Positive Critics of Handicapping Software

[Capper Al]

I'd like to open a dialog about the various software and data packages available and only discuss their virtues and how they work. As a fellow programmer, I was a little hard on spending the big bucks at first thinking that I also could write my own stuff and save money. Now after five years or more of development and realizing that I have to do a re-write of my software package since I've learned so much more from coding, maybe buying into a system wasn't a bad idea. I believe the researcher in me would like to pull data down into a spreadsheet and hunt for patterns. These may be called using filters by software vendors. There are at least two categories of software-- those that only work with Pace and Speed, and others that are comprehensive and take into account more than Pace and Speed. I'm here to listen to those who have good things to say. And ask a few follow up questions.

[JJMartin]

I can do anything with my Excel program, let me know what you want to do as long as you use bris files. Pm me.

[Augenj]

I have something I'm going to present here in the next few weeks and it's pretty comprehensive - jockey, trainer, speed, finish, workouts, lead pack, and closer. After working on this for several decades, to my dismay I found that finding "patterns" in the past over hundreds of races is no guarantee that they will carry forward into the future. You need to apply these patterns to the future in dry runs and select those that stay profitable, if any do remain. More on this later.

[Track Collector]

I have something I'm going to present here in the next few weeks and it's pretty comprehensive - jockey, trainer, speed, finish, workouts, lead pack, and closer. After working on this for several decades, to my dismay I found that finding "patterns" in the past over hundreds of races is no guarantee that they will carry forward into the future. You need to apply these patterns to the future in dry runs and select those that stay profitable, if any do remain. More on this later.

Welcome to the real world of data mining.


..

[JJMartin]

I have something I'm going to present here in the next few weeks and it's pretty comprehensive - jockey, trainer, speed, finish, workouts, lead pack, and closer. After working on this for several decades, to my dismay I found that finding "patterns" in the past over hundreds of races is no guarantee that they will carry forward into the future. You need to apply these patterns to the future in dry runs and select those that stay profitable, if any do remain. More on this later.

I know what you mean first hand, I've gone through hundreds of patterns/systems, almost nothing stays consistent, or it does but results in a negative ROI. Sometimes you have picks that conform to a certain pattern and when you switch to it, the picks revert to a different one. It's a never ending battle. Another aspect is the current winning trend at the track, like when all the low paying horses win for a month straight, no method geared for value can defend against that scenario.

[Robert Goren]

When you have a computer searching for trends, you have to realize that there other people doing the same thing coming up with the same trends. I think this is a dead end for people looking to make money betting horses.

[JJMartin]

When you have a computer searching for trends, you have to realize that there other people doing the same thing coming up with the same trends. I think this is a dead end for people looking to make money betting horses.

If horses don't win consistently according to some factors, then how do they win? Randomly? The problem is not that you can't find factors that are predictive but that they result in a negative ROI. Like I stated in another post, 60% of the time the winner is one of the top 3 speed horses (at least the way I do it). I don't think its a dead end, it just takes a lot of work to get into a positive ROI consistently.

[Robert Goren]

If horses don't win consistently according to some factors, then how do they win? Randomly? The problem is not that you can't find factors that are predictive but that they result in a negative ROI. Like I stated in another post, 60% of the time the winner is one of the top 3 speed horses (at least the way I do it). I don't think its a dead end, it just takes a lot of work to get into a positive ROI consistently. 40 years ago you could have done that. I did for a while before I went out into the real world of business. Today you are fighting whales with lots of money and people and huge data base programs and super fast computers that access that you don't have. I believe you have to find stuff happens so rarely they haven't caught it. The easy stuff, they bet off the board. Using Speed Ratings every which way imaginable falls into the easy stuff for the whales. But hey, it is your life, just don't waste too much of it beating your head against a wall.

[Capper Al]

I was more interested in commercial handicapping software or homegrown developers sharing similar experiences. Thanks for the offers of help, but I don't need it. My own code is 14,000 lines of C++ feeding into two excel spreadsheets, one for maidens and another for non-maidens. Each spreadsheet has about 20 tabs with some going past 364 columns. What I'm finding is that I don't have enough time to do it all-- work, play, wager, family life, and code. This is what commercial software could have offered me the most-- my time.

[PICSIX]

The secret is....wait for it....wait for it..."fast horses are fast & slow horses are slow" :ThmbUp:

[jk3521]

Yeah, but sometimes those slow horses are just fast enough to win. :D

[RaceBookJoe]

Yeah, but sometimes those slow horses are just fast enough to win. :D

They just have to be "less slow" than the others :)

[Tom]

Try HTR. You have pace/speed, velocity, trainers, workouts, form cycle, any number of factor you can use, and if you don't like speed/pace. turn it off - totally ignore it. The two Robots can do massive amounts of testing whatever you like, and you can easily change the parameters. Once you have a spot play isolates, you can save it and run it every day. Some guys have literally scores of spot plays. The number of things you an test fast is awesome. And the exports into comma files for Excel or databases is massive.

You can test a play and get the results listed out for you by day, by week.
Try a demo of it - see what you think.

You asked in another thread how Ralph and I got that one horse at Arlington - he literally screamed BET ME of the HTR screen. The new toys Ken just gave us last month are pretty good - still learning what they can do.

And, as a bonus, your get all the T-Bred charts for every track every day thrown in, HW Data, so you know it is quality.

[Speed Figure]

The main problem I have with HTR is their are no videos available to see all of these features. I can't understand why Mr. Massa would not have a ton of videos showing all of the things you can do with the program. Their are a lot of things to read, but sometimes action speaks louder than words. This is one reason why I really like Michael Pizzolla, Dave Schwartz, Rdss, I've seen some JCapper features too, because they will show you how things work. Would it really be hard to show all of the things the robot could do on a video?

[JimG]

Capper Al,

Depending on what you are looking for, I think JCapper, HTR, or HSH could fit you since you are looking for comprehensive commercial software with research capabilites. Beyond that, it is a matter of which feels right to you. There are winners and losers with all three.

Jim

[Capper Al]

Thanks for your replies.

[nat1223]

couldn't agree more! the typical computer software programs on today's market will not produce long term profit. those inferior programs are trying to compete with larger databases created by groups of computer gurus! 40 years ago you could have done that. I did for a while before I went out into the real world of business. Today you are fighting whales with lots of money and people and huge data base programs and super fast computers that access that you don't have. I believe you have to find stuff happens so rarely they haven't caught it. The easy stuff, they bet off the board. Using Speed Ratings every which way imaginable falls into the easy stuff for the whales. But hey, it is your life, just don't waste too much of it beating your head against a wall.

[sam4022]

Being a programmer / system analyst for a fortune 100 corp for many years, I could design full user run systems under only Excel, followed by being the Director of Advance Technology with free access to almost NSA powered mainframes I have learned the following:
1. You need to segregate your target populations by distance, speed or pace, earnings, jockey/trainer, last race data, dirt, turf, etc.
2. Develop test populations, usually thousands of races each, for what appear to be the most promising conditions (won last race, etc.).
3. Make some educated assumptions (maybe eliminate races with grey horses having odd numbered saddle cloths).
4. It takes many months/years and thousands of hours of computer time to 99% of the time reach a dead end or perhaps a play once every 1,154 races at a fair grounds meet.
THEN.....
5. Pick yourself up and start over.....and over....and over.

Having developed two (2) decent performing systems (T.I.P.S. and SOS) your time is better spent reading the boards like this one and being with your family. I did enjoy developing the systems but I really enjoy playing with the grandchildren much more. But as an exercise in mental agility and dogged persistence it's second to none.

[nat1223]

Being a programmer / system analyst for a fortune 100 corp for many years, I could design full user run systems under only Excel, followed by being the Director of Advance Technology with free access to almost NSA powered mainframes I have learned the following:
1. You need to segregate your target populations by distance, speed or pace, earnings, jockey/trainer, last race data, dirt, turf, etc.
2. Develop test populations, usually thousands of races each, for what appear to be the most promising conditions (won last race, etc.).
3. Make some educated assumptions (maybe eliminate races with grey horses having odd numbered saddle cloths).
4. It takes many months/years and thousands of hours of computer time to 99% of the time reach a dead end or perhaps a play once every 1,154 races at a fair grounds meet.
THEN.....
5. Pick yourself up and start over.....and over....and over.

Having developed two (2) decent performing systems (T.I.P.S. and SOS) your time is better spent reading the boards like this one and being with your family. I did enjoy developing the systems but I really enjoy playing with the grandchildren much more. But as an exercise in mental agility and dogged persistence it's second to none. single programmer trying to do this would never have time to make a bet and be profitable.

[sam4022]

Hey Nat1223,

There were a minimum of 3 of us and as many as 5 and it was still difficult. But you are definitely correct them days are gone forever.

[SteveeBee]

the typical computer software programs on today's market will not produce long term profit. those inferior programs are trying to compete with larger databases created by groups... My apologies in advance if this has been discussed before:
Meet player and owner Dana Parham who bets 300 million per year on North American racing including small time tracks.

Interviewed by Meadowland's host Ken Warkentin.

Part, 1 - http://www.youtube.com/watch?v=AKqzD3jdBvc

Part, 2 - http://www.youtube.com/watch?v=Uf_Yno4Jg-8

[redeye007]

Bear in mind that the $1 BRIS file is missing the two most important and critical factors that lead to success in computer handicapping, BRIS Class Rating and BRIS Race Rating. The absence of these two factors can be directly attributable to the dismal performance of software that uses the $1 file. :)

[Tom]

Who says the $1 files lead to abysmal results?
I find them to part of a few very successful programs that do not use the class ratings at all.

[Fingal]

So do I. What works for each person is subjective, but sometimes more is less.

Albert Einstein said it best-

Not everything that can be counted counts, and not everything that counts can be counted.

[dkithore]

Bear in mind that the $1 BRIS file is missing the two most important and critical factors that lead to success in computer handicapping, BRIS Class Rating and BRIS Race Rating. The absence of these two factors can be directly attributable to the dismal performance of software that uses the $1 file. :).

A fine observation.
Do you know which one is more important of the two. I tried to search the definitions but they confuse me more. Can you clarify how to interprete with some clear cut example. appreciate it.

[thoroughbred]

Capper Al

Your post shows an interest in the basis and inclusiveness of handicapping software.

This link shows the mathematical basis behind the CompuTrak program and will provide you with a view of its broad aspects.

I apologize, in that I believe this was shown here some years ago, but it may be appropriate to your query.

http://www.revelationprofits.com/docs/Engineering%20Analysis.pdf

Disclosure: I am the creator of the program.

[Pensacola Pete]

Do any of the whales make money anymore? There are so many of them that it would seem they step on each others' toes.

[DRIVEWAY]

My apologies in advance if this has been discussed before:
Meet player and owner Dana Parham who bets 300 million per year on North American racing including small time tracks.

Interviewed by Meadowland's host Ken Warkentin.

Part, 1 - http://www.youtube.com/watch?v=AKqzD3jdBvc

Part, 2 - http://www.youtube.com/watch?v=Uf_Yno4Jg-8

Points of Interest

1.) He knows of 5 other wagering groups similar to his(total 1.8 billion)
2.) He has 20 employees involved
3.) They target their odds line i.e. a Parham 7-1 that's at 9-1 odds with 1 second to posttime receives a wager sufficient to drop odds to 7-1. This tells me that they are searching for a breakeven on the wagers and live on the rebates.
4.) Wants reform of breakage methods - especially on show.
5.) Would love a betting exchange. They could post odds that ensure them of a small profit or breakeven.

Conclusion
The game is about better information, instantaneous access to the pools and REBATES.

[Capper Al]

Capper Al

Your post shows an interest in the basis and inclusiveness of handicapping software.

This link shows the mathematical basis behind the CompuTrak program and will provide you with a view of its broad aspects.

I apologize, in that I believe this was shown here some years ago, but it may be appropriate to your query.

http://www.revelationprofits.com/docs/Engineering%20Analysis.pdf

Disclosure: I am the creator of the program.

Your volume of work is similar to mine. Lots of it. Thanks

[thoroughbred]

Your volume of work is similar to mine. Lots of it. Thanks

Capper Al,

This is interesting. I'd like to read it. Do you have a link, or a write up for viewing.

Thanks.

Thoroughbred