Pace and Cap  - Sartin Methodology & The Match Up

Pace and Cap - Sartin Methodology & The Match Up (http://paceandcap.com/forums/index.php)
-   General Discussion (http://paceandcap.com/forums/forumdisplay.php?f=5)
-   -   Finding +EV (http://paceandcap.com/forums/showthread.php?t=11699)

rdiam 07-02-2018 06:29 PM

Finding +EV
 
I hate to "beat a dead horse" (pun intended), but the secret to "beating" the races is not picking the right contenders/pace lines, but finding horses whose public odds differ significantly from their probability of winning to overcome the track take and breakage, i.e, +EV. Horses whose odds are significantly below their win probabilities should be bet against, while those whose odds are significantly above their win probabilities should be bet.

According to Glendon Jones, author of "Horse Racing Logic", favorites will be overlaid to their win probabilities about 20% of the time, and underlaid to their win probabilities also about 20% of the time, which neatly lines up with the current win % of favorites these days of around 40%. This implies that in 20% of the races we should be looking to bet on the favorite, and 20% of the races we should be looking to bet against the favorite.

In the remaining 60% of the races, the favorite will be "properly" bet to reflect its win probabilities, and these are the races to look for "classic" overlays. However, statistical studies show there is a favorite/longshot bias in the paramutuel win pool, such that only 10% of horses showing 5-1 odds will be overlaid to win probabilities, and only 5% of horses showing 10-1 odds will be overlaid, and only 2% of horses showing 20-1 odds and higher will be overlaid. This focuses on a discussion about a 20-1 cutoff in another thread, where the "real" reason for passing on most of these horses is that their odds rarely exceed their win probabilities.

So we will need to be betting favorites around 1 out of every 5 races: the criteria for such a bet that the favorite's odds are 1-1 or higher and the SECOND favorite's odds are 7-2/3-1 or less AND has very little chance of winning. Even better if the third public choice also has very little chance of winning.

We will also needs to bet AGAINST the favorite around 1 out of every 5 races: the criteria for such a bet if that the favorite if bet down to at least 3-2, preferably 1-1 or less AND cannot win today. Most of the time our bet will be on the second public choice and/or the third public choice.

Roughly 60% of the time the first 2-3 public choices will be fairly bet or overbet, and this is the situation to try to find the "classic" overlay: my rule of thumb in this case is to bet such a horse as long as its odds are equal to or greater than half the number of horses starting: that is, in a 6 horse field, 3-1 minimum, in an 8 horse field, 4-1 minimum, in a 10 horse field, 5-1 minimum. The Sartin 2-horse bet in such a case according to Kelly (which will maximize bankroll growth) is to bet this "classic" overlay or overlays combined with enough on the "fairly" priced favorite to breakeven if/when the favorite wins. Alternatively, bet a single horse to win, and back it up with a place bet big enough to breakeven if the horse comes in second.

Bottom line, you are badly mistaken if you think you can create +EV without watching and using the tote board to structure your bets.

Richard

Lefty 07-02-2018 10:05 PM

Overlays, underlays, have all been discussed hundreds of times through the yrs. But no matter how you structure an odds line, whatever criteria you use, I contend it's all the SWAG system, Scientific Wild Ass guess.

mick 07-03-2018 12:21 PM

Richard, that's a fine book. I haven't read it in several years but seeing your excellent post, I pulled it off the shelf. According to the dust jacket, Glendon Jones has a master of science degree in applied statistics from the University of Toronto. (I have a degree or two, but I don't have one of those.)

I understand Lefty's point. For several years, back in my "paper and pencil handicapping" days, I tried to construct my own odds line for each race and frankly, it was little more than SWAG. For all that effort, I don't think I improved my ROI a penny. Perhaps the only benefit was that it kept me out of some races that I shouldn't have bet anyway.

Still, the concept of value has stayed with me. I have always been a big fan of Dr. Bill Quirin and his book Winning at the Races remains one of my favorites. After much thought and watching at least a dozen Dave Schwartz videos, I have decided that an odds line based on measurable factors, weighted by impact value, might just be the answer, at least for me. (Dave was an early Sartin guy who also was influenced by Dr. Quirin and his philosophy and software blend the two.)

So, I'm waiting patiently on our friend Ted to release the new version of RDSS with the tote fix, among other things, and then I'm going to build a large RDSS database that contains all the horses in each race, not just the winner or the place horse, and after I have a thousand races or so, I'm going to do some serious impact value analysis.

One of the best players that I know on P&C is Richard Mathie who has created his own four-factor system from vast statistical studies which he has done by hand! He can tell you the winning percentage of all contenders from their four-factor profiles. I have spent many hours on the phone with Rich handicapping races and he almost never loses. He is using his percentages to determine value, both in the horse and in the race. He is a very disciplined player and only bets when there is value. (Needless to say, he doesn't bet many races.) In a manner of speaking, he's creating an odds line with his statistics, an objective odds line, not a subjective SWAG version like Lefty mentioned.

Brian Chamberlain 07-03-2018 01:12 PM

I totally agree with y'all. Bill Quirin's book featured in Brohamer's Modern Pace Handicapping. If you believe RDSS odds and find the contenders the public doesn't agree with and have the discipline to pass races, this game becomes more fun.
Happy Independence Day!
Brian

rdiam 07-03-2018 06:27 PM

I always like to remember that betting horses in the US is still a parimutuel game (except for Betfair in NJ), so the key feature is that we are not really betting ON or AGAINST horses, but we are betting AGAINST the other bettors in the parimutuel pool. In reality, we are betting against people or groups of people WEIGHTED by their bankroll size, rebate size, ability to do late calculations on bets and ability to make these bets as late as possible. It is NOT a level playing field. "Friend" of Pace and Cap Bill Ziemba has co-edited a book "Efficiency of Racetrack Betting Markets" which describes the math and statistics being used by Bill Benter, the late Alan Woods, and others. It is considered the "Bible" by those looking to make probability estimates in a horse race. If it were easy, then everyone would be doing it, and the +EV wold disappear.

The purpose of my post above was to suggest a "shortcut" to making these probability estimates -- let the tote board do it for you. A bettor needs to overcome a "vig" of at least 25% to just break-even mathematically in this game. The simplest way to do this is find a horse that is bet on the tote at 3-1 or less that cannot win (25% of the pool and the public's probability estimate right there for the taking). The public has to make a mistake or there is no +EV and no bet. This eliminates the "vig", and if your handicapping is any good, you are now betting into a remaining pool that will return +EV.

In addition, mathematically there is a special relationship (as pointed out by Jones in his book) between the tote favorite and the tote second choice, so that if you deem a favorite bet down to 3-2 or below as a "false" favorite who has little chance of winning the race, you are most likely getting +EV on the tote board second choice. Even better if your handicapping indicates another horse. Similarly, if you are going to bet the favorite at 1-1 or higher, it will likely be +EV if the second favorite is bet down to 3-1 or lower and your handicapping says it cannot win.

Note that this is a blunt force method -- no precise probability estimates or "scientific wild ass guesses" were needed or used, just handicapping judgement as to whether 1-3 horses can or cannot win (with a commensurate mistake by the betting "public") and comparing their relative odds. It is also a method that flies somewhat under the radar (although it is mentioned by Mike Maloney in his "Betting with an Edge".

One more thing -- sometimes we forget that the only thing that matters in betting (or investing for that matter) is bankroll growth. Bankroll growth is a function of win % and the number of bets. If there are not enough bets the win % becomes irrelevant even when it is quite high. When betting Kelly (or some fraction thereof) where we are betting our "edge" divided by the odds, we need and want the compounding feature to maximize bankroll growth. Not having enough betting opportunities truncates this feature.

Richard

Bill V. 07-03-2018 08:24 PM

Thank you
 
3 Attachment(s)
hello Richard

I want to thank you, I really appreciate your work.

It just paid off for me. This horse ratings-wise was fine, there were only afew others in the field.
What caught my eye was just as they were near the gate Mr Grey had dropped
to 2/1 from the 8/1 ML

I was so happy when I saw MR grey win ( place) for me at 4/1

I appreciate your eye-opening post.

Mr grey was the outstanding choice
The tote board tipped me off

Attachment 46446

Attachment 46447

Attachment 46448

dippeli 10-20-2018 01:48 PM

Quote:

Originally Posted by rdiam (Post 116237)
Horses whose odds are significantly below their win probabilities should be bet against, while those whose odds are significantly above their win probabilities should be bet.

Absolutely!

Quote:

Originally Posted by rdiam (Post 116237)
Bottom line, you are badly mistaken if you think you can create +EV without watching and using the tote board to structure your bets.

FALSE!

You can create situations that are +ev IN THE LONG-RUN without looking the tote board.

mick 10-22-2018 06:14 PM

Quote:

Originally Posted by dippeli (Post 118073)

FALSE!

You can create situations that are +ev IN THE LONG-RUN without looking the tote board.

Always interested in new concepts. Is there more? :)


All times are GMT -4. The time now is 09:31 AM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.