PDA

View Full Version : At The Races program review


Dick Schmidt
07-04-2002, 05:30 AM
This is a review of a new computer program I recently completed. I got permission to post it here in hopes that some of you may find it interesting. Before you ask, I have no financial connection to the program other than buying a copy. I have never met the programmer, Gary Pizzigati, but he sure does very nice work. Fast, too. I've known Jack Burkholder for some time and he is best described as a consummate professional.

Dick Schmidt



At The Races

Computer program by Gary Pizzigati and Jack Burkholder

Review by Dick Schmidt


This is one of the finest pieces of software available for the modern, computerized Internet handicapper. Such a shame most of them will never consider buying it. Not that it’s expensive, hard to use or not needed; it is none of those things. It’s what it doesn’t do: pick horses. This is a betting program, and thus falls into that vast morass of dead programs and ideas called money management. Too bad, because this is the program most of us need far more than another “horse picking” program. On the off chance that there are some professional level players out there reading this, this program is what you have been looking for.

What At the Races does is access the internet through your regular internet connection (modem, cable, DSL, whatever), gather minute by minute toteboard information, point out betting opportunities, do “Dr. Z” calculations on the place and show pools, show you all the exacta payouts, including overlays and underlays, allow you to dutch your win and exacta bets, show you when you make more money wheeling a horse in the exacta than with a straight win bet and in general give you the information you need to make a bet. Since you provide your own handicapping information, it will work with any program, system or method available, including homemade ones. If you have a computer and a connection to the Internet, At the Races will work for you.

Because the documentation that comes with the program is a bit skimpy, I spent an hour on the phone with Jack Burkholder, the program’s designer, going over the fine points. Jack has been a professional player for over 20 years and he designed this program to do exactly what he wanted from a betting program. His professionalism and experience are obvious in the way the program operates. As time goes on, more “bells and whistles” will be added to the program, but right now it is a fully functioning, professional level tool. Anyway, here is a brief overview of how the program operates, as I understand it. Both of the principles behind this program are very willing to answer your questions by e-mail and are open to suggestions for improvements, so if I leave out anything, you can always get more information from the source.

When you first open the program, you first need to click on the Internet Options button and tell the program how you are accessing the Internet. I use a high-speed cable modem and found that I could ignore the questions about proxy names and ports, I assume because my connection is always on. Anyway, I just told it I was using a proxy for connection and away we went. It appears that if you are connected to the Internet when you start the program, it will find your connection.

Next choice is to decide which Internet toteboard to use. You can choose between “R”, the Racing Channel and “S” the BRIS Supertote. If one doesn’t carry the track you want, chances are the other will. This flexibility is typical of the program. Given the amount of bickering that goes on between the various racing entities, you never know who will be carrying what next week. At the Races lets you find the track you want. Should things change and other toteboard options come along, I’m sure the program will be updated to accommodate them.

Once you choose a toteboard option, you will be shown a list of available tracks. Lots of tracks. Dog tracks. Harness tracks. Australian tracks. Tracks you never heard of. It’s wonderful the amount of information available on the Internet. It’s also wonderful to be able to utilize it. You can view the track toteboards in two ways, either one at a time or scrolling through a list of selected tracks. If you want to see the tote action at just one track, select it and tell the program to show it to you continuously. The tote will be updated every minute. Scratches are posted as they occur. You can also tell it to show you one track without updates. You can edit the track list to just those tracks you are interested in and have it cycle through once or continuously. Amazingly flexible.

Let’s assume you want to watch the tote for the next race at Belmont. It is 20 minutes to post, and you want to follow the tote action. You select Belmont as the track you want and tell the program to update continuously. Now the fun begins. The first screen you will encounter shows the win prices (the actually return for a $2 bet), the maximum and minimum place and show prices with Dr. Z bets highlighted in blue (the actually advantage is shown under Place and Show ROI/$1) and the W/P/S pools, both total and for each horse. There are also two columns labeled Pitz Exacta and Pitz Double. These show you when it is better to bet a horse in a dutched exacta or double wheel instead of win. In other words, say you like the four horse. If he is paying $6.80 to win, it may be more profitable to bet him on top of all the other horses in the exacta pool. This is a complex bet, as you must bet a different amount on each combination, but the program will figure it out for you in a blink. Many times you can turn a low paying favorite into a much more profitable situation in the exacta pool. Lots of useful stuff, but there is even better stuff just ahead.

In the lower left-hand corner of the screen is a dutching box for win bets. You click on the white box under the horse’s number you want to bet and it will tell you how much to bet on each. It is set by default to bet $100 a race, but you can change that by clicking on the Win Bets white box to the right of the dutch box. If you want to dutch exotics, you can set them as well. Click on the box and then enter any amount you wish, from $20 on up. A neat feature is that you can also hedge a bet. In other words, say you like the favorite and two others. Since you can’t really make money betting favorites, you decide to hedge, betting the favorite to break even and splitting the rest of your money between the other two horses. If the favorite wins, you get your bet back, and you’ve put more money on your longer priced horses. To do this, hit the box a second time, and the label will change from Dutch to Hedge and will turn from green to blue. Hitting the box a third time cancels the bet.

Across the top of the screen, just under the track name, are a series of small buttons. The one labeled WPS Payoffs is checked. Move your cursor and check the next one, Odds. The first two columns show the current odds and the morning line for each horse. The favorite is shown in red, the second favorite in blue and the third in purple. As the odds shift, the colors change with them. The columns labels 1st Exacta and 2nd Exacta show the amounts bet on each horse, expressed as odds. The lower the odds, the more money bet in the exacta pool on that horse in that position, either on top or underneath. In our race at Belmont, the amounts bet each minute are displayed in the column labeled > 12 (greater than 12). The amounts are expressed as odds. Think of each minute of betting as being a separate pool. Starting with 12 minutes to post, each individual minute is displayed and recorded, again expressed as odds representing the money bet just during that minute. Therefore you can see that even though the 4 horse is 8-1 overall, he may have been the even money favorite during the fifth minute before post time when the trainer made his big bet. You can check to see how he did in the previous minutes and follow the action during the next few. Those who use the “Talking Tote” to make their bets have found their ultimate tool.

The next button is Exacta Stats. It shows all the possible exacta combinations first as a payoff, then as a percentage. If the percentage is red, the horse is that percent underplayed, based on the “fair” price of the win odds. Blue and the horse is that percent overlayed. Before you get excited and decide to bet just the overlays and get rich, most of the time Jack has found that it doesn’t work that way. In exactas at least, the big underlays are usually the ones to bet. If you don’t believe it, try it on paper for a while.

After that, we find the Exacta Bets screen, my personal favorite. Here you not only see the familiar exacta payoff matrix, but the tools needed to dutch and hedge exactas. Since the win dutch box accompanies us no matter which screen we are using, I find that here I have all the information I need to play a race. Directly under the exacta matrix are automatic plays labeled Low 2, Low 2 Box and so on. If you hit the Low 3 button, it will display the bets for dutching the three lowest paying exactas. Low 3 box shows the dutch for the 6 combinations of the lowest three odds horses. Play with these and you’ll soon see how they work. To erase what you’ve done, click the CLEAR box in the upper left hand corner.


Continued in part II

rrbauer
07-04-2002, 11:56 AM
Dick Schmidt wrote:
Jack has been a professional player for over 20 years and he designed this program

Comment:
Next time you speak with Jack ask him if a program called "Super Dutch" had any influence on "his" design.

rrbauer
07-08-2002, 09:47 AM
I have been involved with developing software to support horse-race betting for over 20 years. Because of that interest and experience, new tools such as, At The Races, get my attention. What follows are some thoughts and ideas that address the subject in general.

Background
Before simulcasting and wide-spread network access to betting pools existed, when all betting (pari-mutuel) on a track’s races occurred at that track, it was possible to enhance payoffs by carefully analyzing displays of payoff data to determine if a selection(s) offered better value in one pool than another. Basically a selection’s win odds were compared to the odds that would be available by betting the same selection to “win” through the exacta or daily double. Occasionally a cross-pool anomaly would occur and a bet could be structured to gain the advantage being “offered”. Similarly, a back-wheel bet could be used to “guarantee” a selection’s odds in the second half of a daily double where it was suspected that the selection would be bet down in the win pool when that race became available for betting. Placing the bets as close as possible to post time was essential.

Initially, the analysis was done manually with pencil and paper and perhaps a calculator. Later on, programmable calculators replaced the manual analysis. And, later yet, portable computers replaced the calculators. As simulcasting expanded to become nationwide and worldwide, pool and payoff volatility increased. Today, in the U. S. about two-thirds of the bets (in dollar terms) on the simulcast network, occur in the last two minutes of betting. That fact, combined with the latency factors in updating pool and payoff displays, makes the betting end-game highly volatile. Consequently, betting decisions based upon the public-access pool and payoff displays are subject to errors, even when there are zero minutes-to-post, because the betting decisions are driven by interpretation of display data that is still in a state of flux.

I have been experimenting for several months with a set of personal tools that “grab” and analyze the pool and payoff data from the Racing Channel displays for WPS, Double, Exacta and Quinella bets. The grab process was developed for me by a third party, with the resultant data formatted to fit the requirements of the analysis process that I developed. Two general outputs are provided: One is a betting plan based upon analysis of the pool/payoff data in relation to handicapping data that I provide. The second is a betting plan based upon analysis of the pool/payoff data, in relation to the implicit probabilities of winning based on the win-pool data. The latter is primarily concerned with finding imbalances across the various pools and exploiting them (arbitrage).

The betting plan tied to handicapping data is subject to two basic risks: First, the risk that the handicapping is flawed; and, second, the risk that the pool/payoff data is flawed. The second betting plan is subject mostly to the pool/payoff data flaw.

I am sorry to report to you that the results so far have been less than encouraging. And, in both cases, the major culprit has been the volatility of the end-game data. In the first case the bet amounts and type vary significantly with each pool/payoff update, near the close of betting. The other approach yields very few betting opportunities and about half of the opportunities uncovered involve betting the favorite to show.

I believe that products such as, At The Races (ATR), however slick, (and ATR is very slick) will be compromised by this data-volatility demon to the extent that most of their intended value and promise, as products, will not be realized by their users.

My review and specific comments regarding the use and features of ATR are covered in Part 2.

Dick Schmidt
07-10-2002, 02:28 AM
Richard B,

I have been using ATR for a few days now, and I agree that making betting decisions based on toteboard information is almost impossible given the late money that pours in. Then again, I never did think betting the tote was a very good idea.

One thing that I have found to be very useful is the ability to track the betting action minute by minute. Starting with 12 minutes to post, the action for each individual minute is recorded, and trends are easy to spot. Say you have a horse who is the morning line favorite at 5/2. As the betting progresses, he drifts down to 2/1. Then with 3 minutes to go, he drops to 1.7/1 (ATR does everything in decimals). 2 min and he's 1.4 to one. You can see the trend and the money coming in on this horse and project he'll wind up around 4/5. Same can be done if the betting on a horse is light, and he drifts up. Of course a huge bet right at the end can upset all this, either positive or negative, but at least you have some data to go on. I've not seen many huge bets drastically change the odds, though I did it myself once at Great Lakes Downs (note to self, $200 is too much!).

So far I'm just using ATR for win betting, but I plan to spend some time looking at exacta dutches next week. Those pools are more stable and much harder to move. I'll see how it works out and let you all know.

Dick

Dick Schmidt
07-12-2002, 05:33 PM
Making this post to keep the two parts of the review together.

Dick

PaceAdvantage
07-12-2002, 05:58 PM
Man, I'm getting lazy....will post the review on the permanent REVIEW site hopefully tonight...

Jake
07-12-2002, 11:48 PM
Richard, thank you. Very informative post.


Originally posted by rrbauer
I have been involved with developing software to support horse-race betting for over 20 years. Because of that interest and experience, new tools such as, At The Races, get my attention. What follows are some thoughts and ideas that address the subject in general.

Background
Before simulcasting and wide-spread network access to betting pools existed, when all betting (pari-mutuel) on a track’s races occurred at that track, it was possible to enhance payoffs by carefully analyzing displays of payoff data to determine if a selection(s) offered better value in one pool than another. Basically a selection’s win odds were compared to the odds that would be available by betting the same selection to “win” through the exacta or daily double. Occasionally a cross-pool anomaly would occur and a bet could be structured to gain the advantage being “offered”. Similarly, a back-wheel bet could be used to “guarantee” a selection’s odds in the second half of a daily double where it was suspected that the selection would be bet down in the win pool when that race became available for betting. Placing the bets as close as possible to post time was essential.

Initially, the analysis was done manually with pencil and paper and perhaps a calculator. Later on, programmable calculators replaced the manual analysis. And, later yet, portable computers replaced the calculators. As simulcasting expanded to become nationwide and worldwide, pool and payoff volatility increased. Today, in the U. S. about two-thirds of the bets (in dollar terms) on the simulcast network, occur in the last two minutes of betting. That fact, combined with the latency factors in updating pool and payoff displays, makes the betting end-game highly volatile. Consequently, betting decisions based upon the public-access pool and payoff displays are subject to errors, even when there are zero minutes-to-post, because the betting decisions are driven by interpretation of display data that is still in a state of flux.

I have been experimenting for several months with a set of personal tools that “grab” and analyze the pool and payoff data from the Racing Channel displays for WPS, Double, Exacta and Quinella bets. The grab process was developed for me by a third party, with the resultant data formatted to fit the requirements of the analysis process that I developed. Two general outputs are provided: One is a betting plan based upon analysis of the pool/payoff data in relation to handicapping data that I provide. The second is a betting plan based upon analysis of the pool/payoff data, in relation to the implicit probabilities of winning based on the win-pool data. The latter is primarily concerned with finding imbalances across the various pools and exploiting them (arbitrage).

The betting plan tied to handicapping data is subject to two basic risks: First, the risk that the handicapping is flawed; and, second, the risk that the pool/payoff data is flawed. The second betting plan is subject mostly to the pool/payoff data flaw.

I am sorry to report to you that the results so far have been less than encouraging. And, in both cases, the major culprit has been the volatility of the end-game data. In the first case the bet amounts and type vary significantly with each pool/payoff update, near the close of betting. The other approach yields very few betting opportunities and about half of the opportunities uncovered involve betting the favorite to show.

I believe that products such as, At The Races (ATR), however slick, (and ATR is very slick) will be compromised by this data-volatility demon to the extent that most of their intended value and promise, as products, will not be realized by their users.

My review and specific comments regarding the use and features of ATR are covered in Part 2.

Dick Schmidt
07-13-2002, 03:55 AM
PA,

Lazy doesn't begin to describe it! This review should have been posted days ago! What are we paying you for?

Dick

Oh, wait, we're not . . . Nevermind.

Dick Schmidt
08-12-2002, 05:26 PM
I'm posting this to "bumb" this up so people will stop sending me e-mail asking where it is.

Dick

andicap
08-12-2002, 06:14 PM
And once again, living up to your self-described "surly" moniker.

;)

PaceAdvantage
08-12-2002, 07:07 PM
I posted the review on the 'Reviews' section of this website some time ago....I guess nobody goes there....LOL

BillW
08-12-2002, 07:36 PM
Originally posted by PaceAdvantage
I posted the review on the 'Reviews' section of this website some time ago....I guess nobody goes there....LOL


Maybe someone ought to do a review (of the review section that is).:cool: