A. I. programs

[Jaguar]

Jackad, I would add that any program from David Schwartz would be strong. He holds the distinction of producing the first really powerful A. I. prograim, Thorobrain.

I would also mention the latest version of All-Ways, an amazing program, and enormous in its complexity, except that the full version is so costly that I can't imagine anyone- in this day and age- buying it when other powerful discs cost so much less.

Cost is definitely a consideration concerning RaceCom, as well, but the one somewhat ameliorating factor might be that RaceCom can be accessed on the web- via credit card- on a sort of per diem basis(if I understand their current policy correctly).

But, certainly between $250.00- $600, there are super programs out there.

In handicapping, as in romance, beauty is in the eye of the beholder. Probably die-hard pace handicappers would shun the type of comprehensive A. I. programs I buy.

While I, when a friend shows me his latest genius pace program at the Teletheater, experience a feeling of light-headedness, quickly followed by mild nausea and gastroenteritis- followed later by waves of contemptuous hysteria. All the rest of the day the words: "He paid $450 for that piece of crap." echo through my mind.

You might want to check out Joe Zambuto's program as well, what a smart man- and a superb handicapper.

All the best,

Jaguar

[JimG]

Jaguar,

What is your personal software of choice? Out of curiosity, have you taken a look at Equisim? If so, what are your thoughts on it?


Jim

[Jaguar]

JimG,

My software of choice is an old A.I. program no longer on the market, called Multi-Strats. I do not use the program out of the box, rather, I have configured the 25 strategies the program uses to rate horses by setting the weights(or statistical impact values) of the strategies according to the values I have derived from using Multi-Strats to analyze thousands of races.

Curtis Martin gets the credit for developing this fascinating and powerful program over a 15-year period, culminating in the "Titan" version, which was issued in '98 or '99.

I do not handicap claimers or maidens, only allowance and stakes events.

Curiously, the program was sold with no hint as to pre-setting the strategy weights, the user had to discover their own values by entering winners and slowly and painfully allowing the program to configure itself.

Another anomaly, the program merges distances- such as 6f, 7f, etc., for sprints- in its analysis. The program does the same thing with route distances, whereas Multi-Strats should have been designed to measure discreet distances, that is, all 6f races should be compared only to 6f and all 1m races compared only to new 1m races, etc.

However, several glaring statistical oddities notwithstanding, Multi-Strats has unique strengths, such as the capacity to accurately measure a horse's connections(trainer-jockey patterns), and to determine a horse's form cycle, and to analyze speed and pace in a very effective way(the sprint ratings are the best I've seen).

Often, Multi-Strats will give a high rating to a horse which the pace programs downgrade, and when Multi-Strats does that, the horse is usually a double digit value in the win pool.

If I had to recommend one program to a handicapper, it would be Thorobrain5. Jim love took David Schwartz's brilliant creation and made it slicker and easier to use. Happily, the program comes with some models provided.

In my opinion, the guys with the best models get the money, because programs like Multi-Strats and Thorobrain tell you when to go in and when to sit on the sidelines. For example, I don't bet unless Multi-Strats tells me I have a 77% or better chance of winning the race. I'm looking to cash a ticket, I'm not looking for action. If I don't get more than one pick today- at a given track- I don't worry, because tomorrow I may get 3 picks and I will always take the best of it.

Regarding Equisim, which I have used and studied: Brilliant concept, brilliant execution. But, too limited in scope.

Pace analysis just ain't good enough, Rocky. These days it's the connections that count. We need the patterns, the models- to get the money.

To paraphrase Marlon Brando in On the Waterfront, "All the time it was you, Amylcar, all the time it was you."

Just watch the pace handicappers staring dourly at their laptop screens after their program's big pace horse finishes up the track, and take a look at the losing tickets on the floor around those pace handicappers. Lots of luck, fellas.

Hey, when a horse lays off 15 months, comes back and wins in a walk-over, does that tell me something. Yeah, it does: it's the connections guys- . "See what I mean, Vern?"(Jim Varney, you were great, RIP).

Best of luck!

Jaguar

[JimG]

Jaguar,

First of all thanks for your well thought out reply. I appreciate you taking the time.

Regarding pace horses finishing up the track...that is true. Of course, big class, big fig, or big trainer connection horses finish up the track as well. You seem to enjoy impact values. I do to. Problem is the program I would like to use with them is All-Ways which charges minimum of $7 a pop for their datafile. For making models, that makes the cost out of the question for all but the very wealthy or those who just want to play one track. If they ever went to unlimited, I would jump there in a minute.

Based on a couple of things you have said on this board recently, I wrote Thorobrain and they gave me some reports to look at for todays races. I'll try to give some initial feedback on TB later. One thing we do agree on...Dave S. is extremely smart and any program he designs is not for the faint of heart. Much time is needed to digest the nuances of a program that Mr. Schwartz is involved with.

I've heard of multi-strats but never used it. Sounds like your having good success with it.

Don't sell Equisim short, the programmer Nathan seems very dedicated and I predict it will get better and better. It takes into account connections when tuning simulations. One thing you said I would like in a program though...is one that has confidence levels so it can help guide you WHEN is a good time to lay the money down.

Thanks again for your reply.

Jim

[jandrus]

thats how you really field about my new progams thanks a lot

[Jaguar]

Truck, you old Marine, you know I wasn't referring to you- but you do know what's what.

Once you educate me about the Equisim upgrade you mentioned to me at the old stand- is it out yet?- I can take a look at the "connections" factor mentioned by one of our fellow handicapping enthusiasts on this board.

Were it not for the eagle eyes here on PA, I would not even have been particularly aware of that factor, and I would have had to wait for you to detail it for me.

By the way, there is murmuring in the back halls about a potential buyout of MS "Titan" in its current guise- and of re-doing it for Windows. -What a monster task that would be. It's alot of code in DOS. In Windows it would be much bigger.

I am informally advising about kludges within the program, such as combined distances, etc., etc., -all the old stuff we hate- so maybe those problems will be corrected in a newer version. -Let us keep our fingers crossed. - Just give me strategy 16 in the new program and I'll be a happy camper.

If, wonder of wonders, anything firms up, I'll E-Mail you. But, I
suppose by then the PA sharpies will already know about. And, knowing you, you early adopter, you will probably have a beta version installed on your laptop.

Also, since you own even more software than I do, kindly keep me clued in on any dynamite new programs which come your way,
since they always do wind up on your hard drive.

Hope we hear something positive about MS soon, too bad Dr. Sengbush isn't involved.

All the best, old Pal-

Jaguar