Why don't favorites win more often?

[tanda]

I believe that favorites are continuing to win at a 32-33% clip. This remains unchanged from histroical levels. Why? And what handicapping implications does the answer have?

The following are reasons why winning favorites should have increased in the last 10-15 years:

1) Lotteries and spreading casinos have peeled away the "numbers" players. Thus, the remaining players are, on average, better skilled then in the past.

2) Proliferation of books on handicapping tending to increase handicaaping skill.

3) Ditto regarding software.

4) Ditto regarding information contained in past performances (mainly pace and speed figures, pedigree and trainer data).

5) Shorter fields which give every horse a better chance to win.

Based on this, shouldn't the "crowd" be handicapping better. Yet, the crowd is not, judged by winning favorites and its 2nd choices, etc.

What are the implications?

1) These improvements in data/tools have not actually increased handicapping ability, or

2) These improvements in data/tools have increased handicapping ability.

Duh, I am sure you are saying. But read those statements again.

If the crowd has improved its handicapping skills, but cannot improve its win percentage, then the implication is that improved handicapping does not necessarily lead to improved results. How can that be so?

Or, if the improved data/tools do not lead to increased handicapping ability, then how good are they? And of what use are further improvements?

Maybe it means that no matter how good your handicapping is, the randomness of horse racing is such that no methodology can yield more than 33% winners if every race is played. For example, let's assume that there is a "best horse" in each race, taking into account all facets: pace, speed, form, class, trainer, jockey, weight, post position, distance, surface, track bias, etc. Maybe even if you were able to identify that horse, you would only win 33% of your races. Maybe there is a ceiling.

For those of you who believe not, then how do you explain the public's inability to improve its results (not just in the area of the favorite, but 2nd choice, etc.)?

[Dave Schwartz]

Tanda,

The win percentage on the favorite in a given type of race is based more upon the average odds than anything else.

The average odds on the favorites are falling (as evidenced by the loss on favorites going up). To me this indicates that the public is in greater agreement about who the best horses in the race are.

In other words, instead of the win pct going up, the mutuels are coming down.


Regards,
Dave Schwartz

[tanda]

Dave,

So the public is doing worse. Same win percentage but lower mutuels. Again, why not better?

For example, in the past the crowd won 33% in a 9 horse field at 9:5. Now they are winning 33% in an eight horse field at 7:5. You would think that with shorter fields and lower odds, the hit percentage would go up. But it has not.

Another way of stating it. Would you agree that the crowd is smarter than in the past? I think the crowd is. Why not better win percentage?

TThe crowd was picking 33% in a nine horse field without speed figures, pace figures, pedigree and trainer data. Now with all those things and shorter fields there is no improvement. How useful are those things? I think they are useful, but it is curious.

For example, you would think those things would bump the percentage up to 35 or 36%, but no.

I do not have any ideas, I just find it interesting that the rate is consistent (at least as far as I know, any body know different?).

[BIG HIT]

I think one reason is we as handicappers still have same old feeling about horseracing people would rather think they were cheated than they picked the wrong horse.Yes there is more info but that only brings more decision as to which way to go what is the best way to handicap a race.All this info just like a chaos race in it self.In a way you could make a case for handicaping being harder now then before given all the info books and software on the market now.

[GR1@HTR]

Why? Me thinks because the ultimate goal of the majority of handicappers is not to pick the winner...But to pick the best play...ie value...From what I understand, horseplayers today are more likely to "pound" the favorite regardless of rationale than compared to years ago...

[Rick Ransom]

tanda,

A very interesting question and one which I have wondered about myself many times. I think Dave has it right. Favorites used to lose only about 10%, now it's more like 20%. Why? I think it is because there is more information available but most of the additional information is redundant.

If I like a horse because of speed, and you like a horse because of pace, and another guy likes a horse because of class, most of the time it will be the same horse and when it is the odds will plummet. When you see a situation like this, you should look for another horse that is only good in one of the factors. That horse will win an only slightly smaller percentage of the time but will return much higher odds.

[Rick Ransom]

While I'm at it, I might as well irritate the Sartin disciples. Here goes. I think if your win percentage on your top horse (betting every race) goes over about 30%, you should probably look for ways to lower it because you're hurting your ROI by using dependent factors. I know some of you will disagree, but all I can go by is my experience, and I've never seen anything above that level that didn't hurt ROI in more than 20 years. But that's just my opinion. I could be wrong.

[Aussieplayer]

Not sure how relevant it is to the actual question you're asking, but do remember that the win percent of favourites IS different - in different "types" of races.

I have found favs. in some distance/class categories with a low percentage such as 22%, and other categories with 38% for instance.

But yes, even here (and the UK as far as I am aware), the "overall" favourite percent is right on 33% (using last 10 years data).

Which is interesting - as the average Aussie field size would be a few extra horses than the overall USA favourite percentage!

Cheers
Aussieplayer

[so.cal.fan]

Aussie Player is right. Every type of race has a different rate of favorites.
Today's Racing Digest in So. Calif and No. Calif. gives the favorite percentage for each different type of race.
They range from a low 15% to higher than 50%.
It is very interesting, the question Tanda raises- why?
Is much of our information wrong?

[hurrikane]

An idea....perhaps 60% of the races truely have no horse that is the favorite. They are just too contentious to claim one horse as the fav. However...one of the horses goes off at 2-1 and the other 5-2. the 5-2 wins...so the fav lost. but how much difference was there really between the two...not much I think. many races you have one horse at 2-1 and 3 horses at 5-2 and the others at 3-1. This happens a lot in a short field.

Is there really a big enough difference in these horses to state that one shold be the favorite over the others.
Probably not..but still one is considered the favorite.

Perhaps the calculation should be how many horses win when the spread on the odds between #1 fav and the #2 fav exceeds say......2 pts. That would give you a more meaningful result. IMHO

IMO the real answer is in the PP data itself.

As far as I know nothing has changed in the way PP data is collected during a race for years. I believe that the timing at the calls of a race are all set based on estimates of time from the frist horse that hits the marker and the spotters judgemt of lengths behind the leader.

All PPs, computer data bases and programs are based on this data. So it should not be a big surprise that the public and any computer program is going to come up with (more or less) the same list of contenders which will result in a favorite and the other low odds runners.

What this says to me is that 30 to 33% win rate is as accurate you can get with using the data collected and presented for handicapping. It doesn't really matter how good you are as a handicapper or how good you are as a computer programmer you can only get to a a certain level of accurracy with the data you are using. You know garbage in garbage out

I'm not a sheets player but this is the only method I know of that uses a different method of timing and at least hs the potential to come up with something different then the masses. (different but not necessarly better).

This dosen't even address the question of how much of the outcome of a race is dependent on PP's and how much is based on factors that can't be handicapped at all. In general I think handicappers put way to much importance on the numbers and have expecations that runners will run to the so called "form" much more then then they normally will.

[tilson]

Just my 2 cents......1st of all i think it is a misconception that there are "fewer players today"......pari mutuel handles are way up from where they were 20 years ago and that's even when you adjust for inflation.
Just because these players are no longer actually AT THE TRACK doesn't mean there are less players.
Secondly i agree with Dave Schwartz.... while the win of the fav might still be 33% the average pay out is a bit lower now on that 33% then it was 20 years ago.
A difference of only 3 or 4% in the percent of favs winning times their average pay out actually represents a rather large shift of where money is going and HOW it is being bet.

[Rick Ransom]

Here's another idea. The favorite is determined by the amount of money bet, not the number of people betting on it. So it is mostly determined by the large bettors. What if the large bettors don't have better information than the small bettors these days. This would also explain why toteboard systems don't work as well as they used to. Big money is not necessarily smart money anymore.

[GR1@HTR]

Another Theory:

http://www.jeanie.com/horsfaqs.html

[Rick Ransom]

GR1,

That's more amusing than Mickey Rooney's dice!


so.cal.fan,

Do you use Today's Racing Digest frequently? If so, what information, if any, have you found to be useful?

[so.cal.fan]

This is one of the more interesting threads I have read since I have been on the internet.
I too, have wondered for years why favorites have won at approximately the same rate for over 100 years, in spite of all the so called "advanced technology".
There must be something that is so obvious, everyone misses the point.
I suspect it has to do with class and condition, but we are probably all trying to make it too complex.
Rick Ransom,
I use Today's Racing Digest during the Hollywood Park meet, because I get a comp.
They do a good job of reporting So. Calif. racing information and their website-www.todaysracingdigest.com has some great free stuff. Namely the week in review, which gives you a lot of information you will not find in the form on trips, trainers, how strong a field it was, etc. They also give some interesting charts on track profiles, classification, winning favorites.
I should buy one everyday during Del Mar, but I am too cheap.
I'm sure it costs me, I looked at someone's copy yesterday at Santa Anita, after I had bet a loser. Had I have read the Digest. I would not have made the bet. I guess that was my Karma, for getting all those free copies, and then being too cheap to buy one! LOL

[tanda]

SoCal Fan,

Their Week In Review is available for free at their website, so you can be cheap and still save yourself from betting on losers. Like I do.

[Tom]

I agree that this is a state of the are website for racing info. The week i review is priceless, and downloading the older issues still gives you the track variants and other info.
The site is a model for anyone with the ambition to build a good racing site.
Tom

[so.cal.fan]

A guy I know at Santa Anita said he thinks the reason favorites don't win at a higher clip is that they change the track all the time.
I don't quite buy that, but I am not sure. I don't know what they did with the tracks years and years ago, he says they left them alone. It seems to me, with the better equipment we have, they keep the tracks more uniform, especially in rainy weather.
I wonder if certain handicapping methods get "trendy" from time to time and start dominating the pools?
I have long believed that one day the pace handicappers will win, the next day, the sheets guys, the next the class guys, and sometimes the "horoscope" guys (but not as often).
I think any method that is logical will produce approximately the same amount of winners over time. The trick is to get onto what the public is not, and at the same rate of winners. Not so easy to do, but we are all tying.

[Topcat]

Tanda,

i think it is an excellent question with implications for racing and handicappers. I have posed a variation of your question to many handicapping luminaries including Steven Crist, Barry Meadow, Dick Mitchell etc.

The question I ask is why 32-33%. Why not 12.5% which is about random picking from today's current field size or why not twice random whiich is 25% why not 50%?

Yes I know that over certain types of races and certain meets that the favorite win % varies but the constancy of favorites over any large sample winning 32% is amazing. Only coin tosses approach this sort of lack of deviation. If it was a coin toss our answer would be easier to find. We would expect a favorite winning % of about 50%. So what is the favorite win % of 32% correlating to?

Maybe the answer lies in our coin toss or in match races. Take a look at the the top two win %=it is a little over 50% or about 52% the 2nd ranked favorite % s are alwo remarkeable constant over any large sample.

One of the answers or hint of an answer that I favor suggest that each race is essentailly a match race of sorts. The two most likely wnners versus the rest of the field.

While this doesn't answer the question(s) it does suggest that the answer may lie within the carding of the race.


Even still grandpa in 1930 was still doing as well as we are today.

[Rick Ransom]

so.cal.fan,

That guy you referred to may not be right about favorites, but I'm sure he's right about the trend in track maintenance in general. At Santa Anita and Del Mar, the track superintendant is Steve Woods, and he is well known for eliminating track bias. The guy is an expert at it and we'd all (especially you Sartin disciples) better get used to it. If any of this has changed over the last couple of years, let me know.

[4thandlong]

I tend to agree with hurrikane on this one. How many races have legit favs vs. vulnerable and false favs? We see races every day where no horse deserves favoritism but one has to be by default.

Has the number of formless fields and races with no legit chalk changed over the years? I suspect not. Combine that with racing luck and the fact that horses aren't machines and maybe there's your answer.

[so.cal.fan]

My son's theory:
The quality of horses has improved, they breed more horses,
and that the competition is much more than it was years ago, and so that balances out the favorite percentage.
I recall a friend, years ago, who used to attend Santa Anita in the 1930's 40's and 50's.
He said that back in the 30's only a few big stables won all the races.
My husbands theory is that for the past hundred years, that only 50% of horses in races have any chance at all to win, that out of those you can boil it down to 2 or 3 and there is your percentage that has remained stable over the years.
My theory is that we have too much information and it confusing the hell out of us! Before everyone laughs, I know people who win consistently using very simplistic, yet logical methods.
On the other hand, I know people who win consistently using very complex methods.
I guess the answer to this interesting question has a question mark at the end of it??????????????????????????????????????

[Rick Ransom]

so.cal.fan,

I think complex methods can only win if they use a lot of obscure factors. If you combine a lot of common factors, you usually wind up with an underlay. I use only four factors, and one of them is a negative factor that improves my ROI. The funny thing is that I've checked a lot of things that are positive for all horses but don't improve the results for my top selections. So, when somebody makes broad generalizations based on studies of single factors, be careful about using it in your handicapping. It might be positive for one person, but negative for another.

Back to the favorite question. How about this: It's a complete coincidence based on the fact that whatever has happened to raise the % has been balanced out by something that lowered it. I guess that's generally what your son is saying, but it could be any combination of things.

[Tom]

Maybe it's just that the favorite is the consensus selecton of everyone. Since all the new information is available to everyone, everyone uses it the same way everyone used the old paradigm of information-a few win, most lose. Probably what changes is who the winning group is fro time to time.
As a group, everyone wins 1 in 3 and losses 18% of their bankroll. This meas that there is a large group oout there that wins less than taht and loses more than that. So for some, just betting the favorite every race would be an big improvement in their handicapping.
Tom

[so.cal.fan]

Since I am a So. Cal. fan, I don't know anything about other circuits, but here big stables, big name trainers win a very, very high percentage of the better class races.
I play only claiming races, but if I was specializing in the better races and the straight maidens, I would follow the better barns-PERIOD.
It seems what has changed here, is there are more of the better barns than in past decades. You have about 6 big names that you can't throw out of these types of races. Many times there are 4 or even 5 of them in the SAME race. This has got to balance out those percentages, at least to some degree.
It would be fascinating to have a study of the favorite percentages for the different types of races over a hundred year span!
It might just give us an answer.
When the purse money in big, the big names usually go after it.
An example is Pomona Fair meet. When the purses got big enough in their stakes races, the big name trainers ran their second string stakes turf horses in them, and won them all.

[Rick Ransom]

so.cal.fan,

I've seen entire meets in Southern California where the top trainer showed a decent win % and profit. Of course, it was usually followed by a meet where the same strategey lost about the usual 20%. Go figure!

[so.cal.fan]

Lots of stables go real hot for one meet, then cool off.
It sort of makes sense. Even the guys who can afford all the best of "everything", can't keep them going that long without a rest.
Unless a guy claims or gets in a lot of new stock, he is usually going to cool off.
Also, more often than not, barns get sickness or rashes going around the barn, and that sets everyone back. It happens more than people realize. That is why I don't like to see big gaps in the work patterns or see horses in the paddock with scars of skin rash on them. Again, you see it more often than you would think.
You hit on another good reason for the favorite percentage to balance out, Rick. A hot stable goes cold, but not before running several losers, it takes a while for the public and insiders to get off them.
I love this thread, everyone has come up with some very logical explainations.
Perhaps it is a combination of all of them?