RSS Feed for this Blog

Monday, April 21, 2008

Chalk Players Eating Poly

- I wanted to do a simple calculation - to determine the winning percentage of favorites on main track races at Keeneland this spring. Shouldn't be too hard, right? Just check out the result charts for each day.

Unfortunately, it comes at a cost; free result charts are only available for around a week these days it seems. Otherwise, you can buy them at BRIS for 50 cents a shot (per day/per track). Seems a pretty restrictive stranglehold on information, as noted from time to time by other members of the TBA. By contrast, I can go on ESPN.com, and easily find a box score, complete play-by-play, and a game story for every single major league baseball game going back to 2002!

Anyway, back to the subject at hand - enterprising as I am, I managed to get all the charts for free using Formulator....except for opening day, which I just guessed from the win prices. Out of an even 100 races conducted on the Polytrack this spring, favorites have won 20, for a percentage of (obviously) 20%. So those who complain that there's no form on the Poly can point to a number which is far below the universal average of 33%.

I'm strongly in favor of 33% winning favorites; it's the accepted norm, whether at the flats or harness, big track or small, and I think it's just about right. Enough favorites for it all to make at least a little sense, and with which to go to the well when you need a single or just a little confidence. But also plenty of opportunities to take a stand and make it pay off. 20% is pretty low, though it's not that big of a sample in the long run. If the Keeneland meeting went on for two months, statistical probability would likely nudge the figure closer to the norm. I'd check out the stats for the three prior Polymeetings at Keeneland to get a bigger sample....but I figure it'd cost me around $22.50.

25 Comments:

Anonymous said...

Unfortunately, it comes at a cost; free result charts are only available for around a week these days it seems. Otherwise, you can buy them at BRIS for 50 cents a shot (per day/per track). Seems a pretty restrictive stranglehold on information, as noted from time to time by other members of the TBA. By contrast, I can go on ESPN.com, and easily find a box score, complete play-by-play, and a game story for every single major league baseball game going back to 2002!

Alan,

This is just another way to attract players(and gamblers) to Horseracing!

Dave B said...

Given a choice between 20% and the 40-45% that is the norm at our local harness track, I'll take the 20%. Nothing more boring than a day with 8 chalk winners in 11 races!

Teresa said...

Been griping about this, as you know, for months. It seems outrageous that someone can "own" a set of facts that are essentially in the public domain.

Anonymous said...

Synthetic tracks just add another huge dollop of uncertainty to a game that already has too much uncertainty.

If this fad, and that's all it really is (Michael Dickinson to the contrary notwithstanding) doesn't die of its own absurdity, it will be another nail in the coffin of this game.

Ballyfager (Sometimes Google/Blogger will remember me, more often not and THAT is going to be a nail in the coffin of THIS game for me).

Handride said...

Horseracing where customers are #28. Calling #3, Calling #3.

Thanks for the link.

Anonymous said...

The Whales mostly place races with definable favorites, or pass. Few play the exotics.

This is why overall handle is down.

I, and many smaller players, search for false favorite to play against. Without the whales, few opportunities exist.

Handle declines further.

If left to the two dollar better, the spread between longshots and favorites will continue to decline, when combined with the uncertainty of the result makes the races even less playable.

Disappointed Stronarch seems to be caving to the politicians and will likely install another artificial track at Santa Anita. He has a one time opening to opt for tradition but is failing to stand firm.

Nick said...

I'd be interested to see what the number is for other synthetic tracks. With the possible exception of Del Mar, I don't get the impression that there is as much gnashing of the teeth elsewhere as there is at Keenland.

El Angelo said...

Theresa:

I wish they were available too, but I'm going to disagree that they're "public domain". Someone, be it equibase, DRF or Keeneland themselves, is putting in considerable time, money and human capital into making the charts. It'd be beyond moronic to make you pay for them the day or the race and the next few days, but I definitely can appreciate why they charge for their product after that point.

Angelo

Harl said...

20% sounds about right while the handicapping public catches up to these new nuances of the game. I imagine there are still a lot of casual players handicapping Keeneland as a dirt track or (gasp!) like a Cushion Track or Tapeta surface. There are differences in each and one the public figures that out, that 20% will move closer to 33%. In the meantime, I am more than happy to have a slight advantage over those who are still trying to figure it out or spending their time bitching about it instaed of doing a little research. Isn't that what this game is all about?

Teresa said...

I don't disagree entirely, el angelo, that Equibase deserves some compensation for their recording of and presentation of their observations.

My problem is that the information they compile is readily available in multiple forms (by people at the track, by people watching races on TV, etc.), and my understanding is that Equibase could sue me for copyright if I decided to start my own little database of readily observable information and then make it public. If I'm wrong about that, I'd love to hear it.

If I'm not, it seems absurd that one company exclusively gets to own information that is as common as the score of last Friday's Rangers game, then charge people to see it when it's no longer easily accessible.

Erin said...

Harl - Are you saying you have been able to come up with new handicapping rules and standards for poly? Because it seems to me that's part of the frustration - no consistent pattern, no clear type of horse that poly flatters. Personally safety should be paramount, but after the loses by Pyro (and the other shorter-priced horses in the Bluegrass) and Tomcito, I find Baffert's words - poly making good horses bad and bad horses good - echoing in my head. Brilliance seems to getting stiffly punished. I will be interested to see how Monba and Cowboy Cal progress, and continue to remind myself that Street Sense handled both the Keeneland surface and dirt last year.

Matt said...

Alan,

Great point about ESPN...but it's even better than that for baseball stats. One can go to Baseball Reference.com and find lifetime statistics on any player to play the game...or to Retrosheet.org and find a box score for almost any game ever played in baseball history...all for free. (Retrosheet is a great site for baseball junkies) And all of that information was compiled by somebody.

If I want a lifetime past performance of one horse it costs me $5.00 at DRF. Pretty insane.

I realize Equibase and everybody else want to make their $$$, but basic information such as results should be freely available. Charging for the Beyer figs, or pace figs or whatever else is fine, but things like which horse won and the final time should be free.

El Angelo said...

Theresa:

There's absolutely nothing I can think of stopping anyone (hey, including us!) from starting a site that simply records the results of races at major tracks; ie, who finished in the money, what they paid, the conditions, etc. That actually might be a useful site to create.

Angelo

alan said...

Erin - One handicapping rule and standard I've been using with the Keeneland Polytrack is that form between it and turf is 100% interchangeable. I've had a decent amount of success using that assumption, and think it's a good place to start.

Anonymous said...

Anonymous 8:50

This is Stronach's chance to make a right decision for a change. If he stands up to the politicians and bureaucrats and flatly refuses to install another synthetic track at SA what will they do to him?

Nothing.

Anonymous said...

Anon 12.04, agree, but for whatever reason he appears to be caving.

His new dirt surfaces, with brand new bases, have performed beautifully, yet you do not see anyone doing comparisons of breakdown rates pre and post their installation.

That would be interesting.

as for the charts, equibase clearly copyrights every chart and posts in pdf form so they can not be easily copy/pasted. Suppose if you wish you could recreate the whole thing yourself and post for free, and then compile pp's for each horse, but not easy to do for the hundreds of tracks in the country and world.

MLB has 15 games per night six months of the year, with about 30 players on average per game.

There is a LOT more data involved in racing, with nowhere near the level of demand.

alan said...

Don't know if you noticed that Stronach seemed to support a switch to dirt when he spoke to HRTV on Sunday.

But he apparently changed his mind after speaking to horsemen later that day (not sure if that's the correct chronology however).

Erin said...

Alan,
I did notice that turf form carried over to Poly at Keeneland last fall, but this spring it doesn't seem that easy. I haven't looked at as many races this spring though - thanks for your response, I'll keep it in mind.

Also, I'd add that my comments refer strictly to Keeneland's synthetic - I do believe this synthetic course is an extreme example of what we're seeing elsewhere.

Teresa said...

El Angelo and anon: another TBA blogger and I had planned to do this for NYRA tracks over the winter...reserved a blog name, all ready to go.

As anon pointed out, it's incredibly time-consuming, even to record only the order of finish and prices. So, OK, Equibase gets kudos for taking the time to do it, and a right to make some money off of its efforts.

But there's a difference between being proprietary about the information itself, and about the presentation of the information. Why can't NYRA keep a complete order of finish and prices archived on its website?

I was daunted by the time commitment of getting this going...but if a few more hands want to join the effort and share the labor, I've got the blog all ready to go...

alan said...

Teresa -

Curious if you and the other blogger looked into the full legal ramifications of that? Because even though the information is public, I imagine that they could conceivably have what's known as a business method patent. In this case, it could hypothetically be a patent on the method of compiling a race result into a chart form with odds, comments, etc.

Teresa said...

We didn't, Alan, though it's exactly the sort of thing we wanted to put to the test.

Though I don't think we ever envisioned (I didn't, anyway) anything like full race charts. It really would be order of finish and the prices: basic information.

This came up for me when the Donna Wormser drug positive on Travers Day came out in the fall...I wanted to see who finished second and what the exacta paid, and I could do that only by buying the chart (which of course I didn't do). So I sort of decided that I wanted to test the waters, to see what fur might fly if I watched race replays, wrote down the order of finish and prices, and posted them daily.

Of course, I didn't only want to tweak...also wanted to get what I consider basic info out there.

I need to go now, as my inbox is filling up with e-mails from lawyers...

steve in nc said...

I'm not surprised about the lack of winning Polytrack favorites for two reasons:

1) The field size at KEE is much bigger than the average meet. I'm guessing maybe 10, compared with maybe 7.5 or less for NYRA in April? If those guesses are in the ballpark, then (if my math reasoning is correct) KEE would raise it's % of winning favorites to around 27 or 28 just by cutting field size to NYRA's level. That accounts for about half the difference. People analyzing the % of wire-to-wire winners also need to take field size into account.

2) Obviously I'm not alone in having difficulty handicapping KEE's races. I've done very little of it, and my win rate was normal for me, but I rarely have so many races on dirt where all my top picks are nowhere, as I had in my short foray last week at KEE.

But I hope they keep the variety of surfaces. The more factors, the more undeserving chalk. Last weekend, I lost money trying to beat Gayego (first time dirt), but enjoyed betting against the low-priced first time Poly horses in the Blue Grass.

rgustafson said...

Alan, you should have checked the turf races. According to my stats, favorites have won 4 out of 23 races or between 17%-18%, so why no complaints about the turf.

QQ said...

Alan, I skimmed the comments here, so maybe I missed it -- but has anyone suggested that their local library subscribe to the Form or look into a database deal with Equibase? This is the kind of thing libraries exist for -- preservation of information for all of us, not just those who can afford it. But without input from their patrons, libraries don't always know about these resources. Of course, those lucky enough to be in Lexington might stop by the Keeneland Library to consult the entire DRF archives.

Harl said...

Yeah, it seems turf form translates to Keeneland Polytrack quite a bit, but it's not an end all. I think a lot of horseplayers assume all synthetic tracks play the same and that's a big no-no. Dirt form holds steady on Cushion Track, but not as transferable to Polytrack. A Tapeta surface is a different animal; I'm still trying to get a handle on that.

Just because a horse has a solid "synthetic" record, doesn't mean he'll post similar results at Del Mar as he/she does at Santa Anita or Hollywood. I think that is one area where you can get an edge.

There are other things to look at such as trainer intent that come into play at Keeneland, but that's been a solid angle to play since before they installed a synthetic surface and one I'll continue to play even if they're running on concrete.