer the indignity of enduring losing streaks of any consequence in the process".

The difficult part is to actually prove it.

[Tom]

We need a novice.
Anyone know a novice?

[OCF]

It is extremely easy for anyone here to stand up and say that..."I can transform a novice player into a winner in a matter of only a few weeks...simply by introducing him to a few simple techniques, which would allow him not only to win money overall...but to do so without having to suffer the indignity of enduring losing streaks of any consequence in the process".

The difficult part is to actually prove it.

Understood and agreed.

Again, isn't it possible that in many cases our greater sophistication just takes us further down rabbit holes? I'd even change the "possible" to "likely".

[thaskalos]

Understood and agreed.

Again, isn't it possible that in many cases our greater sophistication just takes us further down rabbit holes? I'd even change the "possible" to "likely".

If you are implying that "unsophistication" might be considered an asset in this game...then I would have to disagree.

[raybo]

It is extremely easy for anyone here to stand up and say that..."I can transform a novice player into a winner in a matter of only a few weeks...simply by introducing him to a few simple techniques, which would allow him not only to win money overall...but to do so without having to suffer the indignity of enduring losing streaks of any consequence in the process".

The difficult part is to actually prove it.

Yeah, such a claim is easy to make without having to back it up. At least some of us who make similar claims, actually provide examples of our work. That can't be said of the poster who made that statement. He has never attempted to provide anything, of substance, to back up his supposed expertise in racing. Just another troll trying to be somebody he's obviously not.

Almost anybody can talk a good game! Put up or shut up.

[OCF]

If you are implying that "unsophistication" might be considered an asset in this game...then I would have to disagree.

No, I'm not implying that. I am implying that what is regarded as sophistication is often times not of much use. Especially in the pari-mutuel environment where many of the $ are operating from within a fairly small range of increased sophistication.

[raybo]

We need a novice.
Anyone know a novice?

Yeah, I have introduced several to the game, and most of them are no longer novices anymore. And some of those are already profitable. It can be done, but it takes some commitment, and more than a little "intestinal fortitude" (which is in short supply in this game).

[DeltaLover]

Yeah, such a claim is easy to make without having to back it up. At least some of us who make similar claims, actually provide examples of our work. That can't be said of the poster who made that statement. He has never attempted to provide anything, of substance, to back up his supposed expertise in racing. Just another troll trying to be somebody he's obviously not.

Almost anybody can talk a good game! Put up or shut up.


What traynor claims is feasible in theory. Besides that though, as most of us here, I am reluctant to accept it, mainly because of the decades I have put on this game while still remaining just a student of it.

I am sure that most of us maintain a similar view.

In other words, Traynor is contradicting something that looks obvious for most of us, based on our experience; this does not necessary means that he is wrong.

What is missing from is provocative statement, is the proof! Presenting the 'curriculum' he is going to teach to the novice, is the only way to support his statement.

Until then, although I cannot completely rule out his theory, I will remain skeptic about it but certainly not claim that he never provided anything as a substance neither that he is pretending to be an expert in racing.

In my opinion he has a deep understanding of the game and his positions should at least be taken seriously and not easily rejected.

[thaskalos]

No, I'm not implying that. I am implying that what is regarded as sophistication is often times not of much use. Especially in the pari-mutuel environment where many of the $ are operating from within a fairly small range of increased sophistication.

IMO...the only way to win at this game is to outplay the other players by a significant margin. And the other players, as a whole, get progressively smarter...simply because the bad players eventually drop out, and only the better players remain. And, as the game continues to be played by mostly the better players...it will be discovered that even some of the better player are now experiencing heavy losses...and they too will eventually drop out. So now...the game will be played by only the even better players. And so it goes...

We eventually get to a point where most of the wagering dollars are bet by a small minority of the players...while most of the betting public is making insignificant wagers...which, even collectively, cannot be considered to be a major force on the tote board. That's where I think we are today...

If a player hopes to profit in today's game...then he has to outplay even that small nucleus of heavy bettors who affect the tote board the most. And that's what I call "sophistication"...

[aaron]

IMO...the only way to win at this game is to outplay the other players by a significant margin. And the other players, as a whole, get progressively smarter...simply because the bad players eventually drop out, and only the better players remain. And, as the game continues to be played by mostly the better players...it will be discovered that even some of the better player are now experiencing heavy losses...and they too will eventually drop out. So now...the game will be played by only the even better players. And so it goes...

We eventually get to a point where most of the wagering dollars are bet by a small minority of the players...while most of the betting public is making insignificant wagers...which, even collectively, cannot be considered to be a major force on the tote board. That's where I think we are today...

If a player hopes to profit in today's game...then he has to outplay even that small nucleus of heavy bettors who affect the tote board the most. And that's what I call "sophistication"...
Perfect example of what the game has become is the 1st race at Belmont today. Jacobson off the claim wins at 6/5.In the race Ness off the claim at 8-1.Rudy has a horse off a win at 5/2. There is also a trip horse,number 6 at 3-1. If you go for value in this situation,which all the books say you should you will lose.If you bet the Jacobson horse in this situation you will probably end up showing a profit. This is how the game has changed. In years gone by you would definitely try to beat the Jacobson horse,now I am not so sure this is the way to go.

[TrifectaMike]

The premise is a simple one, but remarkably sophisicated at the same time. Specifically, there are attributes that indicate a potential winner. Those attributes can be combined to create a synergy. The underlying logic is sound. What was lacking was the data analysis necessary to determine which attributes are significant, and the most effective weighting of those attributes in relationships with each other. Originally, the developer believed that simply adding the values together would be sufficient. It was--for a time. There is more complexity here than you can imagine. The significance of the attributes to any model should NOT be left to the regression method.

Using the basic premises (rather than uncritically copying the attributes and weighting of others) required considerable research (to determine the most effective attributes by track, distance, class, etc.) and considerable regression analysis to determine the most effective weighting to assign to each attribute and to each combination of attributes.

The attributes change. So do the weightings assigned to the attributes and combinations of attributes. It takes some pretty serious programming and data analysis to stay on top of it. It is not a question of the weightings changing. They change, because they were assumed to be deterministic. The weights have a distribution. Ordinary regression assumes they are fixed. For ex: Take a simple regression... Y=Ax+b. We estimate A and B and predict Y as x is observed. Assume your bet is based on Y. How confident are you in Y? Bet a buck or bet $500? It is NOT as simple as presented here.

In this frameworl, things get very complex very quickly

However, a simplified version can be learned fairly easily by a novice--no PhDs or quants required--and successfully applied. It was that simplified version that I had in mind when I suggested a novice could learn to be profitable in a relatively short period of time. This I would like to see!

Igor Kusuyshyn developed a similar approach for harness races, and some still use (continually adjusted, updated, and modified versions of) Kusyshyn's basic approach successfully.

Simple? I don't think so.

Mike

[DeltaLover]

IMO...the only way to win at this game is to outplay the other players by a significant margin. And the other players, as a whole, get progressively smarter...simply because the bad players eventually drop out, and only the better players remain. And, as the game continues to be played by mostly the better players...it will be discovered that even some of the better player are now experiencing heavy losses...and they too will eventually drop out. So now...the game will be played by only the even better players. And so it goes...

We eventually get to a point where most of the wagering dollars are bet by a small minority of the players...while most of the betting public is making insignificant wagers...which, even collectively, cannot be considered to be a major force on the tote board. That's where I think we are today...

If a player hopes to profit in today's game...then he has to outplay even that small nucleus of heavy bettors who affect the tote board the most. And that's what I call "sophistication"...


Excellent post Thask.

What I find to be the most interesting of it, is the closing statement.

To have any hope in today's game, you definitely need to outplay the heavy hitters, this is where you can make the score.

I think this is possible.

A $30M per year whale seems to have a different objective and approach to the game than a serious $300K bettor who naturally relies less on rebates and more on a large score.

In contrary to the public belief, I am reluctant to believe that all of those 'whales' are able to show profitability by the end of the year.

I have absolutely no personal experience with 'whales', but approaching the topic analytically, I cannot see how they can be so profitable, especially given the huge operational cost implied by the size of their teams and the sophistication of their platforms. I compare them to most of the hedge funds, who although might show profitability for a certain period they ultimately suffer huge loses that eventually take them out of business.

[castaway01]

Excellent post Thask.

What I find to be the most interesting of it, is the closing statement.

To have any hope in today's game, you definitely need to outplay the heavy hitters, this is where you can make the score.

I think this is possible.

A $30M per year whale seems to have a different objective and approach to the game than a serious $300K bettor who naturally relies less on rebates and more on a large score.

In contrary to the public belief, I am reluctant to believe that all of those 'whales' are able to show profitability by the end of the year.

I have absolutely no personal experience with 'whales', but approaching the topic analytically, I cannot see how they can be so profitable, especially given the huge operational cost implied by the size of their teams and the sophistication of their platforms. I compare them to most of the hedge funds, who although might show profitability for a certain period they ultimately suffer huge loses that eventually take them out of business.



If you're receiving a 10 percent rebate, you don't have to be profitable. You just have to get back at least 90 percent of what you wager.

[OCF]

I think all would agree that the betting public as a whole is WAY more sophisticated today than it was during Ainslie's time...and comments such as these would be viewed with great skepticism today...especially if they came unaccompanied by any proof.

If a player hopes to profit in today's game...then he has to outplay even that small nucleus of heavy bettors who affect the tote board the most. And that's what I call "sophistication"...

With all due respect, I could be wrong but I don't think those uses of sophisticated and sophistication can be reconciled.

I do agree with your last post as a whole though.

[thaskalos]

With all due respect, I could be wrong but I don't think those uses of sophisticated and sophistication can be reconciled.

I do agree with your last post as a whole though.
IMO...sophistication comes in levels...as does intelligence.

We can say that one person is more intelligent than another person...without necessarily implying that this person has a high level of intelligence. Sophistication works the same way.

When playing a difficult game such as ours...a certain level of sophistication is required just to survive without sustaining heavy losses. To actually win money on a regular basis would require taking this sophistication to an entirely different level.

[traynor]

The quotes you used might have been Ainslie's...but the decision to apply them to the game as it is played today was clearly yours.

I think all would agree that the betting public as a whole is WAY more sophisticated today than it was during Ainslie's time...and comments such as these would be viewed with great skepticism today...especially if they came unaccompanied by any proof.

I think that what I said is accurate, and applies every bit as much today as it did when Ainslie wrote it. If anything, the current obsession with speed and pace figures, as well as relative similarity of computer applications "analyzing the available data" makes it even easier. Specifically, the reference is to using a relatively simple numerical evaluation of key performance indicators (for lack of a simpler explanation). It is no more difficult to teach someone how to apply such an approach now than it was when Ainslie wrote it.

If you are referring to Ainslie's specific factor values applied to races today, I agree completely. However, as I have repeatedly stated, the attributes and values are derived from regression analysis of current reality--not copied from some musty old book found in a musty old bookstore in a back alley somewhere. Ainslie's approach is what I think has great value--not necessarily the factors and values he advocated.

I think skepticism is a good thing, having more than a bit of it myself. I also believe that the only really acceptable proof is that which one creates from her or his own results. I think any astute bettor--given the source (Ainslie) and a description of the methodology (also Ainslie's) could create a similar method specific to his or her track relatively easily. That is the best proof of all, and one I strongly recommend.

[traynor]

It is extremely easy for anyone here to stand up and say that..."I can transform a novice player into a winner in a matter of only a few weeks...simply by introducing him to a few simple techniques, which would allow him not only to win money overall...but to do so without having to suffer the indignity of enduring losing streaks of any consequence in the process".

The difficult part is to actually prove it.

Again, anyone interested in wagering can develop his or her own approach fairly simply, using Ainslie's process as a beginning template. Expecting someone else to "prove" such a thing is unrealistic, given the overall skepticism about anything outside one's personal comfort zone of knowledge and experience. If proof is desired, one can readily create it for himself or herself. If rejection of any proof is desired, one can readily tweak one's own efforts to assure negative outcomes.

In particular, I think you (and many others) may have been stung so many times with the purchase of "systems and methods" in the past that you are personally biased against anything that seems even remotely like those previous "bad experiences." Bear in mind I am not selling anything--only suggesting possible areas of value that might warrant further study.

[traynor]

What traynor claims is feasible in theory. Besides that though, as most of us here, I am reluctant to accept it, mainly because of the decades I have put on this game while still remaining just a student of it.

I am sure that most of us maintain a similar view.

In other words, Traynor is contradicting something that looks obvious for most of us, based on our experience; this does not necessary means that he is wrong.

What is missing from is provocative statement, is the proof! Presenting the 'curriculum' he is going to teach to the novice, is the only way to support his statement.

Until then, although I cannot completely rule out his theory, I will remain skeptic about it but certainly not claim that he never provided anything as a substance neither that he is pretending to be an expert in racing.

In my opinion he has a deep understanding of the game and his positions should at least be taken seriously and not easily rejected.

Again, I think the only acceptable proof would be that which one creates for himself or herself by applying the techniques in the real world at his or her own track and in her or his own wagering or research. Nothing else is acceptable, as far as I am concerned.

That is not a spurious suggestion. I think that anyone who actually wants to win (rather than simply endlessly argue in support of his or her preconceived notions and beliefs) would be more than willing to exert the effort necessary to prove or disprove the value for himself or herself. Conversely, I think anyone who wants to kick back and do nothing while demanding "proof" is unlikely to succeed at much of anything worthwhile.

[traynor]

Simple? I don't think so.

Mike

Neither do I. I was not referring to a full-blown, professional-level, industrial strength wagering application, nor to the techniques used in developing such. Nor was I using precise language in the description. Given the parameters of the current discourse community, I chose to restrict the description to terms and phrases likely to be within the knowledge range of most readers--what is usually referred to as "inclusive language choices."

[traynor]

Again, I want to emphasize that I am not selling anything, and feel no compulsion whatsoever to "prove" anything, nor to hand out a fully developed schema that can be immediately applied by anyone too lazy to do the work himself or herself. I might point out areas that I consider useful, but I have very little inclination to carry anyone to that area and spoon-feed them new knowledge.

I don't think it is especially difficult to analyze available data, find attributes, evaluate them, and combine them in a synergy that is productive, and apply them. I think that almost any motivated bettor of average intelligence who is willing to exert a bit of effort could do the same thing--if he or she restricts her or his efforts to a particular track or circuit, rather than trying to cherry pick "suitable" races from 20-30 different tracks.

It ain't rocket science, folks. No PhDs or quants necessary. Just a bit of effort, a bit of thinking, and a few hours away from lala land.

[thaskalos]

Again, anyone interested in wagering can develop his or her own approach fairly simply, using Ainslie's process as a beginning template. Expecting someone else to "prove" such a thing is unrealistic, given the overall skepticism about anything outside one's personal comfort zone of knowledge and experience. If proof is desired, one can readily create it for himself or herself. If rejection of any proof is desired, one can readily tweak one's own efforts to assure negative outcomes.

In particular, I think you (and many others) may have been stung so many times with the purchase of "systems and methods" in the past that you are personally biased against anything that seems even remotely like those previous "bad experiences." Bear in mind I am not selling anything--only suggesting possible areas of value that might warrant further study.

Traynor...no one here expected anything from you...until you yourself suggested that you would present a methodology here, through which a complete novice would be able to join the ranks of the winners in a matter of a few weeks.

When people asked you if this methodology was forthcoming anytime soon...you replied that your original plans for the unveiling of this mystery method were canceled...because your idea was not "well received" by the more skeptical among us. Our feathers were supposedly ruffled by your suggestion. And now that we skeptics are explaining to you that you have misunderstood us -- and that we'd love to see you go on with your original plans -- you are telling us that we can do it all ourselves, if we are so inclined.

I don't care if you never do what you say you originally planned do here: I just wish you wouldn't use our skepticism as an excuse.

[raybo]

Traynor...no one here expected anything from you...until you yourself suggested that you would present a methodology here, through which a complete novice would be able to join the ranks of the winners in a matter of a few weeks.

When people asked you if this methodology was forthcoming anytime soon...you replied that your original plans for the unveiling of this mystery method were canceled...because your idea was not "well received" by the more skeptical among us. Our feathers were supposedly ruffled by your suggestion. And now that we skeptics are explaining to you that you have misunderstood us -- and that we'd love to see you go on with your original plans -- you are telling us that we can do it all ourselves, if we are so inclined.

I don't care if you never do what you say you originally planned do here: I just wish you wouldn't use our skepticism as an excuse.

Theory and substance are completely different things, and while some here think the theory is great, and are more than happy to listen to it, there ae others of us who think substance is the only thing that has value, regardless of what theory is involved. You and I would probably fall in the latter group, while he falls into the former, all theory with no substance.

We all knew he would simply dodge the subject, as he always does. I think he really missed his calling, he should have become a politician.

[traynor]

Traynor...no one here expected anything from you...until you yourself suggested that you would present a methodology here, through which a complete novice would be able to join the ranks of the winners in a matter of a few weeks.

When people asked you if this methodology was forthcoming anytime soon...you replied that your original plans for the unveiling of this mystery method were canceled...because your idea was not "well received" by the more skeptical among us. Our feathers were supposedly ruffled by your suggestion. And now that we skeptics are explaining to you that you have misunderstood us -- and that we'd love to see you go on with your original plans -- you are telling us that we can do it all ourselves, if we are so inclined.

I don't care if you never do what you say you originally planned do here: I just wish you wouldn't use our skepticism as an excuse.

I expected a joint effort--a collaborative effort, if you will, in which more experienced handicappers would work together to create a way for novices to get off to a running start, on the assumption that more winners would equate to more bettors that would equate to larger mutuel pools that would equate to benefit to all. It was the fact that new bettors are completely turned off by the notion of years of effort being required to cash a few winning tickets and make a couple of bucks, and prefer other wagering activities, that provided the initial motivation.

I don't want (and didn't want) worshippers, chelas, acolytes, or mentees. Or even pats on the head (or anywhere else). I just wanted to put something together to give the newbies a shot at winning (which is substantially more motivating than losing for all but the most jaded).

I think self-claimed (and self-earned) knowledge is the most valuable. I think knowledge handed over on a silver platter is worthless. I think any experienced handicapper can read Ainslie's book, discover how he put it together, and do it themselves. If they are not motivated enough to do that, so be it. I have no interest in being the prima ballerina.

[TrifectaMike]

Neither do I. I was not referring to a full-blown, professional-level, industrial strength wagering application, nor to the techniques used in developing such. Nor was I using precise language in the description. Given the parameters of the current discourse community, I chose to restrict the description to terms and phrases likely to be within the knowledge range of most readers--what is usually referred to as "inclusive language choices."

Ok, how about you use some "exclusive language choices" and speak to me. I will not hold it against you.

Traynor, your posts have been all over the place... from Delphi to regression is "evil"...large samples, small samples, to Ainslie is the answer. Who is next to offer the "Key to the Mint", George Illich. Ther are both gone from this earth, but George's wife had all his notes, but she is also gone.

If a Mike Warren didn't exist, you could easily step into that role.

Folks, Traynor is NOT talking theory. He is simply talking "bullshit"!

Mike

[traynor]

Ok, how about you use some "exclusive language choices" and speak to me. I will not hold it against you.

Traynor, your posts have been all over the place... from Delphi to regression is "evil"...large samples, small samples, to Ainslie is the answer. Who is next to offer the "Key to the Mint", George Illich. Ther are both gone from this earth, but George's wife had all his notes, but she is also gone.

If a Mike Warren didn't exist, you could easily step into that role.

Folks, Traynor is NOT talking theory. He is simply talking "bullshit"!

Mike

I don't think you could afford my rates. However, don't let that stop you from dazzling the rubes with obfuscatory prose and impressive displays of esoteric concepts claimed to define reality for everyone else.

And once again, no PhDs or quants are required to win. It is only a horse race.

[raybo]

Ok, how about you use some "exclusive language choices" and speak to me. I will not hold it against you.

Traynor, your posts have been all over the place... from Delphi to regression is "evil"...large samples, small samples, to Ainslie is the answer. Who is next to offer the "Key to the Mint", George Illich. Ther are both gone from this earth, but George's wife had all his notes, but she is also gone.

If a Mike Warren didn't exist, you could easily step into that role.

Folks, Traynor is NOT talking theory. He is simply talking "bullshit"!

Mike

I think that's what we have been saying, just not in those words.

[traynor]

Ok, how about you use some "exclusive language choices" and speak to me. I will not hold it against you.

Traynor, your posts have been all over the place... from Delphi to regression is "evil"...large samples, small samples, to Ainslie is the answer. Who is next to offer the "Key to the Mint", George Illich. Ther are both gone from this earth, but George's wife had all his notes, but she is also gone.

If a Mike Warren didn't exist, you could easily step into that role.

Folks, Traynor is NOT talking theory. He is simply talking "bullshit"!

Mike

I don't suppose anyone has noticed that the PhDs and quants seem to come up short in applying their knowledge in their own wagering. They all seem to need someone to pay them for their "knowledge." And when the initial blush wears off, and the "employer" discovers that the neat sounding theories don't work as expected, it is off trolling for a new "client" for one's "consultation services."

Of course it is an incredibly complex problem that can only be solved with methods and techniques way beyond your ability to comprehend. And--of course--that is why one has to hire a purveyor of those complex methods and techniques. In your dreams, cowboy. Been there. Done that. PhDs impress me not at all.

[traynor]

Oh, dear, I forgot the lala land litany:

"Never mind all the BS and theory. Post your selections so we can see how you do picking winners in the real world."

[TrifectaMike]

I don't suppose anyone has noticed that the PhDs and quants seem to come up short in applying their knowledge in their own wagering. They all seem to need someone to pay them for their "knowledge." And when the initial blush wears off, and the "employer" discovers that the neat sounding theories don't work as expected, it is off trolling for a new "client" for one's "consultation services."

Of course it is an incredibly complex problem that can only be solved with methods and techniques way beyond your ability to comprehend. And--of course--that is why one has to hire a purveyor of those complex methods and techniques. In your dreams, cowboy. Been there. Done that. PhDs impress me not at all.

Traynor, I'm sorry to inform you that I don't troll for clients. I am the client...if anyone here is a cowboy, it is you.

Mike

[traynor]

Traynor, I'm sorry to inform you that I don't troll for clients. I am the client...if anyone here is a cowboy, it is you.

Mike

Yuh. I believe you. How about posting those selections? It would be nice to see how a real expert struts his stuff.

[PaceAdvantage]

Yuh. I believe you. How about posting those selections? It would be nice to see how a real expert struts his stuff.When was the last time you posted a selection? Haven't people called you out to post some picks? I believe your condescending reply was that you were too important/busy to do such a thing...I'm paraphrasing of course...

[dkithore]

I am not referring to claims of any kind (other than your assertion of profitability using figures and pace scenarios). I am referring specifically to a way to encourage new bettors to participate in thoroughbred racing by bypassing the long years of painstaking "learning," prolonged losing streaks, and substantial investment in software and data downloads that many seem to believe is essential to just break even, much less earn a profit.

The method I referred to is a simplified version of a (much) more complex approach used successfully by several wagering syndicates (that would most likely qualify as "whales").

Traynor!!

I am a newbie! Where do I start?

DK

[magwell]

Oh, dear, I forgot the lala land litany:

"Never mind all the BS and theory. Post your selections so we can see how you do picking winners in the real world." Really all you have to do is handicap a day or two of your picks and put all the nonsense to rest, It's called "put up or shut up"......;)

[traynor]

When was the last time you posted a selection? Haven't people called you out to post some picks? I believe your condescending reply was that you were too important/busy to do such a thing...I'm paraphrasing of course...

Not too long ago. On the harness thread.

[PaceAdvantage]

Then I guess I stand corrected...my apologies...

[cj]

Not too long ago. On the harness thread.

I didn't realize this thread was about drivers...sorry, couldn't help myself!

[traynor]

Traynor!!

I am a newbie! Where do I start?

DK

Read Tom Ainslie. Theory and Practice of Handicapping and The Compleat Horseplayer. He presented a very simple approach--isolate key factors that are associated with winners more strongly than with non-winners. The factors he used are somewhat outdated, but the principles are sound. If you primarily bet one track, study the results charts for races compared to the past performances. What you are looking for is a pattern in the past performances (or in specific factors) that are associated with winners.

In essence, look at the past performances with the race results in hand, and ask yourself how you could have predicted that outcome. It is a lot easier to stick to one specific type of race intially, to avoid confusion. The predictive factors for a dirt sprint will usually be different than the predictive factors for a turf route. Because they are so frequently carded, mid-level claiming sprints are a good place to start.

When you see something that looks like it could have predicted the upcoming win, circle it with a red flair pen. It shouldn't take too long to discover you are circling similar areas in different races--positions, beaten lengths, win percentages, jockey stats, trainer stats, whatever (I have no idea what kind of data you are using for PPs).

[traynor]

I didn't realize this thread was about drivers...sorry, couldn't help myself!

Everyone should be allowed at least one snarky remark per day. :)

[thaskalos]

Read Tom Ainslie. Theory and Practice of Handicapping and The Compleat Horseplayer. He presented a very simple approach--isolate key factors that are associated with winners more strongly than with non-winners. The factors he used are somewhat outdated, but the principles are sound. If you primarily bet one track, study the results charts for races compared to the past performances. What you are looking for is a pattern in the past performances (or in specific factors) that are associated with winners.

In essence, look at the past performances with the race results in hand, and ask yourself how you could have predicted that outcome. It is a lot easier to stick to one specific type of race intially, to avoid confusion. The predictive factors for a dirt sprint will usually be different than the predictive factors for a turf route. Because they are so frequently carded, mid-level claiming sprints are a good place to start.

When you see something that looks like it could have predicted the upcoming win, circle it with a red flair pen. It shouldn't take too long to discover you are circling similar areas in different races--positions, beaten lengths, win percentages, whatever (I have no idea what kind of data you are using for PPs).

Somehow...I was under the impression that you were against "backfitting"...