Understanding workouts (better)

[bellsbendboy]

In the last ten years or so the availability of workout data has increased exponentially. Led by the DRF with rankings, bullets and staff reports and supported by other publications, often on a fee basis, workout data is ubiquitous. Add in; bloggers, private clockers and the tracks own websites and you have a pile of info to evaluate.

Workouts are a large part of my handicapping and over the last five decades I have come to realize; I have much to learn. With that printed and my reading of another thread that somewhat skirted the issue, I thought I would
add some insights. In no particular order;

1) Twenty five percent of thoroughbreds are runoffs and although they work fast in the morning; they will never win a race. "Morning glory's" indeed.

2) Ten percent of workouts are in the dark, ten percent are missed and smaller percentages have the wrong horse, wrong time or inaccurate distance.

3) Most tracks change radically during the morning as the sun rises. Certainly a horse working five minutes before the renovation break is catching a totally different surface than another horse that works five minutes after the break. Many tracks are fast in the morning but dull in the afternoon, and others can be the opposite.

4) Trainers choose among the following;
A) Small rider-larger rider or jockey

B) In company or alone

C) Drugs? Many horses work on bute, Lasix etc.

D) On the flat, or from the gate?

E) Which pole the worker breaks off from, or often between poles. Many trainers work their horses into the turn which greatly affects the time. Some trainers work around both turns.

F) If in company; with a stablemate, or another trainers' horse. With an unstarted maiden or with a stakeswinner? A 2yo or older.

G) With a change of equipment? Blinkers of all different types, tongue ties, caulks, bandages, stickers, draw reins etc.

H) In some fifty percent of works the trainer does not take his/her watch when they go to view the work, as the time is irrelevant.

As you can see from this abbreviated list the variables are endless. Using raw times to evaluate, or bet horses is primitive at best, although you will cash, infrequently. As a general rule, I put little stock into the time of the work, and much more into the schedule a horse is on and the spacing of the works. Most top trainers select a target race and work backwards.

Much can be learned by putting a horse into your stable mail and "plan" the works yourself. The only horse I have in my stable is Street Sense and I put him in there when Carl was quoted that he put off his retirement because he had a couple of good ones. If I had to guess, I would expect 'Sense would work five eights Friday the 10th, five eights the 16th and blow out a half the 22nd. That would set him up nicely for the Travers.

In conclusion, this post may be redundant to many, and old hat to others, yet works are a very good indicator of how your horse will run. The time of the work(s), however should be evaluated on a case by case basis. My favorite workout anecdote concerned the great Seattle Slew. Before his debut, Billy T. worked him and "slew" saw another worker seven or eight lengths ahead of him. Slew ran that horse down and Billy looked at his watch seeing "Slew went a half in :45 flat! Seattle Slew NEVER worked in company again!! A side note that you might find interesting is at www.news.bloodhorse.com/viewstory.asp?id=40068 Crack staffer Claire Novak reports on her visit to the clocker stand at Saratoga this past Sunday. If the link fails bloodhorse.com, then a day at Saratoga Sunday aug 5th will get you there. BBB

[jotb]

You came up with all that and left out probably the most important thing and that is how many horses that work in the morning are given a different name to the clocker before the work.

Joe

[46zilzal]

Only thing to look for is frequency. Forget the rest for if you are not there to see them, you know NOTHING about what happened, who was riding, where they rode, was it a even move or sped up at the end, in company or alone?, was it near the start or after the break when the track was re-surfaced, and, remember there are no drug tests for workouts.

[DJofSD]

Did I miss it -- nothing in the list about who was up during the drill. Exercise rider or jock?

If a horse goes out onto the track but just jogs around the circuit a couple of times, is that a work out?

As more tracks convert to an all weather surface the amount of time needed to renovate is decreased and more time for more horses to get out there.

I agree w/ zz -- the number of times with an official work, how far they work and how soon after a race tell me more than anything else.

Be careful of info overload (or paralysis by analysis).

[Tom]

HTR has a WO rating - calculated strictly from the data files for that card - that is awsome, to say the least. It has withstood testing of large samples, time periods, and not only improves performance of other factors, it sometimes shows a profit on its own for certain tracks and times. It looks at patterns in the works.

[46zilzal]

HTR has a WO rating - calculated strictly from the data files for that card - that is awsome, to say the least. It has withstood testing of large samples, time periods, and not only improves performance of other factors, it sometimes shows a profit on its own for certain tracks and times. It looks at patterns in the works.

I wouldn't hold my breath waiting on that. The more I learn about workouts from actually BEING AT THE TRACK, the less the TIME means and the frequency depends on the horse and the trainer's specific plan for said horse. They have individualized programs.

[Tom]

Do we have to give back the money???

It has been proven over two years and hundreds of thousands of horses to work.
Far more data to support this rating than your E/L numbers. FAR more.

[Fastracehorse]

Did I miss it -- nothing in the list about who was up during the drill. Exercise rider or jock?

If a horse goes out onto the track but just jogs around the circuit a couple of times, is that a work out?

As more tracks convert to an all weather surface the amount of time needed to renovate is decreased and more time for more horses to get out there.

I agree w/ zz -- the number of times with an official work, how far they work and how soon after a race tell me more than anything else.

Be careful of info overload (or paralysis by analysis).

It's not that easy.

Trainer Doug O'Neill says he likes to save the horse for the race. He frequently will race a horse without a work between races - and they have won.

Work outs for me are important but I haven't seen a pattern that is the 'Holy Grail' of works.

fffastt

[Fastracehorse]

You are right - times aren't everything.
Some of the best trainers ( Dutrow comes to mind ) work them slow.

What about Cristophe Clemente?? He can train a grass horse to win long off a series of slow 4 f works.

I agree that works are both fascinating and confusing.

On rare occasions, the works stands out as being a sign of a huge effort coming. But normally they are ambiguous or benign in terms of pushing the button.

One thing I do like are firsters with one fast work - or firsters who have worked long and quick but have a light work tab. What else do have to go on with firsters besides looks?? Pedigree??

fffastt

[Jeff P]

I developed a factor in JCapper called Workout Brilliance which is based primarily on raw workout time. Despite everything you read in print about workout time being irrelevant, and everything you read about workout times being inaccurate and misreported... the rating has proven itself to be statistically relevant when applied over significant data samples.

My calendar year 2006 database shows the following:

WOBRILL
By: Workout Brilliance Rank (2006 1st time starters removed)

Rank Gain Bet Roi Wins Plays Pct Impact
1 -5593.80 56182.00 0.9004 4814 28091 .1714 1.3548
2 -10511.50 59542.00 0.8235 4374 29771 .1469 1.1615
3 -15458.50 59846.00 0.7417 3935 29923 .1315 1.0396
4 -14442.20 57026.00 0.7467 3475 28513 .1219 0.9635
5 -13060.60 48320.00 0.7297 2812 24160 .1164 0.9201
6 -10400.50 36826.00 0.7176 1926 18413 .1046 0.8269
7 -7379.00 25138.00 0.7065 1160 12569 .0923 0.7296
8 -3994.10 16048.00 0.7511 689 8024 .0859 0.6788
9 -3323.80 9454.00 0.6484 369 4727 .0781 0.6171
10 -1659.00 4966.00 0.6659 196 2483 .0789 0.6240
11 -835.10 2376.00 0.6485 65 1188 .0547 0.4325
12 -536.20 924.00 0.4197 19 462 .0411 0.3251
13 -73.50 156.00 0.5288 3 78 .0385 0.3041
14 -66.00 66.00 0.0000 0 33 .0000 0.0000
15 -2.00 2.00 0.0000 0 1 .0000 0.0000
16 -2.00 2.00 0.0000 0 1 .0000 0.0000
17 -6.00 6.00 0.0000 0 3 .0000 0.0000
18 0.00 0.00 0.0000 0 0 .0000 0.0000
19 -2.00 2.00 0.0000 0 1 .0000 0.0000

This factor (WoBrill or fast workout time) is one of those that I refer to as a hidden positive. What I mean by this is that it is a factor the betting public historically tends to overlook as they bet the races. Believe it or not this factor alone allows you to improve your handicapping to the point of you being able to outperform most other handicappers at your track. And at the other end of the spectrum... look at the dismal results you'd get if you were to restrict your betting to horses that rank poorly vs their fields in this one factor. As a general rule, you can become a much better handicapper simply by avoiding horses with terrible WoBrill relative to their fields.



-jp

.

[46zilzal]

raw workout time? when did workouts have a variant?

[ddog]

Jeff,

The stats you have listed here, what does the database consist of as to tracks and type of races,etc?

thx

Also, on Toms post about the HTR workout rating.
It does make money, I think it may look at races/works/etc to come up with the nbr but in the correct timeframe/race it is very strong.

[Jeff P]

raw workout time? when did workouts have a variant?I'm not using a variant at all... simply raw workout time without any consideration whatsoever for the surface the workout was run on... Believe it or not that one piece of information alone... despite everything you've ever read or heard about workouts - can be used to improve overall results.


-jp

.

[46zilzal]

It is ONE thing to utilize some 2nd hand evaluation and a completely different view when you are there looking at what takes place first hand.

Frequency si, time no

[Jeff P]

Jeff,

The stats you have listed here, what does the database consist of as to tracks and type of races,etc?

thxThe database includes approx 200,000 starters... races at all distances on dirt turf and poly... at A,B, C, and some D tracks... NY (AQU BEL SAR) KY (CDX ELP TPX) FLA (GPX CRC TAM) IL (APX HAW) SOCAL (SAX, HOL, DMR, FPX) NOCAL (BMX, GGX, & the Fairs)... as well as a scattering of other tracks I decided to play such as OPX DEL MTH TUP EMD LAD FGX and a few others.


-jp

.

[Jeff P]

Frequency si, time noThat's pretty much what everybody tells me. In the end you're going to believe what you want to believe.

Actually, I've found that BOTH have statistical significance.


It is ONE thing to utilize some 2nd hand evaluation and a completely different view when you are there looking at what takes place first hand.Humor me though... if workout time truly does not matter... explain the statistical significance of the data I've posted.





-jp

.

[spilparc]

I don't pay any attention to workouts with one exception. If a horse has been off for say, six weeks, that's fine, but if there has been no work in the interim, then I severely downgrade the horse. I feel this is especially true with older, cheaper horses.

I liked two horses yesterday that had been laid off with no work. I tossed them both--neither won.

[Tom]

Humor me though... if workout time truly does not matter... explain the statistical significance of the data I've posted.
-jp

.

Don't hold your breath.:D

Thanks for sharing your findings. I don't know what HTR uses, but the WO rating is one powerful little rascal, like slat to the stew.

I agee too, about raw times - one wise handicapper at HTR turned me onto two specific lenght/time workouts that have drastically improved my play of layoff horses and FTS.

[Robert Fischer]

I just don't get this stuff. 90% roi for botox what the hell? So you are saying if a horse works the fastest you only lose 10% of your money for the year 2006 ? How can you bet like that ?
sure that is interesting to some extent...

You have to know the horse and the trainer.

then you look at the pattern and speed compared to what is expected from the horse and trainer.

sometimes a fast work is a negative.

you have pletcher's false favorite at arlington to be run overe the all weather track . this horse quits every time at 6 furlongs on the turf. He just can not rate! now his last work was a bullet. he is 9/5 in the ml , so you are structering your 10 cent superfecta around him finishing 3rd or worse. He just fired a bullet, this chump couldn't even rate in the morning , no way can he rate today from out the gate with goofy rene douglas on board!

or you have street sense and he works 5 in 102. Common sense tells you to assume the last three went in 37 or better... and you have people doubting if he is in shape.....

do you guys really bet on statistics from a large sample of a previous year and apply that to a specific race for today? :confused:


learn your horses

learn your trainers

learn your surfaces

learn how to handicap


then you might pay a small bit of attention to a work time or pattern taking into account the known as well as the unknown and undocumented

[kenwoodallpromos]

Look at works and races as a streaming movie, not a still photo.
Anytime you look at any factor as a static event you are in trouble.
A good horse should show solid works all the time; if not, either the horse or the trainer is inconsistent. The worst negative is multiple works slowing down lately.

[xtb]

do you guys really bet on statistics from a large sample of a previous year and apply that to a specific race for today? :confused:


learn your horses

learn your trainers

learn your surfaces

learn how to handicap



Well Jeff, looks like you're just going to have to return all the money you've made. Oh, and don't forget to take your name down from the leaderboard of the PA contest, you obviously don't know what you're doing. ;)

[Jeff P]

I just don't get this stuff. 90% roi for botox what the hell? So you are saying if a horse works the fastest you only lose 10% of your money for the year 2006?Yes.

How can you bet like that? I personally don't bet like that.

do you guys really bet on statistics from a large sample of a previous year and apply that to a specific race for today? What I do is this:

I do my own research and use what I discover to create computer models (sets of rules) that define spot plays. In JCapper a spot play is called a UDM or User Defined Model. A UDM can contain as many factor combinations as I want in the model.

IMHO, the .90 roi represented by WoBrill rank=1 makes a pretty nice starting point for creating a UDM (or spot play.) Compare it to the .75 roi you'd get if you started with the set of all horses everywhere. So what I will do is take a factor like WoBrill rank=1, use it as the foundation for a UDM, and go to work on improving things from there. I'll do research into what happens when WoBrill rank=1 horses are combined with other factors. When I discover something that adds to the effectiveness of the model then I'll add that factor (whatever it happens to be) to the model.

Once I am happy with the model I will test it by confronting it against a validation sample - fresh races that weren't part of the sample used in the model's development. If I'm happy with the model's performance against the validation sample, I'll consider using the model in my own live play.

Workout Brilliance is just one factor - a factor that with research and work - can be used to build models that perform well going forward.

I don't really handicap today's race in the traditional sense at all. In fact I gave up traditional handicapping completely almost 15 years ago. I purposely almost never look at traditional past performances any more. More accurately, I would say all of my "handicapping" is done during the R&D phase of my model(s). On race day, if a horse qualifies according to the defined rules of one of my models - and I perceive the horse to be underbet relative to my own probability assessment - then I'll pull the trigger.

That's about as accurate a description of my own process (in one post) that I can give you.


-jp

.

[Robert Fischer]

Jeff - Whatever you are successful with, and enjoy doing :ThmbUp:. I apologize for being a bit of an AH ;)

[nobeyerspls]

learn your horses

learn your trainers

learn your surfaces

learn how to handicap


then you might pay a small bit of attention to a work time or pattern taking into account the known as well as the unknown and undocumented

Finally, the mention of the word "surface". I am primarily a surface handicapper looking for favorable surface changes that produce outstanding improvements in performance. The nuances take time to master but the reward is worth the effort.
Workouts are involved here though as they often point to surface preference. For example, a horse's first five starts are poor efforts on the Woodbine poly. They ship to Fort Erie to run on the dirt track. You note a decent workout on the WO training track, a dirt surface, among several not so good poly works. Even without a class drop, this horse is a great play at long odds.
Surface can also work in an inverse way as you pointed out in your example. Identifying an odds-on favorite with no chance of winning can be as rewarding as locating a live longshot.

[nobeyerspls]

I don't pay any attention to workouts with one exception. If a horse has been off for say, six weeks, that's fine, but if there has been no work in the interim, then I severely downgrade the horse. I feel this is especially true with older, cheaper horses.

I liked two horses yesterday that had been laid off with no work. I tossed them both--neither won.

The no work thing can get tricky. I bred and raced a filly with conformational problems, the most pronounced being her right foreleg offset at the knee. The periostial elevation shortly after foaling helped a little so she was able to train. Eventually bone chips showed up so we removed them surgically and, after appropriate time off, returned her to training. If you looked at the form for her return you would see no works since her last race several months prior. She won easily and I then sold her, disclosing her problems to the buyer. We were able to pony her back into shape. For those unfamiliar with the term it simply means long gallops under tack with another horse alongside. This keeps the rider's weight off thus reducing pressure on the knee.
With older male horses, two minute licking can be as productive as workouts. You gallop for a few days and then you go a mile in as close to two minutes as possible. This sequence is repeated until the horse is fit enough to race. That pace is somewhere between a gallop and a workout and is very effective at legging up. If you want a nice price in the return race, simply throw in a couple of intentially slow works spaced far apart.

[spilparc]

The no work thing can get tricky.

Actually, the no workout after a layoff thing is a new "guideline" for me. It's something I've been following very closely. Normally, I don't pay any attention to workouts. But I did happen to notice that some horses I liked (and bet) hadn't been worked in a long time and 90% of them did very poorly. So far, one of them has won, and he was laid off for about 5 weeks. Again, for the most part, these are older, cheaper horses.

Nothing is written in stone, and I'm certainly flexible. If a plethora of layoff horses with no workouts start winning, then it's back to the drawing board.

[classhandicapper]

I've found 3 useful applications of WOs.

1. When a decent quality lightly raced horse that is eligible to improve suddenly shows a much improved WO line, it often signals a big move forward in form. It helps a real lot if you have back issues of the DRF so you can look up back workouts. You want to make sure the recent WOs since the last start are actually the best WOs the horse has shown and not just the best recent ones.

2. If a horse usually works regularly for a trainer that tends to work all his horses regularly and he shows up in a race with a longer gap than usual between starts and there is also a gap in the WOs, the horse almost certainly had a problem.

[Tom]

#3 is a secret?

[jotb]

I don't pay any attention to workouts with one exception. If a horse has been off for say, six weeks, that's fine, but if there has been no work in the interim, then I severely downgrade the horse. I feel this is especially true with older, cheaper horses.

I liked two horses yesterday that had been laid off with no work. I tossed them both--neither won.

Hello Spilparc:

If a horse has been away more than 60 days, I do believe the horse has to have a published work within the last 30 days. In regard to older horses, you really shouldn't breeze them constantly especially if they have been running. They are fit enough! Your situation yesterday worked for you but even though you saw no published works on these horses that does not conclude that the horse has not been going to the track in the mornings. Anyway, you can see if a horse is racing fit by just looking at the belly. If a horse has been sent back to the farm chances are he or she has gained weight and you will notice this in paddock inspection. It's rare that a trainer would run a horse without some type of morning work. The trainer will risk that horse to possible injury if he is asked to run in a race. Just like humans if you sit around doing nothing and then go out and run hard and fast you might pull something or for sure be sore for awhile.

Joe

[pandy]

Excellent report and it doesn't surprise me. I started putting more emphasis on workouts (especially fast works) about 7 years ago and it may be the single most important thing I did to produce a consistently profitable ROI. I will give you a few things I've learned about workouts--

1). If two horses seem evenly matched and one has a fast recent workout, give the edge to the fast worker, especially if the horse is trained by someone who doesn't beat their horses in the morning. For example, a fast work by a Bruce Headley horse is hard to gauge because he always works them fast, but not many trainers push them like he does

2). Beware of fast workouts by cheap speed horses, especially in maiden claiming sprints. Some of these "sprinty movers" are simply pullers, or speed horses that like to run and are hard to hold. They consistently work fast but in the actual race they show early speed and tire. This is even more pronounced on Polytrack. There have already been a few of these fast working cheap speed horses that got over bet this meet at Del Mar and popped and stopped. A fast workout by a horse with a one-paced route style or a firster with a route pedigree is much more powerful to look for. Horses that are bred for speed tend to work fast, but it doesn't mean that they are good horses. If a horse with a low dosage index goes :46 for 4f, chances are that horse is a runner because the horse is bred to route and probably does not have the twitchy sprinty action, but a good, long, powerful stride. On these artificial surfaces, stride lengths are increasingly signficant. Sprinty movers are going to end up shipping out to dirt tracks.

3). Slow breezes are overrated. For instance, professional clockers often rave about a horse that breezed 6f in 1:15 because the horse was under wraps and breezing. This workout can certainly help condition a horse but it means nothing in terms of showing whether or not the horse can run. A lot of horses look like a million bucks when going slow, it's when they have to go fast that their true class is tested.

[46zilzal]

I continually get a chuckle out of secondary associative data mining.....Horses run races and it is how they do there, not in the barn, not in the morning, not with the trainers stats, no where but in the competition of going up against other horses in races that is germane.

Databases average out the difference of each particular race to the extent that the specific is lost to the general. It is only the evaluation of the interactions of the historical capabilities of horses coming together in a said race that makes any difference. Today now in the match up of today's race...period.

Go to the back stretch and see the specific training outlines of EACH animal. Just like humans they all have their strengths and weaknesses which dictate how they train. To average out all that specificity is downright dumb. It would be akin to your telling me how a baseball player will do based on the way they worked out. Hogwash.

[Tom]

Well, with all due respect, you don't have clue what you are talking about.

Your failure to understand the process in no way affects it's profitablility or effectiveness.

Jeff has shown statistical significance, and so has Ken Massa. What have you offered besides opinion backed by no data?

[46zilzal]

Put enough data together the correct way, and you can prove eggs produce new hair growth.

Better horses feel better, are in better shape and probably work out better........so what?

It is how they race that is significant not how they exercise.

[GaryG]

Tom, don't confuse the man with facts. He already knows everything...:eek:

[46zilzal]

Tom, don't confuse the man with facts. He already knows everything...
Bottom line is that there is no one way to the mountain top. Many roads lead the same way. Workouts aren't on my path, particularly after I began to work on the backstretch.

Gave up databases in the 80's. One cannot draw the specific from the general...

[DJofSD]

One cannot draw the specific from the general...

IOW, correlation is not causation.

[46zilzal]

IOW, correlation is not causation.
If you bring the same 10 horses together, on the same race track, at the same distance repeatedly every 21 days over the season, you will never have the same ten competitors meeting each other. As Hericlitus said in 500 BC "you can never step in the same river twice." Things change.

The racing surface will change, the form cycle will change each animal, physical nature of the animals will change, the riders might change, the nutrition might change as some horses change barns etc. etc.

The competitors TODAY are what we evaluate, not some peripheral information.

Owners pay the bills. They need to know that it is going to have a good return. The joke on the backstretch is that works are always faster the day the owners are there: usually weekends when they come out to see first hand. A Wednesday morning work, with the 150 lb exercise rider up has no observers other than the stable. A Saturday AM work, with a 115 jockey up, needs to impress that owner who pays the bills so it strangely gets faster. Still just exercise one way or the other.

[Tom]

You assume you know how the ratings are calculated, You do not.
You assume you know how people are using the databases. You do not.
You assume you nknw how we use the information obtained from the db's. You do not.

Other than that, you make good points.:bang:

[46zilzal]

In a word EXTRANEOUS information is still extraneous to the real data at hand. I follow, quite successfully, but not exclusively, the Malcolm Gladwell (author of BLINK) mantra of "Don't think blink." Keep data to a minimum effectiveness and not further.

Not everyone's modus operandi, but it works for this handicapper.

[the_fat_man]

You assume you know how the ratings are calculated, You do not.
You assume you know how people are using the databases. You do not.
You assume you nknw how we use the information obtained from the db's. You do not.

Other than that, you make good points.:bang:

I'm kind of curious, Tom, why you assume that ASSUMPTIONS need to be made when it comes to databases and their use.

Let me ask you a question as your post seems to be from the perspective of one who uses and understands databases:

1) do you know what relational algebra is?
2) how about tuple calculus?

or

even basic (propositional) logic.

Ever sat through a graduate level database course that was essentially ALL logic?

Using someone else's database program not only doesn't make you a DB expert
it doesn't even qualify you to criticize someone else for offering a critique of database use in racing.

Gee, One can only wonder what mysterious things you DB guys are doing.

They must be terribly abstruse.:lol:


Let's go to the data, now.

[Robert Fischer]

I don't think it is impossible to handicapp from a database using a set of rules to form spot plays as Jeff was describing. It isn't the way I do it, and that is ok. I made the mistake of bashing it a little, but hey, there are many ways to approach racing.

[46zilzal]

but hey, there are many ways to approach racing.
Exactly, each with it's pros and cons. Just pointing out some cons here.

[classhandicapper]

#3 is a secret?

:lol:

#3. Workouts in company where one of the horses has clearly understood ability.

[citygoat]

I look for a horse that has worked out on today's track not training centers or some other track than today's.

I especially like it on festival days(breeders cup,arlington million .......)

[Tom]

Does this help you out ACE?

I'm kind of curious, Tom, why you assume that ASSUMPTIONS need to be made when it comes to databases and their use.
BECAUSE HE STATES THEM IN HIS POSTS.
Let me ask you a question as your post seems to be from the perspective of one who uses and understands databases:

1) do you know what relational algebra is? YES.
2) how about tuple calculus? NO.

or

even basic (propositional) logic. NO LOGIC IN THIS THREAD.

Ever sat through a graduate level database course that was essentially ALL logic? NO.

Using someone else's database program not only doesn't make you a DB expert
it doesn't even qualify you to criticize someone else for offering a critique of database use in racing. I USE MY OWN - THE OTHERS VALIDATED THE RATINGS and YES IT DOES, WHEN THAT PERSON CRITIQUES SOMETHING HE DOESN'T HAVE A CLUE ABOUT HOW OR WHAT WE ARE DOINIG, LIKE YOU.

Gee, One can only wonder what mysterious things you DB guys are doing.

They must be terribly abstruse.:lol: NO, REALY SIMPLE STUFF - YOU AND 46 ASSUME YOU KNOW WHEN IN FACT NEITHER OF YOU HAVE A CLUE. rif - NIETHER OF YOU ARE READING WHAT I POSTED, JUST QUICK TO CRITIQUE.


Let's go to the data, now.

[the_fat_man]

Do you actually understand what he's assuming? Or why, assuming he's accounting for a certain distinction, his criticism is warranted?

[46zilzal]

EXTRANEOUS is still irrelevant

Ace, Robert Preston, is long since dead.

[Tom]

Do you actually understand what he's assuming? Or why, assuming he's accounting for a certain distinction, his criticism is warranted?

I stated that HTR has a WO factor that not only improves performance of other factors, but sometimes shows a profit stand alone, for limited time periods. HE has since attacked that. Jeff posted a detailed workout of his WO rating that shows statistical significance, and HE attacks that.
Hhis criticism is not based on any knowledge of either factor, how theyare calculated, or how we use them. To me, that is pretty stupid.
You chime in asserting you are some kind of expert in using databases and I say - what do YOU know about the factors or how we use them?

I ask you both again - Do we have to give back the money?

[Tom]

EXTRANEOUS is still irrelevant

Ace, Robert Preston, is long since dead.

Explain to me what is extraneous. Show me DATA that statistically proves it is irrelevant. Until then, you are strutting ignorance - and smartly too, I might add!

[the_fat_man]

I stated that HTR has a WO factor that not only improves performance of other factors, but sometimes shows a profit stand alone, for limited time periods. HE has since attacked that. Jeff posted a detailed workout of his WO rating that shows statistical significance, and HE attacks that.
Hhis criticism is not based on any knowledge of either factor, how theyare calculated, or how we use them. To me, that is pretty stupid.
You chime in asserting you are some kind of expert in using databases and I say - what do YOU know about the factors or how we use them?

I ask you both again - Do we have to give back the money?

The problem is essentially the following:

46 is adamant that EACH case needs to be treated individually and not as a subset of a more general type. I'm assuming that he understands that the issue is one of probability rather than of absolutes.

The interesting part of this:

are there enough INDIVIDUAL X's that don't qualify as GENERAL X's to made this worthwhile. In other words, are there enough winners out there that don't satisfy the factors identified as necessary ones by a particular DB method OR do most of these winners that 46 thinks are outside the set actually satisfy the criteria.

It's hard to believe that each winner has a completely individualistic set of factors and everything else is EXTRANEOUS. Probably, there are one or two unique factors in each case and a whole slew of general ones. Which kind of makes the case for each camp.

[Tom]

I suspect 46 is a "Musterbatater," as he likes to call others.:D

The fact is that every test of the WO factor shows it be a strong positive. It improves other factors when combined with them.

I'll post more later....specifics.

[bigmack]

When guys like T or Jeff post voluminous back tests it's a way of demonstrating the significance of a factor over time. Do you, Fatty & 46, actually think that people are sitting looking at thousands of races to determine if the works of the entrants of a particular race are germane? Those that use programs look at many factors for EACH race and make their decisions accordingly.

With 46 it's understandable cause he's a stone cold nut, but with you Jumbo, haven't you graduated from questioning DB guys about their methodology?

[xtb]

Gave up databases in the 80's.

This is the 21st century, there is no comparison of what you used in the 80's as to what h/w and s/w of today is capable of.

[the_fat_man]

With 46 it's understandable cause he's a stone cold nut, but with you Jumbo, haven't you graduated from questioning DB guys about their methodology?

Listen, you're not following the argument.

DBs allow for the identification of any given number of factors that are present in the lines of MOST winners. Notice, the key here is 'MOST', cause if it were ALL, then you'd have 100% strike rate (assuming you identified the relevant factors).

So, it follows, that there are SOME winners that fail to satisfy these factors ---perhaps not all but some.

There's also the reasonable argument that perhaps it is not these GENERAL factors that are KEYS to a given horse winning BUT rather OTHER, individual factors that are outside the general ones. If this were the case, the GENERAL factors would be EXTRANEOUS (and thus irrelevant).

The world is Humean and we're all wired (and rationally correct) to expect the general case. It's all about the problem of induction, Mack.

The turkey has 3 options, NOT 2, however:

1) trust that the farmer will feed it everyday AND Thanksgiving --and lose its head Tday

2) NEVER trust the farmer and starve to death, BEFORE Tday ever comes around

3) be wary of the farmer, and eat only after he has left --thereby avoid ing starvation and beheading.

[46zilzal]

I suspect 46 is a "Musterbatater," as he likes to call others.:D

The fact is that every test of the WO factor shows it be a strong positive. It improves other factors when combined with them.

I'll post more later....specifics.
Disregarding it means no effect, no relevance, no evaluation. that's all. Others do what they want.

The musterbators are a singular group of rule crazy individuals and that title represents no one else.

[bigmack]

The turkey has 3 options, NOT 2, however:
1) trust that the farmer will feed it everyday AND Thanksgiving --and lose its head Tday
2) NEVER trust the farmer and starve to death, BEFORE Tday ever comes around
3) be wary of the farmer, and eat only after he has left --thereby avoid ing starvation and beheading.
We'd best stop before it yet gets to be another exercise in futility. Let's just say that I prefer to put together puzzles with more pieces and those of my ilk do well by defining which pieces are more integral than others on a race by race basis.

As for turkeys, Wild Toms would kick the daylights out of those punk-ass farm raised slobs. They get what they deserve.

[46zilzal]

This is the 21st century, there is no comparison of what you used in the 80's as to what h/w and s/w of today is capable of.
Relational databases are faster, but not that much more sophisticated than the old dBase IV.

[46zilzal]

It's hard to believe that each winner has a completely individualistic set of factors and everything else is EXTRANEOUS. Probably, there are one or two unique factors in each case and a whole slew of general ones. Which kind of makes the case for each camp.
I never look at the individuals, I look at the race interaction. The race is what to evaluate. It is only in a competitive situation that one can evaluate the merits of any horse. Change the interaction and the same animal is viewed in a completely different light.

Good example: Cigar's Pacific Classic. Take out Siphon and he had a much greater shot. Same horse, no difference except that the lone speed horse would not have been there. The match up caused his demise, not him. Go with him he was screwed. Stay too far back, Siphon runs off on a easy lead and is uncatchable.

[bigmack]

I never look at the individuals, I look at the race interaction. The race is what to evaluate. It is only in a competitive situation that one can evaluate the merits of any horse. Change the interaction and the same animal is viewed in a completely different light.
I ran into a 4 year old in the Turf Club who said the same thing. Not exactly earth shattering thought. Chapter 1, page 1: Handicapping for Dummies.

[xtb]

Relational databases are faster, but not that much more sophisticated than the old dBase IV.


Now I'm the one chuckling.

[46zilzal]

Look at training at almost every other venue in the world and workout times are irrelevant. Just frequency is the important factor. Two venues on earth time them.

What does that say about time relevance?

[Jeff P]

The interesting part of this:

are there enough INDIVIDUAL X's that don't qualify as GENERAL X's to made this worthwhile. In other words, are there enough winners out there that don't satisfy the factors identified as necessary ones by a particular DB method OR do most of these winners that 46 thinks are outside the set actually satisfy the criteria.the_fat_man:
You said something really interesting here. Yes, I can see both sides. There are always subsets within the data itself that run contrary to the big picture implied at first glance. And for anyone willing to do their own research into those subsets, I could see that research paying off handsomely. For example, I can run DB queries for the starters of certain trainers who never work their horses fast... see what those trainers actually do with their horses... and build a successful model of that trainer's starters from there. I have no doubt whatsoever that an approach of that type could work... even if it does run contrary to the big picture. And I would be the first to give MAD PROPS to anyone who would undertake a project like that... because there are some real nuggets waiting to be discovered within the subsets.

46zilzal:
Yes, I know something like Workout Brilliance is extraneous to the horses themselves... I know there are a lot of inaccuracies and inconsistencies going on when morning works are reported... I know that each horse is an individual and has its own storyline and history of physical problems... so the big picture painted by the type of data I've presented is inherently flawed... In fact I'll freely admit that: The big picture painted by the type of data I've presented is inherently flawed.

But in spite of that...

I use stuff like Workout Brilliance in my models is because it is one of those factors that the public consistently overlooks when they bet the races... it represents what I refer to as a hidden positive... and including it in my models makes virtually all of my models perform better.

So whether speed exhibited in workouts is flawed... extraneous... or whatever other label you want to slap onto it... I don't care... If it makes my models perform better then I'm going to use it.



-jp

.

[46zilzal]

[
So whether speed exhibited in workouts is flawed... extraneous... or whatever other label you want to slap onto it... I don't care... If it makes my models perform better then I'm going to use it.



.

that's great. No one is telling you not to.

I cannot imagine depending upon them with so many inherently unknown factors at work. Drugs especially.

[ryesteve]

I cannot imagine depending upon them with so many inherently unknown factors at work.
As long as a factor works, why get caught up on the lack of perfect and complete information? Look at Beyer figs; they're predictive (and overbet) despite such unknowns as ground loss, wind, bad trips, ambiguous variant calculations, pace effects, etc. But I don't think anyone could rationally argue that because of all those unknown factors, there can't be any value to them.

[46zilzal]

As long as a factor works, why get caught up on the lack of perfect and complete information? Look at Beyer figs; they're predictive (and overbet) despite such unknowns as ground loss, wind, bad trips, ambiguous variant calculations, pace effects, etc. But I don't think anyone could rationally argue that because of all those unknown factors, there can't be any value to them.
Beyers? the last chapter of a book telling nothing about what happened in all the earlier chapters? you mean that number?

Three horses get 85's without knowing that one was on an easy lead, the second won a pace struggle and the last was after the pace collapsed. Those numbers? Ambiguous at best.Let's come in at the end of a race and based upon that little end game, describe the entire running of the contest. Whew bridge selling time. Imagine any other endeavor that forgets the other 90% of a contest. Might as well just start watching the last quarter of football games.

When something simplistic sounds overly simplistic, it usually is.

[Tom]

Look at training at almost every other venue in the world and workout times are irrelevant. Just frequency is the important factor. Two venues on earth time them.

What does that say about time relevance?

It says Jeff P has data that says your opinion is wrong.

[Tom]

46 sits here dissing the Beyers, but in fact, the program he uses relies on TrackMaster variants, calculated from class pars, another topic he disses.

Jeff, your comments about subsets is right on. I have a built in DB in HTR that lets me test the WO factor in any number of subsets - track, surface, class, trainer, etc. 46 himself talks about the power of early speed, and bemoans tracks - like Mnr yesterday - that he can't rely on. What better way to find a track's idosynchrisies than a data base? He himslef is actually doing that with a mental database and very small sample size.

But back to the original topic somebody - I forget who, not even sure if he still plays the ponies - about evaluating workouts and I offered that I had a trating that was very useful. Two days later, after countless posts about why it is no good, knowing not a thing about it - here we are.

[Tom]

Fat_Man....I suspect you think I am doing things way more complicated than I am. I am not. I assure you, I keep things simple. I am not using the DB to come up with spot plays or bets or anything like that. My WO rating is used just like you use any other factor - say, jockey. You see a jock taken off and get excited, I see a horse, 3rd off a layoff, with a hight WO number, I feel confident going back to a previous form cycle to get a paceline better than anything right before or after the layoff. 46 assumes I am using time, in fact, time is not part of the WO rating.

[46zilzal]

Another clairvoyant who seems to think he can read another's mind and KNOWS how others do what they do without ever observing even 2nd hand. Keep those Tarot cards close to your chest.

[Tom]

Another clairvoyant who seems to think he can read another's mind and KNOWS how others do what they do without ever observing even 2nd hand. Keep those Tarot cards close to your chest.

That pretty much sums up your contributions to this thread.
Are you saying you use raw times?

[46zilzal]

That pretty much sums up your contributions to this thread.
Are you saying you use raw times?
I use tangerine yogurt splashes next to my selections. Mango doesn't cut it.

[the_fat_man]

I never look at the individuals, I look at the race interaction.

Well, the only way to look at the 'individual' horse is by its performances against others. So, we're talking about the same thing.

[ryesteve]

Beyers? the last chapter of a book telling nothing about what happened in all the earlier chapters? you mean that number?
I knew that as soon as I said no one could rationally argue that Beyers had no value, you would argue that Beyers have no value.

Why critique them and point out their shortcomings? I already did that. The point is that DESPITE all those shortcomings, they are still predictive.

[Good4Now]

46,

Thanks for the most informative instructions on how to record over 8,000 posts.

[nobeyerspls]

It says Jeff P has data that says your opinion is wrong.

You wouldn't expect a simple thread on workouts to turn into the great debate. I certainly won't join the speed figure discussion but I have to tell you about a filly I owned who was in somebody's database. She worked out just fine in the morning when all alone but a physical problem sometimes surfaced when she raced. She would displace her soft palate, but only when the challenge of another horse racing up next to her caused her to swallow hard and get on the bit. Of course the displacement shut off her air and she would stop badly. When alone on the lead and relatively relaxed she would run on.
Databases and figures and all sorts of attempts to digitize performance are dealing with the vagaries of flesh, bone, blood, and muscle. How many other circumstances are out there to skew the data?

[Tom]

Your exception doesn't disprove the rule.
The bottom line is the rating is predictive. Nothing is 100%
Your filly's problems do not skew the data.

[Tom]

I use tangerine yogurt splashes next to my selections. Mango doesn't cut it.

Nothing more to say? Just change the subject?
You either use raw times or adjsuted times.
Which is it?

[DJofSD]

I follow, quite successfully, but not exclusively, the Malcolm Gladwell (author of BLINK) mantra of "Don't think blink." Keep data to a minimum effectiveness and not further.
Yes, in certain situations, blink, don't think, is the optimal strategy.

But at the same time remember Gottlieb, the marriage researcher. He and his graduate assistances spent a long time building a data base of factors both good and bad that were observed during interactions between couples.

Eventually there were 4 factors that were identified as highly predictive as to the likely success or failure of the marriage.

How is a data base of works correlated with other factors used to derive a significant likelyhood of predicting a winner any different that what Gottlieb is doing?

[Tom]

Blink, don't think! :lol:

Good one!

[46zilzal]

The amazing thing that book teaches, akin to the algorithm he reports that was initiated at Harvard and used by Cook County Hospitals' Cardiac Care Unit: find the "essence" of something, identify the minimum number of factors that predict it and forget the rest as unnecessary. That notion has been an epiphany in moving up both % winners and ROI.

Does it miss some horses? Probably, but it makes wager construction a whole lot easier.

[bigmack]

Does it miss some horses? Probably, but it makes wager construction a whole lot easier.
So does betting colors.

[DanG]

The amazing thing that book teaches, akin to the algorithm he reports that was initiated at Harvard and used by Cook County Hospitals' Cardiac Care Unit: find the "essence" of something, identify the minimum number of factors that predict it and forget the rest as unnecessary. That notion has been an epiphany in moving up both % winners and ROI.

Does it miss some horses? Probably, but it makes wager construction a whole lot easier.
Picking your spots …yes, you can narrow your focus and specialize in one discipline.

My experience is when you start playing multi-race bets where you HAVE to get involved with all types of scenarios, you had better have a very broad tool chest or you will be burning money in the long run.

[JimmyQ]

I personally put more crediblity into fast works by allowance horses or stakes horses....claimers are a tricky thing....many times I'm skeptic of these claimers throwing out a blowout work....sure seems like the connections are trying to draw attention to the horse, that he/she is for sale and healthy, so buy today. Also, you really get screwed if the clockers do the connections a favor and post a faster time for such a horse and there is indeed an ailment......but my absolute favorite workout to spot is the one the connections do on race day, and it doesn't happen a ton but if you happen to catch one of these on a claimer that they are just working him out during a race and setting him up for a future race then you can score some nice wins...recently caught a $24 horse at MNR and a $40 horse at hollywood on this exact thing in claiming events just by paying attention to video


JimmyQ

[46zilzal]

My experience is when you start playing multi-race bets where you HAVE to get involved with all types of scenarios, you had better have a very broad tool chest or you will be burning money in the long run.

NEVER play multi-race bets. Find a race that makes sense and is worthwhile, and as Andrew Beyer says crush it.

[DanG]

NEVER play multi-race bets. Find a race that makes sense and is worthwhile, and as Andrew Beyer says crush it.
Now I know your smoking something… :D

“You” never play multi-race bets, or “they” should never be played. Your far from dumb 46…think that one through before you hit the send key.

[46zilzal]

DD last time was in the 70's, Pick 3, 4 ,6? I never play them.

[PaceAdvantage]

Is it just me, or did 46 hijack yet another thread and muck it all up? I'm looking for a consensus here....

[Tom]

NEVER play multi-race bets. Find a race that makes sense and is worthwhile, and as Andrew Beyer says crush it.

Agree, PA...this sure sounds like a "musterbater" to me.
46 has lots of rules. And no data. :lol:

[nobeyerspls]

Is it just me, or did 46 hijack yet another thread and muck it all up? I'm looking for a consensus here....

How about a real-time demonstration to get the thread back to use of workouts (I was, however, really interested in the marriage predictors).

The 4th at Saratoga today has six of twelve maiden fillies who are dropping from msw to maiden claiming. That's usually a good place to find the winner although Ivory Star is one among the claimers who looks live. My Fair El, the consensus best of the day, has a decent last work but a 22 day gap from the prior work. I discount 3f works and don't like gaps so I'll try to beat this vulnerable favorite. A horse named Pizza (why do they do that) comes off of two lousy races earlier in the year. I'll ignore the turf race and go with the notion that something that was broken is now fixed using the workouts as evidence. There are some other things going on here that cloud the issue such as blinkers on and she was front wrapped in her first start.
If she beats the favorite and wins the race doing it, the horizontal plays will get a boost. Then too, if she wins and one or more of the other droppers also hit the board beating the favorite (that can happen) then the verticals pay well.
So fellow travellers on this humbling journey, chime in.

[john del riccio]

How about a real-time demonstration to get the thread back to use of workouts (I was, however, really interested in the marriage predictors).

The 4th at Saratoga today has six of twelve maiden fillies who are dropping from msw to maiden claiming. That's usually a good place to find the winner although Ivory Star is one among the claimers who looks live. My Fair El, the consensus best of the day, has a decent last work but a 22 day gap from the prior work. I discount 3f works and don't like gaps so I'll try to beat this vulnerable favorite. A horse named Pizza (why do they do that) comes off of two lousy races earlier in the year. I'll ignore the turf race and go with the notion that something that was broken is now fixed using the workouts as evidence. There are some other things going on here that cloud the issue such as blinkers on and she was front wrapped in her first start.
If she beats the favorite and wins the race doing it, the horizontal plays will get a boost. Then too, if she wins and one or more of the other droppers also hit the board beating the favorite (that can happen) then the verticals pay well.
So fellow travellers on this humbling journey, chime in.

I beleive that workout info is accurate and helpful or not accurate and not helpful. I am not doubting Jeffs findings because he is obviously a very knowlegeable handicapper, however, I think depending on where those works take place means alot. On the west coast, all works must be set up with the clockers WRT where the horse will be breaking off, the saddle cloth, how long the work will be, and what the gallop out parameters are. On the east coast, is a free for all. I have been ther first hand myself when a different horses name was given to a clocker and no wone blinks an eye. Its a very sad thing, but this happens all the time.

One of the best angles to run away from is a SHORT 3f bullet on a claimer dropping in class. first off, even a sore horse can work 35 & change while becoming unraveled in the last 1/8; if its him working at all is another issue.
claimers that work fast short & drop are the worst bets to make. maybe someone can run a database query on this so we have hard data but i think i have seen thi senough to know why..

the ABSENCE of workouts with claimers coming off a layoff can actually be a positive thing because to the "naked eye" it appears that the horse will need a race when in reality, its been prepped and the works were hidden.

its not smatrt to hcp with the conspiracy theory as the primary focus but witht the claiming game, nothing is rarely as it appears at first glance and
what may seem obviously very well may be a trap to lure or shun perspective claims.

with better horses, this is less of an issue because alw 7 stakes horses have little to hide so to speak.

john

[Tom]

Here is an example of how the HTR WO rating can be useful:


Belmont, Saratoga
MSW races for 2 yos, all distances and surfaces
First time starters
2004-present

Number horses – 1379
Number winners – 145 (10.5%) ROI – 0.70



Add WO rating over 84

Number horses – 80
Number winners – 17 (21%) ROI 0.97

[Greyfox]

Is it just me, or did 46 hijack yet another thread and muck it all up? I'm looking for a consensus here....

No. 46z did not hijack this thread.
It takes "two hands to clap." If a thread moves off topic, it is because other participants are also fueling the movement away.

If several individuals are using data base approaches to make money on horse racing via workouts, as they claim to be, good on them. But as several contributors including 46z have pointed out here, there are inherent perils involved in relying solely on morning workouts. We don't need to go over those pitfalls again.

46z 's approach is similar to the one I advocate.
On track Performance in races takes precedence over Workouts.
A workout can show you that a steed is fit and ready. In today's match-up that animal may or not be a factor depending upon who it is up against.

If workouts can be used as a "stand alone" factor, I suspect that it is in
maiden races particularly with guaging how well a number of first time starters have been doing. Even then, different trainers have different ways of bringing their runners along. Bob Baffert for instance drills his horses awfully hard. McNally and Ellis get good results with less powerful drills.

With maidens the "pattern" of workouts may also be as important as the times. For example, a lengthy break in the works is generally not considered to be a positive factor.

I like to use workouts for "spot play" action, if I use them at all.
If a trainer has been working two or three runners in tandem in the mornings
that may be an idea to watch for. If one of those runners "pops" in Race 6,
and another is entered in Race 8 or a day or two later sometimes some good long shot hits can be made.

For the most part though, betting on workouts alone would lead to the poor house IMO.

[classhandicapper]

I think computer based scans of large amounts of data are a great way to build value and probability oriented insights into the game. I think it's better to then dig for specific situations within that broad data where all the value actually exists.

Personally, I find it amazing that people actually question the usefulness of workouts in evaluating both ability and changing form.

[Robert Fischer]

thats not the worst starting point in the world.

so we have about 6 winners per meet and we are losing 3 points for the average of the last three years. What factor can we now add to make us profitable without cutting the amount of winners to an insignificant amount?





Here is an example of how the HTR WO rating can be useful:


Belmont, Saratoga
MSW races for 2 yos, all distances and surfaces
First time starters
2004-present

Number horses – 1379
Number winners – 145 (10.5%) ROI – 0.70



Add WO rating over 84

Number horses – 80
Number winners – 17 (21%) ROI 0.97

[Robert Fischer]

what about trainers?

if we find that BOb Baffert has only an 84% ROI when having the highest Workout-rating, do we eliminate him? The last several years have shown 97 overall but Baffert was training in California... Then we find out that Baffert also produces 3 times the amount of top work-out ratings than the average trainer. - he trains them all fast...

[46zilzal]

Is it just me, or did 46 hijack yet another thread and muck it all up? I'm looking for a consensus here....
If responding to the questions posed is highjacking.

[GaryG]

An uncharcteristically fast work for some trainers han be sign that a 1st timer will run big. This has been a valuable tool for me at both Mth and Crc. There is nothing like knowing a trainer's habits and strengths, as well as weaknesses.

[46zilzal]

An uncharcteristically fast work for some trainers can be sign that a 1st timer will run big. This has been a valuable tool for me at both Mth and Crc. There is nothing like knowing a trainer's habits and strengths, as well as weaknesses.
Often an uncharacteristic fast work indicates the exercise rider lost control of the animal too. Having observed this many a time, it is one factor amongst many that raw times tell you NOTHING about.

[DJofSD]

Often an uncharacteristic fast work indicates the exercise rider lost control of the animal too. Having observed this many a time, it is one factor amongst many that raw times tell you NOTHING about.

Beat me to it.

BTW, just about nothing is more scary than being on the back of a run away horse.

[GaryG]

I am talking about a workout pattern, not just one unusually fast work.

[46zilzal]

I am talking about a workout pattern, not just one unusually fast work.
Drugs, as there is no drug testing after workouts particularly if the horse drops significantly in price. Who would know?

[GaryG]

Well, I am keeping the money and will continue to use this as part of my arsenal. That is if you don't mind? Rock on zilly.

[Tom]

I didn't know they ran claiming workouts? :confused:

[classhandicapper]

Drugs, as there is no drug testing after workouts particularly if the horse drops significantly in price. Who would know?

There are exceptions to every general handicapping insight in the world.

That's why I suggested that looking at broad data as a great place to start.

However, you can and should take it further.

THere is ZERO question that fast working FTS do better than slugs (a broad general insight).

However....

If you know that trainer "X" always asks a lot in the morning and trainer "Y" rarely does, then you might come a slightly different conclusion about that 100bg work in the running line.

If you know that horse "A" and "B" just worked in company where "A" bested "B" and "B" is already competitive in limited ALW races, then "A" obviously looks dangerous in a MSW as a FTS.

If a horse was working in 48, 101, 114 before his first start where he showed speed and tired and has worked in 46, 58.3, 112 since you should realize that the probability of much improved effort is a lot higher than for the typical horse.

If a horse was working and racing regularly all season but suddenly disappears for 2 months and shows a 5 week gap in the workout line before a few slow breezes coming into today's effort, you should be a little extra skeptical (and vice versa).

Just because there might be some other reason for fast works and not every one of these horses will win does not mean you should ignore the obvioius value of work out information.

Just because some works are missed and others are mistimed does not mean there isn't value in the information as it exists.

[46zilzal]

Horses are individuals like football players, gymnasts etc. to make a blanket statement about their trainers and their abilities simply does not correlate. One size does not fit all.

FREQUENCY not time, and even then if they are racing regularly, gallops suffice.

Horses race in company showing their adaptations to the stresses of pace something NEVER seen in a work. Comparison of racing is where it's at.

[bigmack]

there are inherent perils involved in relying solely on morning workouts.

If workouts can be used as a "stand alone" factor

For the most part though, betting on workouts alone would lead to the poor house IMO.
Sage advice but whoever said anything about "solely" "stand alone" or "betting on workouts alone"?

[classhandicapper]

Horses are individuals like football players, gymnasts etc. to make a blanket statement about their trainers and their abilities simply does not correlate.

FREQUENCY not time.

You are pretty much being too silly to have a serious conversation with.

Even though all horses are individuals, it's just plain silly to say that because WOs don't correlate 100%, they don't correlate well enough to become an important handicapping factor in some situations. They do correlate very well in many specific situations and I've given you a few.

In fact, I would say my single biggest source of long shot winners year in and year out where my TOP SELECTION (most probable winner) won has been a close examination of workouts (negative and positive). Most of the rest of my longshots are just contenders that look good enough to take at the price.

[46zilzal]

good for you.......Flawed logic must sometimes work: with the evaluation of practice sessions, which have hardly anything to do with actual racing.

[GaryG]

:lol: :lol: :lol: :lol:

[Greyfox]

Sage advice but whoever said anything about "solely" "stand alone" or "betting on workouts alone"?

An earlier poster mentioned that HTR had a WO factor that sometimes shows a
"stand alone" profit for limited time periods.

I am aware that most members here are sophisticated enough to know that the work outs alone, while enhancing predictive ability in many instances, would be fools play to try and guess when those limited time periods are.
There are also other visitors to the site with only limited experience at the track that may glean that a hot workout might = a good on track effort. That is why I made the remark that betting on WO's alone would lead to the poor house.

[Tom]

Translation - he doesn't know how to use workouts.
Still waiting for data rather than opinions.......

[46zilzal]

Translation - he doesn't know how to use workouts.
Still waiting for data rather than opinions.......
The mind reader has taken up Linus' avocation with Psychiatric help 5 cents....

Years of experience has caused all the data to be put away since it did not incorporate OBSERVATION of said workouts. I sat with the clocker for weeks and it was his opinion that the correlations were weak. He should know after doing that for over 30 years.

[classhandicapper]

Years of experience has caused all the data to be put away since it did not incorporate OBSERVATION of said workouts. I sat with the clocker for weeks and it was his opinion that the correlations were weak. He should know after doing that for over 30 years.

Did you ever actually look at the DRF and examine the WOs of hundreds of first time starters, layoffs, lighty raced horses on the improve, trainers, to see if various WO patterns did CORRELATE to performance etc....?

My guess is that you haven't or have simply disregarded their potential value because somewhere along the line you heard or read that some of WOs are missed, others are mistimed, etc...

The fact is that you aren't going to get a lot of bets based on WOs because most of them don't tell you anything new or important about a horse. But there are situations where they do and if you don't pay attention you are going to miss an important clue for determining how well the horse is likely to run.

We have been trying to demonstrate that by showing you hard general data and some specific applications but you keep repeating the same nonsense.

It would be like me telling you that because your velocity ratings aren't perfectly correlated to performance there aren't some sitiuations and applications of your model that are useful.

If you choose to ignore some of the applications I highlighted, it's your loss.

[46zilzal]

Did you ever actually look at the DRF and examine the WOs of hundreds of first time starters, layoffs, lighty raced horses on the improve, trainers, to see if various WO patterns did CORRELATE to performance etc....?


yes, for years and the correlations were so few and far between I gave up on collecting them any longer as the work did not reap the reward.

Did not hear anything. I went out to the track and SAW first hand how poorly reported the reality of what went on was published.

QUIOTE:The fact is that you aren't going to get a lot of bets based on WOs

SO very true....waste of effort vs reward.

[classhandicapper]

yes, for years and the correlations were so few and far between I gave up on collecting them any longer as the work did not reap the reward.


Then the problem was that you weren't looking at and/or applying the information properly.

That says nothing about the usefulness of the information in the hands of someone that knows what they are doing with it. That goes double when very generalized data that proves the point has been supplied!

[46zilzal]

That goes double when very generalized data that proves the point has been supplied!
very generalized as to be not useful.......


Practice and exercise are just that. They are not racing......

All the exercise riders and trainers I have talked to in the AM agree with the very weak correlation between them. Not reliable. I will take the word of the people actually working them over some theoretical number crunching.

Fast in the AM does not mean competitive in the afternoon. Far too many errors between practice and the reality of the race.

[46zilzal]

Then the problem was that you weren't looking at and/or applying the information properly.


what do you add a factor, divide by the date, multiply by the trainer standings and add in some standard? One can only look at that data a very limited number of ways despite that fact that about 30% (at times) is incorrectly reported or tainted by a 160 pound exercise rider going 4 off the rail in the irons.

The MOST impressive work I ever recall was five furlongs in 1:02 2/5. Not sparkling time, but it was right after a heavy rain, between races on the training track and the rider lost an iron 3 furlongs into it. OBSERVING the mechanics of that work were far more instructive than the time. It was hardly an epiphany as that three year old was 7/9 that season. What else do all of those DRF lists tell you other than time? Trackman's comments are as rare as dodo birds these days.

[bigmack]

Classic Zilly. Take a feeble point and drill it to the ground. Yappin' with morning clockers trumps the credence of all others.

Course, he has my mic turned off. That must make him, as he would call: a putz :lol:

[classhandicapper]

what do you add a factor, divide by the date, multiply by the trainer standings and add in some standard? One can only look at that data a very limited number of ways despite that fact that about 30% (at times) is incorrectly reported or tainted by a 160 pound exercise rider going 4 off the rail in the irons.

If GENERALIZED data shows value, then that screams that more specific data within the group will have a lot of value.

I gave you some very specific examples of HOW to apply data before.

Instead of asking a question, you keep throwing out meaningless nonsense like this exercise rider crap. Of course it is possible that a single WO is misleading because it was reported incorrectly or the exercise rider was heavy etc...

I can tell you with 100% certainly that a lot of your velocity ratings are full of crap too because the starting gate was in a different position for a single race, the wind gusted, the maintenance crew worked on the track between races, moisture added to the track before racing started has evaporated, etc.... There are dozens of them. However, that does not diminish their overall usefulness to you.

Ideally, we would all like to have perfect information all the time. But we don't have perfect speed or pace figures and we don't have perfect WO information. It is all useful despite that if you know what you are doing with it.

[46zilzal]

Just how do you manipulate this data to compare exercise in isolation (zero pace pressure) to a group contest? All one can tell is fitness through frequency and then again that is subject to the individual.

I have seen literally hundreds and hundreds of quick works that were just that: quick works. When subsequent races occurred, the horse's RACE history was what prompted evaluation. The work just hinted that the physical animal was not hurting too much.

[Tom]

Save your breath (fingertips?:rolleyes:) Class, a closed mind is a terrible thing to face. 46 knows it all, so much in fact, that he doesn't need to verify anything.
He can tell us all that the money we win is a fluke and his is truly a reward for hard work.:lol:

[46zilzal]

not telling anyone to do anything. Exercise is exercise. Racing is mutually exclusive since it isn't even close to the same endeavor.

Having saved thousands and thousands of workouts, evaluated and compared them to race results, the correlation was so weak as to make me want to find out why something as dogmatic as what is prompted by the rank and file has such a weak association. That is when I started to go out and see for myself what went on. My conclusions are based upon years of those observations.

Same thing with this lasix nonsense. Furosemide is negated by the action of anti-inflammatories, but then again when 97% of the horses are taking it, how can it be a differentiating factor anyway?

[Jeff P]

http://www.paceadvantage.com/forum/showthread.php?t=38462&page=2

See post #27 in the above thread... I wanted to point out how extraneous imperfect flawed supposedly irrelevant info about the ML and Workouts can be combined to produce profitable play. I've presented a very simplistic example. But it does hint at what is possible.

46, by the way... I agree with everything you've said about the ideas expressed in "Blink."... specifically the value of simplifying things... which is actually what I end up doing. Yes, some of the models I build using a database can involve hundreds of rules for different factors... which might seem incredibly complex... but in the end - what I see on race day is really pretty simple:

1 - horse qualifies as a play according to the rules of the model

-or-

0 - horse fails to qualify as a play

which makes my decisions during live play about as simple as they can be.


-jp

.

[46zilzal]

http://www.paceadvantage.com/forum/showthread.php?t=38462&page=2

See post #27 in the above thread... I wanted to point out how extraneous imperfect flawed supposedly irrelevant info about the ML
Some, not very many, odds line makers, know the horses on the grounds fairly well. Must be why they keep their jobs.

[classhandicapper]

Here's a specific example of how I might use it.

Let's assume this is a race for lightly raced limited winner ALW level 3YOs.

I start off scanning the PPs doing my normal handicapping. I look at CJ's speed and pace figures for the horses, Beyer figures, trip notes from Logic Dictates, any trip notes I've made myself, my own bias notes, field quality info, possibly a replay or two etc....

Now suppose along the way I notice that one of the horses has a 46bg WO since his last start. That's going to catch my eye. I'm immediately going to check his other recent WOs to see if he's typically a fast work horse. I'm going to look to see who his trainer is and based on experience know if the guy tends to work all his horses fast or not. If I conclude that the work DOES look extraordinary for this horse, I'm going to check a bunch of back issues of the DRF for all the horse's WOs that no longer appear in his PPs to see if he has ever shown this kind of speed in the past. I may even look at all the WOs from that day to see if a lot of horses were working fast that day.

If I conclude there was nothing extraordinary about that work for this horse, the info is irrelevent to me. In other words, if he often works in 46 +/- a little or thereabouts, who cares.

If I conclude that the worked showed something new and much improved for this horse, I'm going to expect that there's a very good chance for a much improved effort.

Whether he will win or not is dependent on his ability relative to the other horses, his running style, and all the other factors that influence race results. But at least I know a bigger than usual effort for a horse like this is pretty likely and I can construct a bet that reflects that view.

Out of a group of horses like this some will run poorly and not improve at all (just like some other top figure horses do). Some will improve and not win. Perhaps some are just mistimed WOs etc... But I know from experience that I see an improved effort often enough to make it worthwhile to do the work because there are profits to be had.

One famous recent example like this was Barbaro. Barbaro was screaming and shouting to anyone that looks at WOs that he was going to deliver a lifetime top prior to the Derby and he did.

[46zilzal]

A good colt runs a good work. Hardly surprising.

Can't make a horse change....babies improve and learn to run for sure. One work, full of sample error might give evidence that the past performances (the real data) correlate with a form cycle improvement, but without the past performance data the work is still just exercise.

[classhandicapper]

A good colt runs a good work. Hardly surprising.

Can't make a horse change....babies improve and learn to run for sure. One work, full of sample error might give evidence that the past performances (the real data) correlate with a form cycle improvement, but without the past performance data the work is still just exercise.

You are either hopelessly closed minded or just a pain in the ass.

I find it hard to believe you could be a winning player with that attitude.

I just gave you one of my winning angles for free and you are still denying its validity. :bang:

[46zilzal]

No, I FIND workouts so marginally relevant to the races as to attribute it to what the majority of the trainers, clockers and riders do: EXERCISE.

Use them. I don't, other than to tell me that the animal is fit to run.

[classhandicapper]

Go back to your cave. ;)

[Greyfox]

I may even look at all the WOs from that day to see if a lot of horses were working fast that day.

.

Thank you classhandicapper for sharing with us what you do.
Sounds like a very thorough approach.

I'm wondering about the above statement though that I've quoted.
The DRF gives you a fraction referring to how the horse compared with others at that particular distance on that particular morning.
For example in Del Mar's 8 th race today # 2 Vegas Runner, a first time starter,
worked 4 furlongs in 48 on Del Mar's poly. That earned a rank of 9/49?
Only 8 runners equalled or bettered it that day. We know that in a quick glance.
I consider that fraction more important than the actual time.

[46zilzal]

I just gave you one of my winning angles for free and you are still denying its validity.
Thanks but I will pass valid or not to you, it is not to me...simple as that....

Having worked on the backstretch, sat with the clocker, working the gate with many of the exercise riders, the track vet etc., I have observed, first hand, otherwise...

Goes all the way back to the 2nd grade when I will told something I had observed to be totally different. Answer was then as it is now. "Am I supposed to believe you or my own eyes?" I will believe my own eyes and 40 years of experience with the races. No disrespect, we just see things differently.

[GaryG]

You are either hopelessly closed minded or just a pain the ass. I vote for a combination of the two, tilted toward the latter...

[Jake]

Speed of workout, relative to other workouts, indicate trainer intent. Fast 3 F bullet last is usually negative indicator, especially in low class claimers. Two or more successive 5f bullets is seldom negative. Trainer specific workout patterns are never weak, because they indicate where the trainer thinks the horse is at with current form. In California, look at Ron McAnally's horses in routes with 7F workout patterns; look at Avila's approach--here good works are negative indicators. There is nothing weak about workout data, the weakness lies only in the lack of understanding of trainer's methods, both to frequency and speed of workouts.

[classhandicapper]

Thank you classhandicapper for sharing with us what you do.
Sounds like a very thorough approach.

I'm wondering about the above statement though that I've quoted.
The DRF gives you a fraction referring to how the horse compared with others at that particular distance on that particular morning.
For example in Del Mar's 8 th race today # 2 Vegas Runner, a first time starter,
worked 4 furlongs in 48 on Del Mar's poly. That earned a rank of 9/49?
Only 8 runners equalled or bettered it that day. We know that in a quick glance.
I consider that fraction more important than the actual time.

I look at that ranking also, but sometimes, I am familar enough with a few of the horses on the work tab that I can make a more meaningful comparison.

As to the actual times, I mostly just handicap NY. So I am pretty geared towards what is impressive there and what is not. If I were to attempt to evaluate works at Del Mar for example, I would probably be lost for awhile until I got more used to the times on that surface. I still have a lot of trouble with the Saratoga training track for example, works with dogs up etc... But I am not looking for small improvements. I am looking for WOs that leap off the pages so I can be fairly sure that all the little things that "46" is talking about are not an issue. I don't get a lot of plays this way, but when I do, I often get a live longshot or solid midpriced horse.

[Tee]

No, I FIND workouts so marginally relevant to the races as to attribute it to what the majority of the trainers, clockers and riders do: EXERCISE.

Use them. I don't, other than to tell me that the animal is fit to run.

Either you use workout information or you don't.


Hypothetical situation - you are on vacation in a strange land :), looking down at a set of pp's at a racetrack you know next to nothing about & find a full field msw race. Half the field are first time starters & for some strange reason the other half are coming back from an extended vacation.

Now how to do you handicap this race?

Now don't cop out & say I wouldn't be betting & I would take a pass.

[46zilzal]

Hypothetical situation - you are on vacation in a strange land :), looking down at a set of pp's at a racetrack you know next to nothing about & find a full field msw race. Half the field are first time starters & for some strange reason the other half are coming back from an extended vacation.

Now how to do you handicap this race?

Now don't cop out & say I wouldn't be betting & I would take a pass.

No rational person would wager without more than that since it would be a pure gamble. Without racing experience these babies may or may not transfer swiftness to competition. Pure guesswork. One never knows how they will react to dirt in the face, being bumped, a myriad of other factors ONLY found in a race.

The baby who first learns to apportion its speed wins. The learning curve varies dramatically. No way to tell the quick learners from the dullards here.

[Tee]

Who said anything about betting? I handicap plenty of races w/o a wager being made.

How would you handicap the race in question?


No rational person would wager without more than that since it would be a pure gamble. Without racing experience these babies may or may not transfer swiftness to competition. Pure guesswork. One never knows how they will react to dirt in the face, being bumped, a myriad of other factors ONLY found in a race.

The baby who first learns to apportion its speed wins. The learning curve varies dramatically. No way to tell the quick learners from the dullards here.

[46zilzal]

Who said anything about betting? I handicap plenty of races w/o a wager being made.

How would you handicap the race in question?
throwing a dart would be as rational as reading the information there......

I see races like this all the time. Turn the page is the advice.

On each days racing card there are many races that have no defined logic. After a perusal, they are left where they are without wasting any more time.

I once was challenged at a public handicappers forum put on by the local track of a race just like this. My advice was to get a cup of coffee. The crowd was sort of shocked. I challenged any one to give me any logic to handicapping so many unknowns and no one picked the winner or even the place horses.

You have to put in your time where there is a chance to wager.

[Tee]

Please, please step stop using gambling/wagering as a way to dance around the question.

I'm asking you to do a simple handicapping exercise. You have all the information in front of your face. A full tab or workouts, pedigree information, trainer & jock information, the tote board, the post parade etc, etc.

A simple handicapping exercise - narrow it down to a couple of contenders & watch em run around the track w/o a wager.



throwing a dart would be as rational as reading the information there......

I see races like this all the time. Turn the page is the advice.

On each days racing card there are many races that have no defined logic. After a perusal, they are left where they are without wasting any more time.

I once was challenged at a public handicappers forum put on by the local track of a race just like this. My advice was to get a cup of coffee. The crowd was sort of shocked. I challenged any one to give me any logic to handicapping so many unknowns and no one picked the winner or even the place horses.

You have to put in your time where there is a chance to wager.

[46zilzal]

Please, please step stop using gambling/wagering as a way to dance around the question.

I'm asking you to do a simple handicapping exercise. You have all the information in front of your face. A full tab or workouts, pedigree information, trainer & jock information, the tote board, the post parade etc, etc.

A simple handicapping exercise - narrow it down to a couple of contenders & watch em run around the track w/o a wager.


Even without wagering, knowing our race course, I would wager on post position or the runners that had started having the best move to the 2nd call assuming this were a sprint on dirt.

Pedigree? I just think of all the full brothers and sisters to champions who were no where to be found at the finish line.

[46zilzal]

Imagine the same scenario is any other sport.

You have to decide on who would win a basketball game just based on team workouts. You know nothing about real defense, passing or any of the other skill without a few match-ups to go by.

A boxing match just by watching the sparing. A baseball game just bay watching batting practice.....

The list goes on an on. Exercise is not the game. It tells you fitness of the participants, but without competition, the translation of practice to game time is all theoretical.

[bigmack]

Tee - You'd be better off doing this. Same effect, no typing

http://i165.photobucket.com/albums/u70/macktime/headbangsoncomputer_t.gif

[classhandicapper]

throwing a dart would be as rational as reading the information there......

I took $3.10 + rebate on Maimonides and $5.20 + rebate on Z Humor yesterday and was very comfortable I had a slight overlay on both. It is very possible to have value oriented insights into races that look like guessing games to most people.

[46zilzal]

I took $3.10 + rebate on Maimonides and $5.20 + rebate on Z Humor yesterday and was very comfortable I had a slight overlay on both. It is very possible to have value oriented insights into races that look like guessing games to most people.
You play the game your way, others will play it their way.....Rebate? okay

[46zilzal]

A friend told me he picked Mike Fox in the Queen's Plate off works.
This is a class colt with ability. Three works since the last race.

[46zilzal]

I chose Mike Fox based upon improvement in the last two comparable pace scenarios The races told me what to do, the exercise told me that this one was in good shape.

[Pace Cap'n]

Has anyone thought to ask Clocker Bob?

[GaryG]

Has anyone thought to ask Clocker Bob?You mean Bob Kachur in SoCal? I knew him at SA in the 80s.

[46zilzal]

Another thing never discussed here is that a 'work out' (unless specifically from the gate) is merely the TIMED portion of a larger move. I have often heard a trainer tell the rider to actually work 7 or 8 furlongs in an announced work of 5. Depending upon the exercise rider, they crank up the horse for that timed portion differently: a few go hell bent the few furlongs before, some very carefully bringing up the tempo as the timed part continues and then they go on, at speed, past the finish line to run out in a very good time which is NOT reported.

All that is reported is the timed portion. Another major source of error. Imagine how relevant a 4 furlong portion is in a seven furlong move. Without the trackman's report, or unless you are there, one never knows the actual trainer intent.

[Tee]

I hear ya, but it's kinda fun.


Tee - You'd be better off doing this. Same effect, no typing

http://i165.photobucket.com/albums/u70/macktime/headbangsoncomputer_t.gif

[Tee]

This is along the lines of what I was thinking about at the eye md's office waiting room this afternoon.

What one might think it too little info is more than enough for another. The same can be said for too much info as well imo.


It is very possible to have value oriented insights into races that look like guessing games to most people.

[Tom]

You play the game your way, others will play it their way.....Rebate? okay

Then why do you constantly keep coming up with reasons why things won't work while others are using them profitably?

You still make assumptions you know what others are doing and are totally wrong. BTW, good thing you had accurate track variants based on man made class pars to get those numbers for Mike Fox, huh?

[46zilzal]

Sacred cows of handicapping are usually accepted as dogma without every questioning them. When I was a "grandstander' (those sanitary individuals who's closest association to a race horse was twenty feet away in the winner's circle), I held that dogma as unquestioned, but once working at the track, where the real up close and personal racing happens, I observed enough and was told enough to question many of those ivory tower aspects of handicapping that I even believed.

Like geometry, one has to prove the dogma to oneself by experimentation not lock step agreement. This goes for everything in time. Things change.

More observation, more hands on experience added to a re-evaluation of these unassailable ideas. There were many questions that the data did not represent. The reality as accepted by the masses was in fact, not what they were told. The information wasn't as rock solid as appreciated from the grandstanding sanitized position. Hell, even the reporting of that information was full of inaccuracies.

The mantra is "question the dogma." Put it to the test, and if it is questionable, don't accept it.

[Tom]

But, stuff you are dissing has been put to the test, and passed, and is being used profitably.
Againk you assume you know what others are doing and you do not.

[Jeff P]

Visual... Jeff P climbing up onto a wooden soapbox, putting two fingers to his mouth... giving a LOUD whistle... and saying "HEY! Can I have your ATTENTION for a minute?"

Here's the thing... We all use data that's flawed in some way.

I take Bris Pace and Speed Figs (no doubt many posters here are already chuckling at the words Bris Pace and Speed Figs) and adjust them in my own convoluted way to arrive at figs of my own. I'm 100 pct sure the process I use introduces error or is flawed in the eyes of almost everybody reading this.

I have my own hard coded algorithm that creates a numerical score for trainer intent - man, some of you guys who understand what trainers actually do would have tons of fun raking that one over the coals.

I have another algorithm that creates a numerical score representing the ability of the human connections... again, someone who has actually ridden in races, trained horses, or worked as a jockey agent might well be rolling on the floor laughing if they heard me describe what I'm really attempting to quantify...

I have a handful of separate algorithms that take spacing and speed of workouts and spacing and speed of races to arrive at a numerical score that represents current form... I'm guessing that anyone who has ever owned race horses would scoff at such an idea.

I have other algorithms that attach numerical scores to running styles... and because it is different than PctE and PctMedian calculated in a pure Sartin sense - I'll guarantee you I've been vehemently asked a hundred times "Why the hell would you do THAT?"

I have another algorithm that evaluates workout speed (a subject already beaten to death in this thread)... no need to even go there again...

I have a some other algorithms that generate probability assessments... and while they have proven to be really accurate over some pretty substantial data samples I've seen them ridiculed by a few posters here because they are based on factors that are different than what other software vendors have published to date...

the list goes on... I have hundreds of algorithms...

all of them flawed or imperfect or irrelevant according to SOMEBODY...

and yet all of them useable to me in one way or another because historical data samples clearly show two things... the factor in question, despite its flaws, is:

1. predictive

2. consistently ignored by the public

and if I want to reverse weight for a factor in a model then there's a third thing:

3. over-used by the public

Guys, this isn't rocket science. It's a gambling game. All you need to win is an understanding of those three things. Whether you create models using sophisticated software algorithms, or sit down with just a Form and a pencil - if you handicap in such a way that you focus on those three things - and bet intelligently - and you do nothing else - you can win at this game.

Visual... Jeff P stepping down off the soapbox now... everybody in the room shakes their heads and mumbles "What a goober," before commencing to handicap exactly the same way they have been doing it for years.

-jp

.

[46zilzal]

Everyone has their way of evaluation. Some use more and a very few use far less. If it works for the individual that is their reality.

[the_fat_man]

Jeff

You seem to be just about the only one in this discussion that actually GETS IT. You're taking flawed data (as all models are incomplete by definition) and you're looking for 'regularities' in that data that allow you to make reasonable predictions. You don't care how closely YOUR DATA represents the actual thing itself, as your model is such that your results are consistent. Your resoning involves your models not the actual game. Great. It works. Go with it.

For the past 2 days, there's a back and forth going on between those in your camp (who don't exactly get it) and 46, whose argument all along is that THiS DATA is not faithful to the thing itself. He wants to handicap with more 'representative' starting points. Why? He's worked on the racetrack and knows how things 'really are'. Good for him.

I can see his perspective, as I'm a trip handicapper, who also happens to have worked on the track and knows, somewhat, how things work. I could care less, in most cases, what the figures say, or what someone's database spits out. If I see something in a race that has proven to be 'predictive' in the past, I go with it. I'm not doing anything different than you or the other database people out there. I'm basically following Hume: something has shown to effective in the past, and I assume it will continue to be effective in the future. When it stops working, I stop using it. I watch races, you play with numbers. We both deal with models.

So, why is this discussion still going on at EXTREMES?

[46zilzal]

46, whose argument all along is that THiS DATA is not faithful to the thing itself. He wants to handicap with more 'representative' starting points. Why? He's worked on the racetrack and knows how things 'really are.
No 46 wants to use race data to evaluate races, not exercise. Just as you use the trip, IN A RACE, not in some a.m. exercise....

[the_fat_man]

No 46 wants to use race data to evaluate races, not exercise. Just as you use the trip, IN A RACE, not in some a.m. exercise....

That would be what 'more representative starting points' refers to.

[bigmack]

So, why is this discussion still going on at EXTREMES?
Start looking at the posts from the human drone, Zilly and you'll find your answer. You're not going to start with your condescendingly pedantic ways again are you Fats? You know, go out and hotwalk and all that jazz? You really want someone who gets it? Here ya go:

Jeff - Whatever you are successful with, and enjoy doing .

Whose way is better? What factors mean something or nothing? We'll all come to our own conclusions. Ok by you?

[the_fat_man]

Start looking at the posts from the human drone, Zilly and you'll find your answer. You're not going to start with your condescendingly pedantic ways again are you Fats? You know, go out and hotwalk and all that jazz? You really want someone who gets it? Here ya go:



Whose way is better? What factors mean something or nothing? We'll all come to our own conclusions. Ok by you?

Mack

Typically, before one responds, one needs to understand what one is responding to. Take a moment, reflect, and see if you can understanding, even remotely, what my previous post 'means'.

You can do it. An education is probably a prerequisite but life experience might get it done as well.

Is that all right with you?

The fkin problem on this site is that there can never be an educated exchange of ideas cause, in most cases, the education is lacking.

[46zilzal]

Isn't it fun? On another website there is a clown who's sole purpose is to counter everything I post . Name is foadoc. Contributes nothing to the conversation except x-rated ranting. Seems his brother posts here: contributes nothing except personality digs.

[bigmack]

The fkin problem on this site is that there can never be an educated exchange of ideas cause, in most cases, the education is lacking.
If I recall correctly your posts about the other sites you peruse through you run into that problem as well. It's always "the other guy" not coming up to your level huh? Like I said before, condescendingly pedantic.

Keep us posted on how DrugS is doing with his selections as your skill is exponentially more game. Or something like that.

[the_fat_man]

If I recall correctly your posts about the other sites you peruse through you run into that problem as well. It's always "the other guy" not coming up to your level huh? Like I said before, condescendingly pedantic.

Keep us posted on how DrugS is doing with his selections as your skill is exponentially more game. Or something like that.

Going off topic, are we? Backtracking?

The problem here, Mack, is that you've TWICE failed to comprehend that I'm not defending 46's position over yours (and the others) BUT rather
stating that both work. If it were me, I'd feel pretty stupid and reASSESS.

As for DrugS, it's both very disappointing and somewhat of a rude awakening that someone as talented as he is can struggle like he has (this, of course, refers only to the selections on his site --for all I know, he's having a huge meet, and could care less what he puts on his site). But hey, he posts on a public site and I'm certainly justified in commenting about it. That's the way things are on the web.

Now, rather than constantly trying to stir up shit, what not offer something of value? I (naively) come here to discuss racing.

Hope this clears it up for you.

[PaceAdvantage]

No. 46z did not hijack this thread.

OK then....my apologies....

[Tom]

If anyone kidnapped this thread, would anyone else pay the ransom?
Maybe over the weekend, TFM and 46 can figure out what this thread was all about.
Maybe there will be peace in the middle east by Monday, too. :bang:

[nobeyerspls]

and yet all of them useable to me in one way or another because historical data samples clearly show two things... the factor in question, despite its flaws, is:

1. predictive

2. consistently ignored by the public

and if I want to reverse weight for a factor in a model then there's a third thing:

3. over-used by the public

Guys, this isn't rocket science. It's a gambling game. All you need to win is an understanding of those three things. Whether you create models using sophisticated software algorithms, or sit down with just a Form and a pencil - if you handicap in such a way that you focus on those three things - and bet intelligently - and you do nothing else - you can win at this game.

.

From a pencil and racing form guy who's trying to remember what an algorithm is, you nailed it.

Predictive - not much due to an overabundance of data and not enough real information, i.e. the type you could scope out with a hidden camera and mike in every stall.

Consistently ignored by the public - several angles that I use that produce live longshots.

Overused by the public - jockeys and beyer speed figures, especially when they combine to produce a 4/5 favorite who will be off the board.

Now let's see: Algorithm - the Arabic system of numeration and calculation .......jeez, where the hell is my pencil......

[46zilzal]

If anyone kidnapped this thread, would anyone else pay the ransom?

Presented many, rational observations that occur at every race track, to some degree, in North America that puts the workout data in need of serious review with multiple reasons that it is flawed. Not a person here could contend that the reasons given are not accurate.

There are so many extra factors at work adding in or subtracting from that reported time as to make it, alone, of questionable relevance.

[Tom]

Not when you have people out here using it, quantifying it, qualifying it, and making money off it. You just have not a clue as to what we are doing and yet you spout of repeatedly about things that are not related to what we are talking about. Just because YOU cannot do it, doesn't mean others can't.

You have presented NOTHING in this thread relevant to the topic at hand.Jeff presented a nice illustration of how he uses WO data and has hard evidence that it increases his models' effectiveness. Your OPINION is that is doens't work.

What a strange dude you are.

[bellsbendboy]

Certainly a bit perplexed at the amount of interest in this thread as well as the esoteric methods employed in analyzing workout data. I found little to argue with, although my knowledge pertaining to computers is primitive at best. I can say that about six years ago, I started using my computer to download the DRF, visit track websites, brisnet, blogs etc. and my R.O.I. exploded.

Workouts are a big part of my game. I agree with the post that used the term "streaming movie" and disagreed with the three furlong work on a dropping claimer being negative.

I would add that the trainer knows his horse and his/her intent is always of value. A good example is when Pletcher worked 'Rags and Circular Quay in company all spring. He is lazy as can be and her with competitivesness they made a perfect "team" always finishing noses apart. She won the Oaks after a 55 day layoff and fitness was never a concern.

Generally, younger horses need more work than older, colts more than fillies, dirt more than grass and certainly allowance types much more than claimers.

For the data base clan, one or two works out of the six or eight are important, and that importance, usually does not relate to time. I suspect filters are added for this and other data pieces, but then a level of personal supposition is added, and that would seem to water down predictability?

Lastly, I try to treat horses as individuals, although in big barns it is difficult, for trainers to give a singular horse "private" attention. One additional angle not mentioned is the gate work for a first time starter. Most jurisdictions require such, but gate approval need not appear on the work tab. If a horse stands well, or breaks and stops after a furlong, a gate card is approved. Case in point, yesterdays last at Saratoga. Prado jumped off a very logical Dutrow speed horse (the rail was up) to ride a first timer for Badgett. The firster "Pure Tantrum" did not show any gate activity on his work tab, and was bumped both sides at the break. He finished very well is bred top and bottom for a stretchout... a good one to watch for. Thanks for the input. BBB

[46zilzal]

The majority of turfers run better on that surface than the dirt.

Most major venues keep workouts on the lawn to a minimum to keep the surface from overuse.

Where do these runners work out? On a surface (dirt) on which they are not going to run, and upon which they don't run as well. These work outs equate to their races?

A prime reason to look for frequency not time. The other day when Shakespeare came back from a prolonged injury time out it was the frequency of works that gave some idea of his being back.

[46zilzal]

You have presented NOTHING in this thread relevant to the topic at hand.Jeff presented a nice illustration of how he uses WO data and has hard evidence that it increases his models' effectiveness. Your OPINION is that is doesn't work.

Only about 10 reasons as to why workout data is not accurate.

This isn't the OFF TOPIC arena.

[the_fat_man]

If anyone kidnapped this thread, would anyone else pay the ransom?
Maybe over the weekend, TFM and 46 can figure out what this thread was all about.
Maybe there will be peace in the middle east by Monday, too. :bang:


Tom

I've only posted a couple of times on this thread.

But, tell me, why is it that Jeff, for example, seems to understand my perspective (and the meaning of my posts) while you STILL seem to think that I'm pushing something threatening?

I suggest you go back and slowly reread what I've written. You might be pleasantly surprised. My position, all along, is that BOTH methods work because, essentially, both methods are doing 'the same thing'. The difference, in other words, is essentially one of degree and not of kind (even though one advocates using the 'thing' itself, while the other advocates using models of that thing.) The point might be a bit subtle and perhaps the presentation a bit abstruse but it's there, man. Look for it.

[Tom]

TFM....I do not diagree with you....BOTH methods can work if you know how to use them. The point I am making - 99% to Zilly - is that that is not what we were talking about. The thread started out how one could evaluate works, and several ideas were posted. Then zilly comes in and argues that nothing works, that we are all wrong, that his way is the truth, the way, the light...:D
and in spite of evidence that people are using WO data effectively, he presents ZERO data and continues to tell us we are wrong. To me, that is the height of ignorance - to tell somebody something can't be done while he is doing it!

He has no idea at all how I use the WO rating or what it is, and yet he continues to argue about it. I have told him several time it doesn't use time, and he still keeps refereing to times. Two posts ago is his latest. The guy is thicker than leg hair on BigFoot!

I just threw you into it because you posted today - I was hoping he would start following you home - I'm tired of feeding this lost puppy!:lol:

I get your viewpoint, I agree with it, I'm just not arguing that is all.


(Maybe I can stick Dan with him???)


WE are on the same page.

[46zilzal]

Isn't fun to be CONTINUALLY misquoted?

Workout data is flawed: gave many reasons why...that's that.

[Tom]

Only about 10 reasons as to why workout data is not accurate.

This isn't the OFF TOPIC arena.

All data, or some data?
See, when you have no clue how someone is doing someting, it is hard to argue intelligently.

Show us all how smart you really are - tell me how I am using the data. Show us you have a clue.

Tick, tock.

[Robert Fischer]

Prado jumped off a very logical Dutrow speed horse (the rail was up) to ride a first timer for Badgett.


You think Prado jumped off ?

What is your reasoning for that? Maybe I will learn something.

my estimate (could be wrong), was that Dutrow thought he should have won last out. He runs him back in 5 days and gets Prado the hell off because Gomez is available.

[46zilzal]

One never knows, unless they observe the work, which are flawed and which are not for all the reasons I have posted and then there are the drugs which no one knows other than the barn.

To trust a printed time without knowing the context of how that time was reached is seriously flawed. The reader is wholly dependent (much like a Beyer which only tells final time without reference to how that time was reached) upon filling in the "blanks" without actually knowing a thing about the mechanisms involved in that time, so frequency of works is the only trusted aspect of this whole exercise.

[classhandicapper]

Only about 10 reasons as to why workout data is not accurate.


I can give you about 25 reasons your velocity ratings aren't accurate either. In some cases IMO they are preposterous nonsense, but for some reason you swear by them and continue to use them.

That must mean that despite those flaws and inaccuracies, the information is still accurate and relevant enough to have predictive and gambling value if used properly.

What you refuse to accept is that despite many of the problems with workouts that you and others have identified, there are profitable applications of them when they are used in combination with a solid handicapping foundation. Maybe you can't do it, but a lot of rest of of us can because we have insights that you don't have and refuse to learn because of your stubborness.

[46zilzal]

I can give you about 25 reasons your velocity ratings aren't accurate either. In some cases IMO they are preposterous nonsense, but for some reason you swear by them and continue to use them.

Strange the final time is a velocity and that is what many are swearing by here without a clue as to which were done which way.

25? I'm game to hear, but velocity is not what we use, it is energy distribution which, over the performances of a horse's history does not change that much.

[classhandicapper]

Strange the final time is a velocity and that is what many are swearing by here without a clue as to which were done which way.

25? I'm game to hear, but velocity is not what we use, it is energy distribution which, over the performances of a horse's history does not change that much.

It doesn't matter whether you call it energy, velocity, or anything else.

Any measurement that is based on the clock will be flawed a lot of the time because of variations on where the starting gate is placed, changes in track speed due maintenance during the card, changes in track moisture due to evaporation (let alone rain), changes in wind direction and speed, gusts of wind, drafting, changes in the relationships between distances because the track is not uniform, etc.... It's almost endless. The best figuire makers try to account for all this stuff, but it impossible to be right all the time.

That doesn't count all the flaws in the models that try to capture and formulize the interreationships between pace and energy use, pace and final time, etc... when tracks vary from day to day and from race to race in that regard also.

Despite all those problems (and many others) and the loads of errors that occur (both minor and major), competent figure makers and handicappers of all styles continue to use them and can be profitable with them if they are skilled and knowledgeable enough.

You are starting from the assmption that because some WO data is flawed it's not usable without realizing that the very backbone of your handicapping is also wildly flawed much of the time.

[46zilzal]

I work the gate every race day and it varies in placement about a foot as I leave by backpack right across the wall from the end of it when they first position it. That doesn't represent the timed portion of the race anyway.

By using multiple lines to access the basic energy distribution of each horse, all those MINOR variations (which other than track changes) interact to negate one another. Horse are not like humans: when they are in form they are amazingly honest and consistent.

Energy distribution is independent of velocity. To see if the track has been souped, there is a variable that allows evaluation of track speed on any given day by energy distribution. At some tracks (Woodbine, inner Aqueduct) following that through a program can elucidate track bias long before the crowd picks it up.

[Tom]

Class, maybe you can get through - I do not use WO time. He can't seem to comprehend that.

And Energy distribution is caluclated from early velocity, late velocity, and a compound called total energy - so remove all velocity, and you have no energy at all.

If one of the velocity segments is wrong, energy is wrong too. Can't have energy without velocity.

And with that, I am vacating this thread - no point in arguing with someone who will not read what is being posted and argue with it anyways.
Gotta go print out my PEDIGREE ratings for today! :eek::lol:

[classhandicapper]

Everyone that makes time based figures of any type tinkers with them to account for track speed and the other variations I mentioned, but none of them is accurate all the time. The game is way too complex to measure all these interrelated things. However, we can get close enough often enough for the figures to have value.

[46zilzal]

If one of the velocity segments is wrong, energy is wrong too. Can't have energy without velocity.


That is why rational people never use a single line. The intrinsic energy distribution of a horse in response to pace pressure will show itself over several contests. Otherwise you have sample error, i.e. using the Blue Grass pace line for example or when a horse is a standard presser, you don't look at the lines it ran LATE in races where the pace scenario was different and collapsed.

[Greyfox]

Energy distribution is independent of velocity. .

As Tom has accurately pointed out, no velocity = no energy.
46Z I think that you meant to imply whether it was high velocity or low velocity
the distribution is measured similarly - an early to total ratio.

Having said that weren't we discussing Work Outs? The thread is not about track energy, unless you are trying to demonstrate that the Workouts will
be faster when the track variant is lower and vice versa, in general.
(Which is why I put more faith in "ranking" of workouts versus raw times.)

[46zilzal]

As Tom has accurately pointed out, no velocity = no energy.
46Z I think that you meant to imply whether it was high velocity or low velocity
the distribution is measured similarly - an early to total ratio.


nope distribution often occurs close to the same way whether the velocity goes up or down as it represents segemental percentages of the total.

I responded to questions put to me on velocity and energy.

[DanG]

nope distribution often occurs close to the same way whether the velocity goes up or down as it represents segemental percentages of the total.

I responded to questions put to me on velocity and energy.
I admire your stamina 46!

In my brief time here something has become apparent to me at least. In the monster threads that go over 100 posts, on average you DOMINATE posts 126 trough 184…

Statistical anomaly…possibly, but I am labeling you a ‘sustained runner by your posting energy distribution. I will not bet you in any thread < 100 posts, but in the super marathons I’m inclined to take the rubber band off the cabbage and plunge my brains out! :jump:

[bellsbendboy]

R. Fischer

Please do not take this the wrong way, but once you stop learning in this game you are toast. Jockey movement is one of the least understood handicapping factors and subsequently, will offer competent cappers opportunity.

Yesterday Prado "jumped" off LHOTSE QUALITY to ride a first time starter.

Why?

Analysis of LHotse Quality: Debuted as a sophomore for a tag for a very sharp conditioner and was taken by Dutrow. After one race he came unglued and returned on the inner ten months later, sans Lasix! Lhotse weakened in two subsequent starts but in reality they were awful. A turf sprint, a dirt claming heat, then a turf sprint showed identical, well under par Beyers of 64! Clearly this horse is going nowhere.

Prado and his agent elected to pilot Pure Tantrum who is an extremely well bred 3yo, has been working very well and has a ton of upside. Note the lone four furlong work, its the most telltale on his tab by far, and stamps him as a nice colt. If anyones, data base indicates why this work is telltale, which I would not expect, they have some game indeed.

Lhotse won at some three to one after getting a very slow opening quarter as lonbe speed with the rail up. He is doubtful to win again, ever, and only did so because of the plethora of train wrecks behind him. 'Tantrum breaks his maiden later this meet and has a chance to earn a bunch. Prado and his agent made the right call, in our opinion. Hope this helps. BBB

[46zilzal]

Statistical anomaly…possibly, but I am labeling you a ‘sustained runner by your posting energy distribution. I will not bet you in any thread < 100 posts, but in the super marathons I’m inclined to take the rubber band off the cabbage and plunge my brains out!
Akin to the knowledgeable Witch Doctor, I understand that stamina is a disappearing trait in the thoroughbred on the dirt and try to do my part.

The last stamina dirt sire to become a chef was Run the Gauntlet or Stage Door Johnny wasn't it?

When there are no dirt stamina contests at graded levels offered, how do you know where proven stamina sires come from?????

[Robert Fischer]

R. Fischer

Please do not take this the wrong way, but once you stop learning in this game you are toast.
Which is why i asked your opinion in the true sense of interest.




Yesterday Prado "jumped" off LHOTSE QUALITY to ride a first time starter.

Why?

Analysis of LHotse Quality: Debuted as a sophomore for a tag for a very sharp conditioner and was taken by Dutrow. After one race he came unglued and returned on the inner ten months later, sans Lasix! Lhotse weakened in two subsequent starts but in reality they were awful. A turf sprint, a dirt claming heat, then a turf sprint showed identical, well under par Beyers of 64! Clearly this horse is going nowhere.

Prado and his agent elected to pilot Pure Tantrum who is an extremely well bred 3yo, has been working very well and has a ton of upside. Note the lone four furlong work, its the most telltale on his tab by far, and stamps him as a nice colt. If anyones, data base indicates why this work is telltale, which I would not expect, they have some game indeed.

Lhotse won at some three to one after getting a very slow opening quarter as lonbe speed with the rail up. He is doubtful to win again, ever, and only did so because of the plethora of train wrecks behind him. 'Tantrum breaks his maiden later this meet and has a chance to earn a bunch. Prado and his agent made the right call, in our opinion. Hope this helps. BBB

You make a good case. What bothers me is that the Dutrow horse was a decent bet to pick up a check in THAT race. Pure Tantrum may do well in the future, I don't guess that he has any big purses or stakes necessarily.
Also the previous ride by Prado could have debatably been better and Dutrow fires the horse right back on 5 days rest as if he expects a check from this horse...

I am not dismissing what you say here, I am actually not 100% convinced one way or another, but the information involved could all be useful.

[classhandicapper]

Class, maybe you can get through - I do not use WO time. He can't seem to comprehend that.



He can comprehend it, he just doesn't want to accept that some people have insights that he doesn't.

I happen to use WO time as well as frequency for hints about ability, trainer intent, current form, potential physical issues etc....

I've been trying to explain to him that the accuracy issues he is so concerned about are not as relevant as he thinks. I'm not worried about measuring WOs to the level of accuracy I might measure speed figures. You don't need to in order for them to be effective.

For example, when I saw Hard Spun's 22 44 in company prior to the Derby, that was significant to me. It told me that Hard Spun was working much better than he ever had and was probably doing very well. But it also told me he was probably going to show a lot of speed, be less likely to rate, and perhaps not be ready for a 10F race. Even if I later found out that the fractions were off for one reason or another, the WO was so extreme it would not change the conclusion.

If people have applications and insights that are not at all concerned with the time of WOs, that's fine too. Some of mine are like that also.

He just doesn't want to accept any of it. :bang:

[Tom]

I use times in some cases, too. In spite of what you may have read, they can be useful.:bang:
I focus more on 4 and 5 furlong works - 3 is too short and 6 has a pace element in it. 4 and 5, IMHO are what the horse is at the time. I think you get it that we are not betting horses off works, but using the works as an indicator of current form. You mention that fast work by HS, and I bet you did not bet him off that but off a horse with proven ability that showed you he was fit and might even be getting better. 46 even said in one post the his energy told him the horse could win and the WO told him it was fit. That is the exact point he has failed to acknowledge all week!:lol: He is not one to discuss anything, though. I just enjoyed watching him dance.

Now, I gotta go check out the works and find me a fast horse to bet at Saratoga...getting ready for a road trip. BillW is holding a rack of ribs for me!:p

[jotb]

R. Fischer

Please do not take this the wrong way, but once you stop learning in this game you are toast. Jockey movement is one of the least understood handicapping factors and subsequently, will offer competent cappers opportunity.

Yesterday Prado "jumped" off LHOTSE QUALITY to ride a first time starter.

Why?

Analysis of LHotse Quality: Debuted as a sophomore for a tag for a very sharp conditioner and was taken by Dutrow. After one race he came unglued and returned on the inner ten months later, sans Lasix! Lhotse weakened in two subsequent starts but in reality they were awful. A turf sprint, a dirt claming heat, then a turf sprint showed identical, well under par Beyers of 64! Clearly this horse is going nowhere.

Prado and his agent elected to pilot Pure Tantrum who is an extremely well bred 3yo, has been working very well and has a ton of upside. Note the lone four furlong work, its the most telltale on his tab by far, and stamps him as a nice colt. If anyones, data base indicates why this work is telltale, which I would not expect, they have some game indeed.

Lhotse won at some three to one after getting a very slow opening quarter as lonbe speed with the rail up. He is doubtful to win again, ever, and only did so because of the plethora of train wrecks behind him. 'Tantrum breaks his maiden later this meet and has a chance to earn a bunch. Prado and his agent made the right call, in our opinion. Hope this helps. BBB


Hello:

Looking down the road for an agent can be a difficult chore. In this case, the agent went for the firster but it could have been he gave the trainer the call from the get-go so maybe the agent honored the call. There is something you might have missed here and that is Dutrow wheeled this horse back on 5 days rest. Prado's agent never figured Dutrow to do this and he already might have had his book marked for another spot for Dutrow and already had the call for Pure Tantrum. As an agent you normally take the now horse especially when they are knocking on the door like LHOTSE QUALITY was unless you know for sure Pure Tantrum would win first out. I would think Pure Tantrum might be heading longer in terms of distance down the road and by then, with a rider like Prado I'm sure the agent already has another horse for that race down the road. Who know if Dutrow even called the agent ahead of time to let him know he was going to wheel him back that quick and then if he did the agent has to think if this horse will make the overnight because of the short rest. If the agent knew Dutrow was going in, I'm sure he was praying the race would overfill to keep Dutrows horse out. The next time this race comes up the agent must make sure Pure Tantrum is going in and the agent needs to see what else he will have for Prado on that day and make sure he has Prado on the better horse for that race.

Joe

[Robert Fischer]

... Dutrow wheeled this horse back on 5 days rest. Prado's agent never figured Dutrow to do this ...


this is probably the true solution to the "mystery".

[BIG49010]

Getting back to the workouts, or lack of them, Dutrow is 6 wins for 24 starts with horses without them. If you look at works for this horse, he was going to the track for a work or race every 5 days for several months, so he was due for a work, Dutrow decided to race and get paid instead. If and only if, he got in the race.

[jotb]

If you bring the same 10 horses together, on the same race track, at the same distance repeatedly every 21 days over the season, you will never have the same ten competitors meeting each other. As Hericlitus said in 500 BC "you can never step in the same river twice." Things change.

The racing surface will change, the form cycle will change each animal, physical nature of the animals will change, the riders might change, the nutrition might change as some horses change barns etc. etc.

The competitors TODAY are what we evaluate, not some peripheral information.

Owners pay the bills. They need to know that it is going to have a good return. The joke on the backstretch is that works are always faster the day the owners are there: usually weekends when they come out to see first hand. A Wednesday morning work, with the 150 lb exercise rider up has no observers other than the stable. A Saturday AM work, with a 115 jockey up, needs to impress that owner who pays the bills so it strangely gets faster. Still just exercise one way or the other.


So very true.

Joe

[098poi]

I probably lost 40 of my 80 IQ points reading this thread but here goes. Per James Quinn (Handicappers Condition Book) from a form perspective if a horse has not raced in 30 days and shows a 5F workout in the last 14 days or a 4F workout in the last seven days the horse should NOT be eliminated based on race recency. The times of the workouts don't matter. (This is in another book not THCB, I forget the title)

[GaryG]

It seemed to me when I read those Quinn books that he lifted many of his form criteria from Scott in How Will Your Horse Run Today, almost verbatim.

[098poi]

He credits Scott in THCB. I highly recommend THCB for those that don't have it.

[Tom]

THCB- Scott or Quinn?

The Handicapper's Condition Book?

[GaryG]

THCB- Scott or Quinn?

The Handicapper's Condition Book?Quinn....his best book by far

[Fastracehorse]

I developed a factor in JCapper called Workout Brilliance which is based primarily on raw workout time. Despite everything you read in print about workout time being irrelevant, and everything you read about workout times being inaccurate and misreported... the rating has proven itself to be statistically relevant when applied over significant data samples.

My calendar year 2006 database shows the following:

WOBRILL
By: Workout Brilliance Rank (2006 1st time starters removed)

Rank Gain Bet Roi Wins Plays Pct Impact
1 -5593.80 56182.00 0.9004 4814 28091 .1714 1.3548
2 -10511.50 59542.00 0.8235 4374 29771 .1469 1.1615
3 -15458.50 59846.00 0.7417 3935 29923 .1315 1.0396
4 -14442.20 57026.00 0.7467 3475 28513 .1219 0.9635
5 -13060.60 48320.00 0.7297 2812 24160 .1164 0.9201
6 -10400.50 36826.00 0.7176 1926 18413 .1046 0.8269
7 -7379.00 25138.00 0.7065 1160 12569 .0923 0.7296
8 -3994.10 16048.00 0.7511 689 8024 .0859 0.6788
9 -3323.80 9454.00 0.6484 369 4727 .0781 0.6171
10 -1659.00 4966.00 0.6659 196 2483 .0789 0.6240
11 -835.10 2376.00 0.6485 65 1188 .0547 0.4325
12 -536.20 924.00 0.4197 19 462 .0411 0.3251
13 -73.50 156.00 0.5288 3 78 .0385 0.3041
14 -66.00 66.00 0.0000 0 33 .0000 0.0000
15 -2.00 2.00 0.0000 0 1 .0000 0.0000
16 -2.00 2.00 0.0000 0 1 .0000 0.0000
17 -6.00 6.00 0.0000 0 3 .0000 0.0000
18 0.00 0.00 0.0000 0 0 .0000 0.0000
19 -2.00 2.00 0.0000 0 1 .0000 0.0000

This factor (WoBrill or fast workout time) is one of those that I refer to as a hidden positive. What I mean by this is that it is a factor the betting public historically tends to overlook as they bet the races. Believe it or not this factor alone allows you to improve your handicapping to the point of you being able to outperform most other handicappers at your track. And at the other end of the spectrum... look at the dismal results you'd get if you were to restrict your betting to horses that rank poorly vs their fields in this one factor. As a general rule, you can become a much better handicapper simply by avoiding horses with terrible WoBrill relative to their fields.



-jp

.

I agree that work times are important and that accuracy isn't an issue for myself.

I'm not surprised that your data says the same.

fffastt

[bellsbendboy]

I enjoy this forum, mostly because of the passion displayed when members defend their handicapping methods, patterns, systems etc. Many of you have spent a remarkable amount of time fine tuning your angle, and that's big time cool.

With that, lets get back to workouts. Assume horses are faster than they have ever been. They are. Accordingly, trainers point to certain races and workouts are a strong key, as to their intent. Largely time and frequency are secondary, if not irrellevant in the overall evaluation, and nefarious activities are few and far between.

Hopefully the maiden will work out for Prado/Badgett and provide some good material for us; Pure Tantrum, has pedigree, and apparently some talent; especially if you view his July 4th work, as noteworthy. It is.

We have mostly learned handicapping by trial and expensive error, and the only "sure thing" strategy we remember is; In a maiden race with exactly TWO first time starters and none of the experienced runners "look good", bet the one with the "worst" works. My recollection is that this opportunity surfaces about thirteen times each year and you can expect two or three winners. Some three weeks ago, there was a HUGE pick six carryover in California and the last race went to a $38 horse who fit this criteria.

Tomorrow the last at Saratoga is a maiden race and features two first time starters. Opinion welcomed! BBB



Any given race may have a zillion different variables

[Greyfox]

In a maiden race with exactly TWO first time starters and none of the experienced runners "look good", bet the one with the "worst" works. My recollection is that this opportunity surfaces about thirteen times each year and you can expect two or three winners.



Thirteen times a year? At one track? At several tracks?
Opportunity? No thanks I'll pass.