PTC Weekly Handicapping Report

[Premier Turf Club]

This week we're introducing a new product we hope to make available in the coming months. A good friend of ours we’ll call “AJ”for now, has spent the last five years developing a black-box computer handicapping model similar to the ones used by teams lead by Bill Benter, Peter Wagner, and Alan Woods. While each of these teams employ a slightly different model, what they all have in common is that they gather data from multiple sources and use a robust algorithm to generate win probabilities for each horse in a given race. For now, we are making these win probabilities available free of charge in the form a pdf that lists a horses name, probability, and corresponding win odds. Our ultimate goal is to make these available in file form that can be easily combined with the PTC tote to identify overlays. You’ll find a brief introduction to this product in Today’s Report, as well as probabilities for Delta Downs, Delaware Park and Hawthorne and an article on the Hong Kong computer handicapping groups.

The report can be accessed free of charge on our website www.premierturfclub.com under the <Handicapping> tab.

[boomman]

Thanks Ian! I will definitely check this out and I've heard the great news: You have just signed Retama Park back on! I congratulate the TTA in Texas for their efforts in bringing the horsemen to the table to show them that this, (and NOT the THG hijacking signal access) is in their best interests!! Retama will now be running until February with the extended Sam Houston dates, and I will be including them now in my wagering "menu"................:ThmbUp:

:) Boomer

[dutchboy]

I clicked on the handicapping tab but only see an old report about turf racing from LAD and CBY.

Am I looking in the wrong spot. I used the link in the message below.

This week we're introducing a new product we hope to make available in the coming months. A good friend of ours we’ll call “AJ”for now, has spent the last five years developing a black-box computer handicapping model similar to the ones used by teams lead by Bill Benter, Peter Wagner, and Alan Woods. While each of these teams employ a slightly different model, what they all have in common is that they gather data from multiple sources and use a robust algorithm to generate win probabilities for each horse in a given race. For now, we are making these win probabilities available free of charge in the form a pdf that lists a horses name, probability, and corresponding win odds. Our ultimate goal is to make these available in file form that can be easily combined with the PTC tote to identify overlays. You’ll find a brief introduction to this product in Today’s Report, as well as probabilities for Delta Downs, Delaware Park and Hawthorne and an article on the Hong Kong computer handicapping groups.

The report can be accessed free of charge on our website www.premierturfclub.com (http://www.premierturfclub.com) under the <Handicapping> tab.

[Premier Turf Club]

Dutchboy,

Please clear your cache. Its up there as of about 11 AM this morning.

[JeremyJet]

Ian,

Who's data is used to generate the probabilities? Woodside Associates?

Do you have data that shows how the probabilities have performed over a period of time? In other words, are the probabilities accurate?

Regards,

JeremyJet

[Premier Turf Club]

The raw data is not Woodside's.

I have been playing around with them for a couple of months now, using them mostly to validate opinions. I'm not expecting people to bet out on them blindly any more than I would on the analysis we provide.

As far as my results go, I can tell you betting on high probability overlays resulted in a 92 cent to 94 cent ROI for me pre-rebate which is pretty good. I thought the results were interesting enough to at least give them away for a beta test.

[JeremyJet]

The raw data is not Woodside's.

I have been playing around with them for a couple of months now, using them mostly to validate opinions. I'm not expecting people to bet out on them blindly any more than I would on the analysis we provide.

As far as my results go, I can tell you betting on high probability overlays resulted in a 92 cent to 94 cent ROI for me pre-rebate which is pretty good. I thought the results were interesting enough to at least give them away for a beta test.

So you're not at liberty to disclose the data provider that is being used at this time?

So when you say you're betting on "high probability overlays", you mean percentages in the higher end of the odds spectrum? You're avoiding overlays on like 50:1, 30:1, 25:1, 20:1 and 15:1? How did you handle the 26:1 shot that just won at Hawthorne, race 5, that was an overlay based on the 16.9:1 odds provided in the odds line?

The [vast majority] of odds lines that players generate need to be [tightened] to reflect the favorite-longshot bias. There's usually too much credit being assigned to the longshots and not enough to the favorites. That's been my experiance and i've been playing around with this sort of stuff for over 10 years.

Regards,

JeremyJet

[lamboguy]

this thing is no substitute for doing your own work. its like buyer numbers, sheets, and other forms of numbers. i personally don't use any of that, i buy no racing forms, but that is me. i apply my memory and insights into the game, plus i have a large amount of horses that are at various different racetracks that i know before they race for a trainer. when i say large, i mean we will have over 400 two year olds on the tracks come 2009, and we have a good idea who has more ability than the other.

[Premier Turf Club]

No high probability overlays means horses at the lower model odds that happen to be overlays based upon public assessment. High odds overlays would be low probability horses. Lamboguy is right, it is no substitute for doing your own homework. We are always evaluating data that we think will help our clients win. The more money people can win, the more they can churn, the better off it is for them, us and the entire industry. The goal is to help people stay alive in the game.

[dutchboy]

It would be interesting to know how a person would do if they kept the betting to horses that are between 3-1 and 8-1 on the projected odds line but are going off at odds that are a minimum of a 50% overlay.

I did not bet the Hawthorne race mentioned but after I saw it mentioned I ran it and the projected odds were 6-1. The race after that at Hawthorne with another big longshot the projected odds were 8-1.

[raybo]

Ian, I enjoyed the Hong Kong article, the references and comparisons to the stock market are spot on. I have believed this for several years, as confirmed by the statement in the signature on my posts. Thanks for including it.

The "artificial intelligence" mention, regarding handicapping software is something that peaks my interest and something I have toyed with over the last few years. My very limited programming skills have thus far prevented me from creating such a program, where the software "learns" from it's own data, in real time, and updates itself automatically, on the fly.

Ray

[jonnielu]

No high probability overlays means horses at the lower model odds that happen to be overlays based upon public assessment. High odds overlays would be low probability horses. Lamboguy is right, it is no substitute for doing your own homework. We are always evaluating data that we think will help our clients win. The more money people can win, the more they can churn, the better off it is for them, us and the entire industry. The goal is to help people stay alive in the game.

So as you go down this road, which is an admirable quest, have you considered that if you can not help your clients to win, you will be teaching that they can not win.

Don't get me wrong, I am not trying to dampen your spirits, what you propose is exactly what the racing industry needs to do, and success could resurrect the industry. While failure could finish it off.

The struggle is for the heart and mind of that person who already sees his/her chances as better with the purchase of a lottery ticket or pull of a handle. Those things are easier then the juggling of data too.

The 15 year horseplayer has already juggled so much data, that he is going back to Ainsley's books for a re-examination of past performance handicapping dogma.

The situation may be such that breaking out anything but the actual key to understanding thoroughbred racing could be disasterous. It may be prudent to examine any data in that light.

jdl

[john del riccio]

So as you go down this road, which is an admirable quest, have you considered that if you can not help your clients to win, you will be teaching that they can not win.

Don't get me wrong, I am not trying to dampen your spirits, what you propose is exactly what the racing industry needs to do, and success could resurrect the industry. While failure could finish it off.

The struggle is for the heart and mind of that person who already sees his/her chances as better with the purchase of a lottery ticket or pull of a handle. Those things are easier then the juggling of data too.

The 15 year horseplayer has already juggled so much data, that he is going back to Ainsley's books for a re-examination of past performance handicapping dogma.

The situation may be such that breaking out anything but the actual key to understanding thoroughbred racing could be disasterous. It may be prudent to examine any data in that light.

jdl


Can you repeat that ?

John

[jonnielu]

Can you repeat that ?

John

Okay

Take the entire pile of "how to handicap information" into consideration.

Has it taught the handicapper that "Yes, you can win" to the extent that he has this firmly in mind. Does he know that he can win?

jdl

[JeremyJet]

Does anyone know what kind of data these people used in their handicapping process? I find it hard to believe they used generic data available to everyone. I don't know, maybe they did.

That's why I asked the question about what data is being used to come up with the odds line. With all due respect to Henry Kuck, I have never met a player that uses the "legendary Henry Kuck ratings" as part of their handicapping arsenal. What makes them "legendary"?

If you're interested in reading about these [computer betting teams], get a book titled "Efficiency of Racetrack Betting Markets". Not sure if it's available in the US, I had to get mine from the UK.

Regards,

JeremyJet

[raybo]

Can you repeat that ?

John

I think what he is getting at is that, for those players who are becoming disenfranchised with, or have already started shifting their interests from, horse racing, staying alive in the game will probably not be enough for them.

If they are now losing money and what they gain from your efforts will still leave them in the red, even though not as far in the red as before, they probably are saying to themselves, "This is too much work and I'm still losing.", or something to that effect.

I guess what he is proposing is that you change what you have been doing so you can make your clients profitable. Easy to say, huh?

Personally, and I've been at this for over 30 years, I have always been a past performance handicapper and have, since almost the very beginning, included class, pace, form, and distance capabilities in my handicapping, The only things I have changed over all these years is the tools I use to look at and compare these racing factors. I went from pencil and paper and the DRF and attendance at the track, to computer automation of the number crunching, past performance data files and online wagering. Everything else has remained steeped in "old school" handicapping, and it still works. Of course, one must always be vigilant concerning changes in the industry: surfaces, training methods, new technology, different wagering opportunities, loss of value due to increased knowledge and the subsequent overuse of that knowledge by the public, etc..

There are many ways to "skin a cat" or handicap races, but the bottom line is, and has always been, experience, patience, consistency, and the demand for value.

[raybo]

Does anyone know what kind of data these people used in their handicapping process? I find it hard to believe they used generic data available to everyone. I don't know, maybe they did.

That's why I asked the question about what data is being used to come up with the odds line. With all due respect to Henry Kuck, I have never met a player that uses the "legendary Henry Kuck ratings" as part of their handicapping arsenal. What makes them "legendary"?

If you're interested in reading about these [computer betting teams], get a book titled "Efficiency of Racetrack Betting Markets". Not sure if it's available in the US, I had to get mine from the UK.

Regards,

JeremyJet

I believe he mentioned the use of advanced algorithms which leads me to believe that they have unique numbers. They have extensive access to historical data while updating their methods/numbers, weekly if not daily. They have the ability to place batch bets at the absolute last second which allows their wagering system to analyze the possible profitability of all types of wagers available with, as close as is possible, the final odds. They are looking for value in their plays, no matter the relative value, just value. They have the resources and technology to "hammer" value plays without diminishing the profitability of those plays, due to the huge pools they are involved with.

Bottom line is that they have a very huge advantage over the public, and probably most of us here, due to their advanced technology and long term game plan.

[JeremyJet]

"advanced algorithms" are all find and dandy, but you still have to be working with quality data. What data is being used? Why is it a secret?

There's so many speed figures out there that it will make your head spin. And all these people want you to believe that their numbers are the best. That's just not reality. The reality is that they want your business.

Regards,

JeremyJet

[raybo]

"advanced algorithms" are all find and dandy, but you still have to be working with quality data. What data is being used? Why is it a secret?

There's so many speed figures out there that it will make your head spin. And all these people want you to believe that their numbers are the best. That's just not reality. The reality is that they want your business.

Regards,

JeremyJet

I doubt that the raw data they use is a secret, the way they use it is. My interpretation of the article, regarding their handicapping methods, is that they don't use someone else's speed figures, pace figures, class figures, etc.. They create their own from raw data and obviously a huge database of historical data for horses, tracks, trainers and jockeys, and no telling what else.

And I don't think they want your business, anymore than Warren Buffet wants your business. They just want your money in the pools.

[jonnielu]

I think what he is getting at is that, for those players who are becoming disenfranchised with, or have already started shifting their interests from, horse racing, staying alive in the game will probably not be enough for them.

If they are now losing money and what they gain from your efforts will still leave them in the red, even though not as far in the red as before, they probably are saying to themselves, "This is too much work and I'm still losing.", or something to that effect.

I guess what he is proposing is that you change what you have been doing so you can make your clients profitable. Easy to say, huh?

Personally, and I've been at this for over 30 years, I have always been a past performance handicapper and have, since almost the very beginning, included class, pace, form, and distance capabilities in my handicapping, The only things I have changed over all these years is the tools I use to look at and compare these racing factors. I went from pencil and paper and the DRF and attendance at the track, to computer automation of the number crunching, past performance data files and online wagering. Everything else has remained steeped in "old school" handicapping, and it still works. Of course, one must always be vigilant concerning changes in the industry: surfaces, training methods, new technology, different wagering opportunities, loss of value due to increased knowledge and the subsequent overuse of that knowledge by the public, etc..

There are many ways to "skin a cat" or handicap races, but the bottom line is, and has always been, experience, patience, consistency, and the demand for value.

What I am saying is, that the target market is something that Ian and the racing industry are stuck with. There is no luxury of targeting. The capture will have to be made from a very large group of people that have grown up in an image driven, instant gratification society. You can't take advantage of that preaching hard work and long study.

They never mowed a lawn for a dollar, and they aren't much motivated to work toward anything, unless it is the weekend. Since they never took hold of the 60 hour work week, they aren't about to pick it up now, they have already rejected any cat-skinning methods that require experience or patience.

Hell, they don't really care if the cat gets skinned or not, the racing industry will need to flag them down and drag them in as it is.

I don't think that it will be a case of "build it, and they will come." I think that it will be a case of "anybody can win at horse racing". And, that will have to be true.

Fortunately, for the racing industry, it is true. Possibly unfortunate, is that someone in the racing industry needs to figure that out, and get it out there soon.

jdl

[JeremyJet]

And I don't think they want your business, anymore than Warren Buffet wants your business. They just want your money in the pools.

Well, if they want my money in the pools, they're not going to get it offering Retama Park as an option ... or any of the other bush tracks they're offering at this point in time.

Regards,

JeremyJet

[PaceAdvantage]

Well, if they want my money in the pools, they're not going to get it offering Retama Park as an option ... or any of the other bush tracks they're offering at this point in time.What U.S.-based and licensed ADW are you using that offers better tracks and comparable rebates?

[raybo]

Well, if they want my money in the pools, they're not going to get it offering Retama Park as an option ... or any of the other bush tracks they're offering at this point in time.

Regards,

JeremyJet

The reference was to what their doing in Hong Kong, not the US.

[JeremyJet]

Ian,

How have the odds lines performed over the past month?

Regards,

JeremyJet

[pktruckdriver]

this thing is no substitute for doing your own work. its like buyer numbers, sheets, and other forms of numbers. i personally don't use any of that, i buy no racing forms, but that is me. i apply my memory and insights into the game, plus i have a large amount of horses that are at various different racetracks that i know before they race for a trainer. when i say large, i mean we will have over 400 two year olds on the tracks come 2009, and we have a good idea who has more ability than the other.



Sounds to me like you guys could use a good old fashioned truck driver to haul your horses around, How bout it??


As always Ian great job keep it up, we need more people like you.


Patrick

[Premier Turf Club]

As always Ian great job keep it up, we need more people like you.
Patrick

Thanks Patrick. I enjoy doing all this stuff, just wish there were more hours in the day to do it. Again a big shout out to our supporters, especially guys like CJ who graciously provide his figures every day. In response to numerous requests, we have added an on-site Tampa report similar to the observations we provided each day from KEE. Joe and I are working with additional tracks to add that to our daily handicapping package that's free to all PTC customers. :)

Jeremy, to answer your question we have not given up on providing odds lines on selected tracks though we have not provided them or even tracked them recently (just too much going on). Our modelling expert is tweaking them even as we speak.

Ian

[JeremyJet]

Jeremy, to answer your question we have not given up on providing odds lines on selected tracks though we have not provided them or even tracked them recently (just too much going on). Our modelling expert is tweaking them even as we speak.

Ian

Ian,

Consider useing the data/speed figures CJ provides you. If the figures are any good you should be able to generate some quality odds lines. I've done this with The Sheets/Thorograph with some great results. Only problem I have with this method is that it's too time consuming the way I do it. But you should be OK since you've automated the process with an algorithm.

Regards,

JeremyJet