Most Significant handicapping factor

[arkansasman]

In building a handicapping model, the most significant factor I can find is last race Beyer. I am sure some of you programmers in conducting your research are finding far better handicapping factors than last race Beyer. So, would anyone like to comment on their most significant handicapping factor they have discovered.

[kitts]

I have read that the last race Beyer or the last race HDW speed ratings by Jim Cramer deliver about 30% winners and a negative ROI. And 30% is hardly "significnt" in this hugely random game, IMHO.

[Figman]

The most significant factor in getting to the probable winner at NYRA (Aqu-Bel-Sar) is the favorite as listed on the morning line. That said, the question arises, "do you want just winners or do you want to show a profit?" If profit is your motive, then other avenues must be explored.

[pmd62ndst]

First, allow me to preface that there is no single handicapping factor that will provide a positive ROI. Now...

By far, the most significant handicapping factor is the final odds. It is the most dynamic and consistant thing out there. Unfortunately, you can't always predict what it's going to be. In some cases, you won't truly know the final odds until the race is over. Whatever algorithm you use to make your selections, it needs to be dynamic enough to change with the toteboard and have enough room for error to survive late money changes.

PMD

[arkansasman]

I understand that the public sets the odds that are a true reflection of the probabilities of a horse winning. This is one factor that is built into my model of 84 handicapping factors. The last race Beyer is but 1 of those factors too. Using Pearson Correlation Coefficient, last race Beyer has the highest correlation of any factor at .295. Using multiple regression and normalized finish as the dependent variable, last race Beyer has a T-value of 21.5.

I was not suggesting that using only this factor you could make a profit. That would be far from making a profit. I was just wondering what other factors some of the other people here were finding to be the most significant.

[dav4463]

I don't think the single factor that points to winners can really be quantified. I believe it is the "class move". It doesn't necessarily mean the horse who has faced the toughest or ran the best vs. tougher company, but the horse that is the best fit for today's race. Trainer intent is the main factor I look at first. If the horse is in a bad spot, it probably will not win today. Sometimes the best move is a sharp horse going up in class, other times it is the horse on a decline dropping to a much easier spot. Other times the horse that has been getting close and stays at the same level has an advantage over those who are dropping but are off form. The hard part is identifying which horse is in the best spot today.

[sjk]

Last race best speed rating probably points to the highest win percentage.

Identifying the horses who can be expected to make a clear and easy lead (won't find one in every race) points to a higher ROI.

I would disagree with the statement that some things can't be quantified. I have quantified every factor that I consider important.

[Overlay]

In his 1994 study Modern Impact Values, Michael Nunamaker found the following impact values associated with a horse's rank in last race speed figure when compared to the other horses in today's race for his entire 12,815 race sample:

1 2.07
2 1.45
3 1.16
Front Half (but not in Top Three) .94
Rear Half .62

The single most effective overall speed criterion in the study, however, was an average of the figures from those races out of the horse's last three starts which were run on the same surface (dirt or turf, irrespective of condition) as today's race (or from all of the horse's last three races if none of them had been run on today's surface). Those values were as follows:

1 2.22
2 1.50
3 1.17
Front Half .93
Rear Half .57

Although I agree that no single handicapping factor should be made to stand alone, these values represented the clearest progression and greatest range of all the individual variables or measures associated with any of the performance areas that Nunamaker examined (unless there was something I overlooked). (I should add that Nunamaker himself treated Beyer figures and BRIS figures as interchangeable when applying these values.)

[Buckeye]

What's the ROI on last race Beyer?

[Overlay]

Nunamaker's $NET for the top last-race speed figure in today's field was $1.79.

[Buckeye]

Well I would say that -10.5 cents is going to be hard to beat with a single factor, but what do I know?

[Tom]

Depends on the race.....different situations propel different factors into the limelight. No one single factor is themost important in all kinds of races. You don't use the saem approach to a grass marthon graded stakes as you do to a maiden claimer at 4.5 furlongs at CT.
At least, I HOPE you don't!:D

[Overlay]

I should have added that Nunamaker's $NET for the top-ranked horse in average speed rating from those races out of the last three starts that were run on the same surface as today's race was $1.87.

[Brian Flewwelling]

Originally posted by Tom
You don't use the saem approach to a grass marthon graded stakes as you do to a maiden claimer at 4.5 furlongs at CT.
At least, I HOPE you don't!:D

Sure, bet the grey, or Florida bred, ... best if both :D

Brian

[formula_2002]

Originally posted by arkansasman
In building a handicapping model, the most significant factor I can find is last race Beyer. I am sure some of you programmers in conducting your research are finding far better handicapping factors than last race Beyer. So, would anyone like to comment on their most significant handicapping factor they have discovered.

Unfortunate for me, the most significant factor is the final odds.
Every group of speed, class, pace and form figure I have , ultimately correlates to the final odds.
It all seems to agree with much of the studies that I have seen posted here of late.

[Overlay]

I agree that the final odds set by the public are remarkably accurate when taken as an aggregate. However, because of the take and breakage, betting on all horses at any particular odds range produces a net loss (except, I believe, for extremely short-priced favorites at 1-2 or below). I think the key to realizing a dependable long-term profit is being able to assign an accurate fair-odds value for all horses in a race (not just the horse that is judged most likely to win), so that you can capitalize on those specific instances when the public has miscalculated, and has become oversold on a particular horse, driving its odds down below fair value, with the result that other legitimate contenders in the race are now held at odds which are higher than their true chance of winning.

[formula_2002]

Originally posted by Overlay
I agree that the final odds set by the public are remarkably accurate when taken as an aggregate.

...I think the key to realizing a dependable long-term profit is being able to assign an accurate fair-odds value for all horses in a race (not just the horse that is judged most likely to win), so that you can capitalize on those specific instances when the public has miscalculated

Ah! the miscalculation..

It's like picking out 33% of the your favorite cookies in a jar that has but 25% of your favorite cookies (time, after time, after time)
As a mater of fact you would have to do it thousands of times to prove that you were good enough to keep doing it.

[sjk]

Even after you have done it tens of thousands of times Formula will still say its impossible.

Snowed in here too.

[Overlay]

It's like picking out 33% of the your favorite cookies in a jar that has but 25% of your favorite cookies (time, after time, after time)

It seems to me that this analogy would apply more to a situation like betting on a casino game, where you're bucking adverse objective probabilities that never change, whereas in racing the odds are subjectively set by the betting public, whose opinion can be wrong. "All" you have to do is to outhandicap the other bettors (by which I mean being more effective at gauging a horse's true winning chances) by a sufficient margin to overcome the take and breakage. (I put "all" in quotes because I realize that this is no easy task.) Also, aren't there statistical tests that can be applied to a relatively small sample of races to prove that a particular outcome is not a fluke, but instead represents an independent variable that can be relied on to keep producing the same results over the long term, without the need to sort through thousands of races to replicate or validate the outcome?

[formula_2002]

Originally posted by Overlay

Also, aren't there statistical tests that can be applied to a relatively small sample of races to prove that a particular outcome is not a fluke, but instead represents an independent variable that can be relied on to keep producing the same results over the long term, without the need to sort through thousands of races to replicate or validate the outcome?

I think the most often used test is the one I have posted on my web page.

I have made a few inroads that show natural odds have similar standard deviations to pari-mutual odds.

If the std dev is similar between the two odds groups, it does not bode well for people who are looking to make a profit playing horses.


I define natural odds as follows;

when actual outcomes divided by expected outcomes =1(+ a very small amount)

[formula_2002]

Originally posted by sjk
Even after you have done it tens of thousands of times Formula will still say its impossible.

Snowed in here too.

I'll make it easy for you
win 40% of 200 plays at 2-1 (a 20% profit).

Do that twice and I'll make you rich!!!

[sjk]

Joe,

Thanks for offering to help. Doing fine on my own.

[andicap]

What about distance? He used sprints with routes??

Trouble with using last 3 races is it disregards the time element. You're lumping horses who have raced three times in last 90 days with horses who were

a) laid off
b) 2-3rd off a layoff so their 3rd race back could have been a year ago and be irrelevant to today's race.

Also with "best last race" surveys you lump in horses laid off a year with horses who raced 21 days ago. Distortions on top of distortions make results unreliable to me.


I'd rather see a DB trial by number of days back on level playing field.


So,
1) Max of 182 days
2) last 3 races todays distance structure/surface/fast only (tho I'll good wf or "good" if I have to.)

or even
1) max of 90 days or 120 days
2) same as above

Even within the same distance structure there are issues. Some say you can't use 1-turn 8f PPs for 9f two-turn races. Some 6f results won't translate to 7f. Etc.

But as a rough guide I would like to see the results of the above query compared to best last race speed figure -- providing the last race was within 45 days.

Of course it would be interesting to compare different methods. Do you get better results using my method than the one Nunamker outlined? Maybe not.
Maybe best of last 3 overall compared to best of last 3 within 6 months, same surf/dis, etc. does better.

I'd love to know.





Originally posted by Overlay
In his 1994 study Modern Impact Values, Michael Nunamaker found the following impact values associated with a horse's rank in last race speed figure when compared to the other horses in today's race for his entire 12,815 race sample:

1 2.07
2 1.45
3 1.16
Front Half (but not in Top Three) .94
Rear Half .62

The single most effective overall speed criterion in the study, however, was an average of the figures from those races out of the horse's last three starts which were run on the same surface (dirt or turf, irrespective of condition) as today's race (or from all of the horse's last three races if none of them had been run on today's surface). Those values were as follows:

1 2.22
2 1.50
3 1.17
Front Half .93
Rear Half .57

Although I agree that no single handicapping factor should be made to stand alone, these values represented the clearest progression and greatest range





of all the individual variables or measures associated with any of the performance areas that Nunamaker examined (unless there was something I overlooked). (I should add that Nunamaker himself treated Beyer figures and BRIS figures as interchangeable when applying these values.)

[sjk]

Hard to believe that as recently as 10 years ago a study of 12k races was the best available. You would hope that someone with a modern (much larger) database would take a look at some of these findings to see if they hold up.

[Overlay]

The impact-value figures I cited for average speed figure from those races of the last three which were run on the same surface as today were from the "all races" summary portion of Nunamaker's study. This general recap was followed in the study by eleven separate subcategories which broke the races down according to age, running surface, distance, and maiden vs. non-maiden. The effectiveness of this speed-figure measure (as indicated by the impact-value range from top to bottom) was higher in some categories than others, but every category showed a smooth value flow. (The overall strongest category was dirt routes for two-year-olds, and the weakest was non-maiden turf routes for three-year-olds and up.) I recognize the weakness of not including a time factor in the data, but couldn't this be compensated for by the appropriate weighting of other proven probabilities related to recency?

I would agree that ten years is an eternity in the context of data gathering, and that by now there must be much larger studies out there, and more sophisticated means of measuring speed. But doesn't racing also feature statistics which have remained unchanged for as long as anyone has been keeping records (such as favorites winning 1/3 of the time, and 2/3 of all races being won by one of the first three betting choices)? The problem would seem to be not so much that a factor loses its effectiveness at pointing out winners, but that it loses its pari-mutuel value through overuse. But as long as the handicapper takes care not to let any one factor dominate the analysis of a race, and instead uses multiple fundamental factors properly weighted, the methodology should have a fairly lengthy life span.

[sjk]

But as long as the handicapper takes care not to let any one factor dominate the analysis of a race, and instead uses multiple fundamental factors properly weighted, the methodology should have a fairly lengthy life span.

I've used the same basic program for about 8 years (with occasional modest efforts at improvement) and have seen no reduction in effectiveness. In fact this has been my best year.

It surprises me that the availability of data and computer horsepower has not had more of an impact on the dynamics of the game.

[Jeff P]

Joe,

Can you help me out a little? I'd like to know if there's a way to attach any statistical significance to the following set of plays. All of them are from this year. I haven't hit 200 yet but am getting pretty close.




UDM Definition:
Profile: v_DS_J1-HiRS-LowRV

Divisor = 999
DIRT SPRINTS (From Index File: F:\FALL_2004\pl_JRating_1_ytd.txt)


Data Summary Win Place Show
Mutuel Totals 439.10 347.40 303.00
Bet -316.00 -316.00 -316.00
Gain 123.10 31.40 -13.00

Wins 107 129 128
Plays 158 158 158
PCT .6772 .8165 .8101

ROI 1.3896 1.0994 0.9589
Avg Mut 4.10 2.69 2.37


By: 21 TheOdds

Min Max Gain Bet Roi Wins Plays Pct
0.00 0.00 0.00 0.00 0.0000 0 0 .0000
0.00 0.50 -3.60 54.00 0.9333 19 27 .7037
0.50 1.00 26.90 144.00 1.1868 51 72 .7083
1.00 1.50 -0.60 54.00 0.9889 13 27 .4815
1.50 2.00 33.80 30.00 2.1267 12 15 .8000
2.00 2.50 23.80 14.00 2.7000 6 7 .8571
2.50 3.00 13.80 8.00 2.7250 3 4 .7500
3.00 3.50 -2.00 2.00 0.0000 0 1 .0000
3.50 4.00 0.00 0.00 0.0000 0 0 .0000
4.00 4.50 8.80 2.00 5.4000 1 1 1.0000
4.50 5.00 7.40 4.00 2.8500 1 2 .5000
5.00 5.50 -2.00 2.00 0.0000 0 1 .0000
5.50 6.00 0.00 0.00 0.0000 0 0 .0000
6.00 6.50 0.00 0.00 0.0000 0 0 .0000
6.50 7.00 0.00 0.00 0.0000 0 0 .0000
7.00 7.50 0.00 0.00 0.0000 0 0 .0000
7.50 8.00 0.00 0.00 0.0000 0 0 .0000
8.00 8.50 16.80 2.00 9.4000 1 1 1.0000
8.50 9.00 0.00 0.00 0.0000 0 0 .0000
9.00 9999.00 0.00 0.00 0.0000 0 0 .0000

[sjk]

Jeff,

Easy enough to plug your win results into the formula on Joe's web page (click significance testing by the blue star); horses = 158, odds range = 1.05, win % =.6772. It gives a confidence level of 4.82 which is strongly positive.

[andicap]

Jeff,

You have to change your signature -- Blockbuster is eliminating late fees if you return if within 7 days of the due date.

:D :D

[Jeff P]

Really? :D :D

[formula_2002]

Originally posted by sjk
Jeff,

Easy enough to plug your win results into the formula on Joe's web page (click significance testing by the blue star); horses = 158, odds range = 1.05, win % =.6772. It gives a confidence level of 4.82 which is strongly positive.

JEFF-SJK
Do not do the above!!
Test by the smallest possible odds range.

I looked at the 72 plays, odds range .50 to 1.00

18.7% profit with a 71% win rate yields an average odds of .68-1.. That represnts a 1.99 confidence rating which is significant for 72 plays.

112% profit for the with a 80% win rate yields an average odds ot 1.65-1. That represents a 3.97 confidence rating which is highly significant fo 15 plays.


A few thoughts;
1. If these were back fitted results, do not get too excited.
If they were "LIVE" plays..cool.

2. If you can repeat the above results, I would think the FDA would approve your picks as a cure for poverty!!;)
If you can repeat the results, I would encourage you to get professioional help to examine the results.


Go slow, but by all means keep going.

Good Luck.

[Jeff P]

Life Cycle of a UDM

Here are a few thoughts, mostly my own way of doing things when it comes to practical hands on significance testing of horse racing data. Some of it is borrowed from what others that have dabbled in this type of thing before I did have suggested. Some of it is specific to JCapper, but the same approach, if done right, should work with any software, or even pencil and paper...

I break my data out into three sets:

1. Development Data
2. Validation Data
3. Live Data

Development Data
This accounts for about two thirds of all the race cards sitting on my hard drive at the point in time when the idea for a UDM first comes to me. I create the UDM by setting up the factors and ranges and then I test it to see what I have. I'll also tweak some of the factors and ranges and then re-test to see if I can improve upon what I have. This is the only place where it's okay for me to backfit stuff. I try not to do too much backfitting here. If the initial idea is sound then I shouldn't have to; but so long as I'm working with separate logistical factors I'm okay with a little. Once I have good results for a UDM against Development Data I move on to the next step: I Validate the UDM by confronting it with races it hasn't seen: Validation Data

Validation Data
This accounts for about one third of all the race cards sitting on my hard drive at the point in time when the UDM is created. This is where I simply test the UDM to see how it performs against fresh races. No backfitting allowed here. Period. I simply want the following questions answered: How does the UDM perform when confronted with fresh data? Does it "Validate" or doesn't it? If it does, then I move on to playing real money against live races. If it doesn't, then I drop it. Why keep it if it doesn't Validate?

Live Data
Here, I'm risking real money on today's races. From time to time, I will check the UDM to measure its live performance. Patterns do gradually change over time. The public catches on to whatever they were missing that the UDM was exploiting and prices get driven down, etc. As long as the UDM keeps performing well against Live Data, I keep playing.

The plays that I posted above are from a UDM that exploits some patterns I began noticing in late June 2004. It has held up reasonably well for me since I began using it in live play.

It's not a cure for my own form of poverty but it sure helps out. And by the way, you're not the first person to tell me to seek professional help!!! :D :D

[sjk]

Jeff,

What I like about your process is that you are able to test with a relatively large amount of data in a compressed amount of time. Then if all is well you move on to live betting.

The live betting is the key. Anyone who cannot get past the testing and into the betting needs to reevaluate their mindset or their general approach.

[pandy]

If the average of last 3 on same surface did better than last race Beyer or Bris, it would be interesting to see how best speed fig in last 3 did. Best of last 3 usually has higher ROI than best last race fig.

[Overlay]

Nunamaker's overall impact values and $NET for ranking the horses in a field by best speed figure out of their last three races (regardless of running surface) were as follows:

IV $NET

1 2.06 $1.80
2 1.54 $1.76
3 1.19 $1.65
Front Half 1.02 $1.63
Rear Half .58 $1.34

Smooth value flow, but (according to Nunamaker's data) average same-surface speed figure from last three races was more powerful in terms of impact values, and produced a higher $NET for the top-ranked horse ($1.87), although still resulting in a net loss, as might be expected with a single factor.

[Overlay]

I should have added in answer to Pandy's comment that (as I noted earlier in the thread) Nunamaker's $NET for the top-ranked horse in last-race speed figure alone was $1.79, so best speed figure from last three races outperformed best last-race figure by one cent.

[RXB]

If anybody tells you that the game can't be beaten, refer them to Monday's 7th race at FG.

When a horse stretching out from 6f to a route for the first time, by Gold Fever (a mediocre route sire) to a ZUPPARDO'S PRINCE (very weak routing) mare, with a body type that reflects its breeding to a tee, gets sent off at 3/5, you know that a large section of the gambling public doesn't have a clue in certain situations.

When the Beyers or last-race close-up finishes or obvious pace advantages tell the story, there's rarely any point in betting because the masses will be on the same horse(s) . But when those things are misleading, then there might be an opportunity. In this case, the public did what it will always do-- pounded the horse with the big double fig.

In the long run 3/5 shots are going to win half of the time. But we are dealing with individual races, not the long run, and each of us will sometimes be able to identify certain situations where the public, as a whole, is obviously wrong in its probability assessments.

[formula_2002]

If anybody tells you that the game can't be beaten, refer them to Monday's 7th race at FG.

In the long run 3/5 shots are going to win half of the time. But we are dealing with individual races, not the long run, and each of us will sometimes be able to identify certain situations where the public, as a whole, is obviously wrong in its probability assessments.

Post 200 such bets (3/5 shots) and win 70% ( a mere 12% profit) and I bow to your wisdom.

[RXB]

I don't take a laptop to the races, so I'm not going to be able to post when I think a favourite doesn't look the part on the track, am I?

[Buckeye]

What's the significance of 200 picks? I'm not trying to be flip (for once!), I'm just trying to learn something.

I may be all messed up and bloodied from yesterday, but as I'm sure you know, one day is no cause to for me to bow.

The main thing I think you're missing Formula, is the mistake of grouping all handicappers (including Database users) together. We ain't all alike! How the hell could we be?

[Buckeye]

One more thing,

don't you know there are handicappers above the second and even third standard deviation? :cool:

[Tote Master]

It always amazes me when I read comments like this. Why? Because there’s never any mention about the most important factor of all! Surprisingly after years and years of discussion, countless books written on the subject, and now even with sophisticated handicapping programs, the answer still seems to elude the majority of players. As generally predicted by the results of using all these things, the fastest animals in the best condition should naturally win every race. If so, then why aren’t all of the most astute handicappers wealthy? Why? Simply because all of these projections are based on a single premise: That every horse entered in a race is really trying to win. Unfortunately, this premise is false. Once you remove it from the argument, the forecasts made that are based solely on using all of the past performances begin to lose credibility. Its not because they’re not accurate, it’s because they leave out the most important factor of all: The Intention to Win. This may seem like a pretty ambiguous statement, because how could anyone possibly measure “intent”?

Well, as I see it and have expressed many times before, the only way to measure it is by thoroughly understanding the value of “how much and how little” and “where and when” the money is being placed on each of the entries. Unless you have direct insider information, this is the only way of knowing (and measuring) true intentions. Why? Simply because it not only takes into consideration all of the other handicapping factors and the conditions of the race, but also includes the added bonus of knowing that the money is certainly of value to those making those large and serious wagers. No matter how you attack this game, this premise is always true. If you can accept that, then all of the other predictions made through incisive handicapping techniques suddenly take on a new and exciting dimension. So much so, that even insider information on its own takes a back seat to seeing the actual money being bet.

As far as I’m concerned there’s no such thing as a “bounce” theory at all when it comes to 90% of all race horses. Sure these animals have their limitations and believe it or not their connections surely know it first hand. All you’re really seeing though is their lack of intention. When you understand and appreciate the money end of the game, you can actually predict the so-called bounce: Simply by knowing they’re just not trying today.

Best of Luck!

[Buckeye]

Here we go again,

"somebody" knows something.

Not only that, they'll tell us what they know!

And somebody else will sell it. :rolleyes:

From the goodness of their heart no doubt.

[hurrikane]

yeah,
I'm waiting for the guy on his deathbed to show up with the answer.

so, the majority of races are won by horses within 1 length of the lead at the first call.

early speed use to be an almost break even bet. The ROI has dropped but the win % is about the same. Public is catching on.

With this I would say predicting the early speed is the most important handicapping factor for making a profit.

Now, if you don't want to make money but want to pic winners..joe is probably the most important handicapping statistic.

good luck all

[andicap]

What a load of crapola.

This post is just an ad for whatever tote software this guy is selling.

Why aren't horseplayers wealthy? Ask Len Friedman.
Ask Richie Schwartz.
Yeah, they're broke. Last I heard Dick Schmidt was out selling apples. Right.
How do you know there aren't wealthy horseplayers? I doubt the ones who make a lot of money -- not the "experts" who write all the books -- are out there screaming about it. Most pros live quietly, don't talk about their profession and keep a low profile. And define wealthy. I would kill to make $100,000 a year at the races --and I bet there are people doing that.

It's that kind of thinking that gives horse racing a bad name and keeps it suspicious in the eyes of the public. "You can't win because it's fixed."

If you're ROI is 15% a year -- an excellent result -- how much would you have to put through the windows to make $1 million? About $7 million. That's a lot of change to bet for anyone.
Psychologically no one can play every day or even every week. It wears you down, especially betting that much money. If you played 200 days a year, that would mean betting $35,000 a day! Who could keep a straight head betting that much??? Not to mention affecting some of the pools.

Of course trainer intent is important. Trainers treat some races as preps or conditioning races. Horses go in and out of form and can't be razor sharp every race. NBA players don't play full speed EVERY play -- they couldn't or they would be exhausted by playoff time when it really counts. And yes, some cheating occurs. But not every race or even in almost every race.

And to say the bounce is mostly a myth -- well, everyone's entitled to their own opinion, but if you're reading this and trying to learn the game, that is a very dangerous opinion to hold. You will lose a lot of bets believing horses don't bounce, that they lose because trainers want them to after running a hard race.

TM raises a questoin: how come the fastest, most conditioned horse doesn't always win? He argues it's not because of the PPs because they are "accurate."
Huh? The PPs are a good approximation of how fast a horse is going, but certainly there is plenty of margin for error. The chart-callers make mistakes. The variant makers err. The distance and track to track adjustments are estimates -- good ones -- but estimates nonetheless. I'm amazed they work as well as they do.

The fastest, sharpest horse doesn't always win because of
a) racing luck/track bias/bad ride/
b) he isn't the fastest sharpest horse after all. It's in the interpretation of his form or the numbers could be wrong.
c) He's not feeling well that day or the surface was too hard/soft/etc.
d) form cycles -- horses go in and out of form quickly
e) bad pace scenario. Horse A may be fastest and sharp but if Horse B and Horse C make him go too fast to the 4f pole, Horse A won't win.
f) Yes, trainer intent is important, but not all-important. Trainer might be darkening form.

Am I arguing that you shouldn't watch the tote for clues? No, just that you don't have to in able to win. Sure, I wonder what's going on when a 10-1 shot with mediocore figures is bet to 2-1. And when a 3-1 ML horse with the best figs is 6-1 until 5 minutes to post and then the crowd bets it then to 4-1.
But I know lots of people who win without even considering those factors.
Would be win more if you were an expert at watching the tote -- yes. But You would have to know how individual barns operated. Which ones bet and when? Early? Late? For certain owners? Is a certain jockey aboard a "go" signal? Morning lines are often unreliable.

I expect to become a consistent winner at the races this year -- I have a coherent betting strategy for the first time and keeping better records. But I will never be rich betting the races for one reason: Psychologically I couldn't bet enough money to win enough. Even if I was $50,000 ahead, something would stop me from betting $1,500 on a race.

That's a major reason why many excellent handicappers don't get very wealthy playing the game. That and lack of discipline. I know very good horseplayers who would have excellent ROIs if they played their best 3-5 bets of the day. But they need action and play too many races. Turns them from the black into the red.










It always amazes me when I read comments like this. Why? Because there’s never any mention about the most important factor of all! Surprisingly after years and years of discussion, countless books written on the subject, and now even with sophisticated handicapping programs, the answer still seems to elude the majority of players. As generally predicted by the results of using all these things, the fastest animals in the best condition should naturally win every race. If so, then why aren’t all of the most astute handicappers wealthy? Why? Simply because all of these projections are based on a single premise: That every horse entered in a race is really trying to win. Unfortunately, this premise is false. Once you remove it from the argument, the forecasts made that are based solely on using all of the past performances begin to lose credibility. Its not because they’re not accurate, it’s because they leave out the most important factor of all: The Intention to Win. This may seem like a pretty ambiguous statement, because how could anyone possibly measure “intent”?

Well, as I see it and have expressed many times before, the only way to measure it is by thoroughly understanding the value of “how much and how little” and “where and when” the money is being placed on each of the entries. Unless you have direct insider information, this is the only way of knowing (and measuring) true intentions. Why? Simply because it not only takes into consideration all of the other handicapping factors and the conditions of the race, but also includes the added bonus of knowing that the money is certainly of value to those making those large and serious wagers. No matter how you attack this game, this premise is always true. If you can accept that, then all of the other predictions made through incisive handicapping techniques suddenly take on a new and exciting dimension. So much so, that even insider information on its own takes a back seat to seeing the actual money being bet.

As far as I’m concerned there’s no such thing as a “bounce” theory at all when it comes to 90% of all race horses. Sure these animals have their limitations and believe it or not their connections surely know it first hand. All you’re really seeing though is their lack of intention. When you understand and appreciate the money end of the game, you can actually predict the so-called bounce: Simply by knowing they’re just not trying today.

Best of Luck!

[RXB]

Excellent post, andi.

[Tote Master]

Sorry to burst your bubble guys, but I’m just telling it like it really is! I’m not trying to sell anything at all. I have better things to do with my time. In fact, I could really care less what the skeptics might think. There just might be some players out there who are continually frustrated with this game. They pour over all the information at their disposal, use the best handicapping techniques, and still wonder why an unknown commodity beats the top selections to the wire. This is not a rare occurrence, it happens all the time. In fact, since the public’s 1st and 2nd choice combined wins approximately 52% of the time, then about 48% of the time something else is happening. If you can attribute that to a so-called “margin of error” then obviously a lot of people must be making the same mistakes? Or perhaps they’re missing something?

You can try to justify why the fastest horses in the best condition lose any way you like, but until you realize that this only a game being manipulated by those in control, you’ll always cling to the idea that the focus should only be on the horses and not the money. Who said anything about being fixed? It’s all just part of the game. I can’t help it if the majority of players ignore what some consider being the most important factor. You call it a “dangerous” opinion? I call it a powerful handicapping tool.

All you have to do is look at the amount of money being wagered in ALL the betting pools in a typical race at any reasonably sized track. If you think for a minute that it’s all being generated by just the $2, $10, or $20 bettors you’re mistaken. Your interpretation of how most view the tote board is a typical scenario. Unfortunately, when you’re looking at just the odds, all you’re seeing is the results of the monies being wagered in the win pool only. That’s just a small portion of what’s actually being bet on each race. Looking at the money means ALL the money! As I mentioned it also includes “how much or how little” and “where and when” the money flows into the betting pools. You really don’t need to know anything about the racing operations, or the connections involved. They all use the same color money, but how its being used is the key.

I can’t argue your points about money management. They’re all valid. Discipline is certainly important to not only selecting the best races to play, but to get the most from your winning wagers. It’s funny though how that word “money” keeps coming up.

I will question your insight into knowing just what the pros are doing and how they live. I also question your knowledge about all the psychological aspects of playing and betting. We all have our own budgets, betting preferences and selection procedures. Mine simply takes in one factor that I believe is the most important and involves a thorough knowledge of the tote board activity. Lets put it this way, if they don’t bet on it, neither will I.

Best of Luck

[46zilzal]

toteborad is NOT the factor it once once since BIG pulses come in from OFF track sources...makes it ver confusing.

[46zilzal]

As far as I’m concerned there’s no such thing as a “bounce” theory at all when it comes to 90% of all race horses. .


tell that to THREE YEAR OLD FILLIES!!

[hurrikane]

yep cj. you are right.

I'm going to do the trial and will give you an update but I don't see any added value on the surface.

[Pace Cap'n]

Does anyone know when exactly the simulcast totals are transmitted to the host track? I notice 46zilzal referred to "pulses", but I seem to recall reading that only one transmission is made.

From watching the pools it would appear that not until the last flash (which we don't see) do the outlying wagers hit the board. How would it be possible to analyze the tote if the majority of wagers have yet to register?

Also, is it possible to analyze any pools other than WPS & exacta? Or necessary?

I have caught the odd horse from noting tote action, which made me watch more closely, but soon decided that such plays are few and far between.

Lastly, I have been known to place all my bets for one card well before the first race has been run. Probably not a prevalent betting scheme, but how many such wagers would it take to skew the tote patterns?

[Tote Master]

I guess there’s nothing like a bit of controversy to stir things up!
(Especially in this game where opinions are always item of the day.)

To 46zilzal: I reply, don’t kid yourself or down-play the power of the tote board today. You’re right the Big pulses come from remote sources, but they only appear near Post Time. In fact, Late money in general is only coming from people trying to jump on the bandwagon. This money has nothing to do with “intentions”. Do you think for a minute that the serious money from the connections is coming from an OTB parlor or an on-line account?
Re: The Bounce theory, you pointed out a single grouping of animals (3-yr fillies). Didn’t I say 90% of all race horses? You could certainly include your group in the other 10%.

To hurricane: I say, Welcome, but please keep your agreement with T-W in mind. It is NOT my intention here to promote or solicit any service. I am simply stating an opinion based upon my knowledge, experience and the results of my game. Whether anyone wants to believe what I’m saying or not makes no difference to me at all. I will simply say that I will never stop using the most powerful weapon that this game has to offer. Can I be any more explicit then that?

To Pace Cap’n: I say that your memory servers you right. There are only a couple of late Pulses transmitted to the host track near Post Time. This money, as much or as little as it might be has only a negative impact on a tote analysis. It certainly doesn’t come from any host track connections that I’m aware of. I have found that it’s really not necessary to include this money in the final analysis. Besides even if it were significant, it would come in too late to make a decision for a wager.......It is certainly possible to analyze any pool if you have the data and the correct formulas, but the mutual, exacta and quinella pools are more then adequate.......I’m not sure what you mean by “tote action”. If it’s the rudimentary “viewing of just the odds” you’re referring to, then I would agree with your findings.......In terms of “when” bets are being placed, I would have to say that the “early money” certainly has the most impact. It isn’t a question of how many early bets are made, its how much the bets are and where the money is going. The whole idea behind any form of tote analysis is to be able repeatedly recognize any skew or bias. Without repeatability there’s loss of confidence in any system, and when that occurs whats the point?

By the way, thanks to those of who have written. I knew I couldn’t be the only Watcher out there!

Best of Luck!

[hurrikane]

I can tell you one thing that keeps me away from the tote.

I have known serveral owners that bet the hell out of horse for no other reason than they own it. Stupid as it might seem..they do it. Over and over and over.

[Hosshead]

I can tell you one thing that keeps me away from the tote.

I have known serveral owners that bet the hell out of horse for no other reason than they own it. Stupid as it might seem..they do it. Over and over and over.This is because many owners feel bad if they DON'T bet (support) their own horse (whether he has a chance or not). And when you've got more than one of these owners in the same race, what does that mean? The owner with the deepest pockets has the best horse?

One time, years ago, when they had "large transaction" mutuel windows, I was in line at DMR. And was betting against the fav. who was a speed horse who had just broken his maiden and was now running against some REAL speed. I won my bet. But the interesting thing was that the "hot" (grey) fav. was owned by Jack Klugman (who I really like). Well guess who was in front of me in line? You guessed it, Jack ! And I couldn't help but see over his shoulder at the window. (he's short/i'm tall). Anyway, man did he bet his horse. Even gave the teller a $70 tip for taking his bet. I wanted to say "Please Jack don't waste your money on this race". Anyway his horse was done at the 1/4pole. And I could see him (after the race) sitting in his box with a real sad look on his face. Big hopes (and dreams) for this horse, gone out the window.

However, I must say that I'd rather see my horses being bet down, than the other way, when the connections don't even want to bet it.

[jfdinneen]

I guess there’s nothing like a bit of controversy to stir things up!
(Especially in this game where opinions are always item of the day...)



ToteMaster,


What you have identified is analogous to using a technical approach to investing in the stock market (monitoring the action - weight of money flows - associated with a particular company's share price) and what many of the other contributers are focusing on is similar to using a fundamental approach to selecting shares (measuring a company's performance relative to its competitors).

You will agree that these two approaches (in the extreme) are uneasy bedfellows - hence the frustration expressed by your detractors. In fairness to them, the word handicapping is traditionally associated with the fundamental school and betting with the technical school. The former focuses on winner selection and the latter on making money. Each school is skeptical of the other's approach and tends to diminish its particular insights to the game.



Best wishes,


John

[andicap]

If the average of last 3 on same surface did better than last race Beyer or Bris, it would be interesting to see how best speed fig in last 3 did. Best of last 3 usually has higher ROI than best last race fig.

Good to see you around these parts, Pandy. Hope your head cold is feeling better. BTW, a guy on this board, Joe G. does some pretty good harness figures. He posted Meadowlands Friday night on the harness board.

I too look at on best of last 3 to get my contenders, but only the last 3 at same distance structure and surface AND only if within 6 months.
Trouble with all those databased queries on best speed methods is they will use the last race or best of last 3 even if some of those races took place two years ago! They'll compare turf to dirt, and sprints to distance races with impunity.

I will look at best last race speed rating to see if I've overlooked someone who might have the best recent form/speed as well as for form cycle considerations.

[Tote Master]

To jfdinneen:

Well, since the “Value of Money” is the bottom line, I suppose you could make that type of comparison. Its funny that you chose that particular correlation, because my tote analysis looks very similar to a stock market report. You’re certainly right when it comes to my braking with tradition. I used the traditional approach from 1970 to 2000. Like most decent players, I had my ups and downs fighting this game. I’ve probably forgotten more about those methods than some on these forums even know about the game. Thank goodness, because I’ll never resort to those painstaking methods again! Besides, my time is more valuable and better spent then by pouring over all those variables found from dissecting the PP’s.

The problem with pure handicapping is that you can get too caught up with the animals themselves by constantly trying to determine how they fit the conditions of the race at hand. Many players lose track of why they’re even handicapping to begin with. If it’s only an ego trip to prove how many winners can be selected, then they’ve already lost. The goal should not be “how many winners”, but rather ”how much money is won” after playing so many races. Once you start to realize that those who put on the show use these horses as a means to an end, then the whole game takes on a different perspective. Remember, they’re the ones who first and foremost place their commodities into the arena. They have first hand information about every aspect of their charge, and always keep in mind, their bottom line is money too!

As far as I’m concerned, when you play a game that’s driven by money you’d better have a good understanding of that facet too. So, I’ll just stick with the money making approach. Thank you. If it’s good enough for the connections, then its certainly good enough for me.

Best of Luck

[Light]

Hosshead Quote:
However, I must say that I'd rather see my horses being bet down, than the other way, when the connections don't even want to bet it.

I have had a similar experience to your Klugman story but in an opposite way. Horse goes off at 20-1. I bet $10 to win on him and box him with the favorite. He wins,but there is an objection by the favorite who finishes 2nd. My horse gets taken down and I get a puny exacta as they put the favorite up. Costs me over $200.As this was the last race of the day,I walk out to the parking lot,pissed off. To vent, I tell the stranger walking next to me "can you believe they took him down". He says"tell me about it,I'm the owner". I reply,"so you had a big bet on him?"He says"not a dime,the trainer didn't give me any clue he was live today".I stopped dead in my tracks and let him walk away.

I was dumbfounded.I realized that owners and trainers don't allways know their own horses. Up till then I allways thought they did.

My point is that in Hosshead's story the owner doesn't know his horse enough and foolishly bets him,overestimating. In my case the owner doesn't know his horse and doesn't bet him:underestimating him. So if a horse does not get action on the board and you like him,just remember there are 2 sides to every coin.

[Tote Master]

Re: Hosshead’s and Light’s stories

I believe that both these tales of the owners not knowing what’s going on, while amusing are not really the norm. I would think based upon what I’ve seen over thousands of races, that most of the time the connections certainly have all the information they need as to whether to play or not. I could also point out a few instances when a long shot won a race and there was no indication of intent whatsoever. Fortunately these occurrences are rare, but they can happen. I’ve taken notice that this can occur particularly on off tracks, when even the connections (and the betting public) aren’t aware of how well the horses might perform. They may not be betting at all, yet their horse wins in a romp at a big price.

I’ll continue to use the side of the coin that shows intent and pass on the others. I value and respect their money as much as they do.

Best of Luck

[Hosshead]

Hosshead Quote:
However, I must say that I'd rather see my horses being bet down, than the other way, when the connections don't even want to bet it.

I reply,"so you had a big bet on him?"He says"not a dime,the trainer didn't give me any clue he was live today"..Light: Even though your horse was 20/1, I wonder if the trainer had a bet on him? Maybe the trainer didn't want the owner to drop the odds? I wonder how often this happens? When the trainer knows something and doesn't tell anyone, not even the owner. And then how long does he remain as the trainer, for not telling the owner the horse was "ready". But then again, as an owner, how could you fire a trainer after your horse just won? catch 22

[sjk]

There is probably only one better way to lose a lot of money than to own race horses and that is to bet your horses every time the trainer thinks they're going to win.