"The new software that breathes fire into the Beyers..."



David K. Don, PROBER's president, grills his top programmer, F.G. Sallustro about WIRED 4-SPEED. What follows is a brief sample from a book to be published by PROBER concerning WIRED 4-SPEED and the Beyers.

DKD
Do you always enter a horse's last three races, even when they differ from today's race?
FGS
Yes. Always.
DKD
What do you truly think of WIRED 4-SPEED's approach?
FGS
It's wonderful. Simple and direct and accurate.
DKD
Why? What makes it wonderful?
FGS
It makes no pretensions. It ranks all the horses from first to last without hesitation. So many "systems" these days are really data-organizers. What they do is present data for the user to evaluate, usually very beautifully, but without answering the central issue: Who is the most likely winner?
DKD
But why concentrate on the likely winner? Haven't you come to know the real issue is value and not picking the winner?
FGS
That's nonsense. I suppose it comes from mistaken probability notions.
DKD
When you say "most likely winner" are you referring to impact values?
FGS
Most definitely not. I wish the whole discussion of impact values would disappear forever. Really, there's no place in handicapping for that approach.
DKD
What about more accepted prediction systems like regression analysis?
FGS
Useless! And misleading, too.
DKD
What approach meets your approval?
FGS
That's easy. WIRED 4-SPEED--and any other method that's based on handicapping experiences. Not data expeditions, but experiences playing real races for many years. Of course, it's not that easy to distill the "laws" that underlie the winning horse, but that doesn't mean it can't be done successfully.
DKD
You're a mathematician?
FGS
Yes, but I've worked in statistics since graduate school.
DKD
So, how can you not use mathematical-statistical models to uncover handicapping rules.
FGS
Please understand that I'm certainly not against using any technique that works, but the models that you mentioned, impact values and regression, have very limited usefulness in horseracing prediction systems.
DKD
Why?
FGS
Essentially, those techniques, like others, have implicit mathematical requirements that are not met by horseracing analyses. In fact, the assumptions are usually violated--which dooms the results.
DKD
What about computers?
FGS
Very useful. For me, essential.
DKD
Yet there are many handicappers who refuse to use one.
FGS
Yes. You're one such handicapper.
DKD
That's right. I've yet to see any advantages.
FGS
Pity. Perhaps I should program your artful approach!
DKD
I can't afford to hire you! (FGS laughs nervously) Besides a computer, what else do you find essential?
FGS
The Daily Racing Form. And a red pen.
DKD
How about services that compute speed ratings or track variants?
FGS
I suppose they could be useful for some, but not for me. All I need is my computer and The Daily Racing Form. Especially since the DRF incorporated Andy Beyer's speed numbers. The Beyers are the single most crucial data in the Form. Absolutely and without question.
DKD
That's quite an endorsement.
FGS
I don't think my endorsement matters to anyone.
DKD
You're being modest.
FGS
No, not at all. Seriously, I wouldn't play the races without access to the Beyers.
DKD
Yet there are many who think the Beyers are misleading.
FGS
Is that so?
DKD
Yes. For example, some claim that a high Beyer earned in the mud overstates a horse's speed potential.
FGS
I don't understand what you mean by overstates.
DKD
Suppose a horse typically earns Beyers of 80s, but then, one muddy race day, runs to a 95. Doesn't that 95 mislead a horseplayer?
FGS
Absolutely not. The race effort was valued at 95. Remember, the Beyer pertains to the race, not the horse. Admittedly, that's a fine distinction, but it's clearer that way. At any rate, I hope you wouldn't expect Beyer to adjust any figure downward merely because it was too fast for that horse. That's tinkering.
DKD
Isn't that what really happens all the time?
FGS
I don't accept that as valid. Listen, how the Beyers are derived has been clearly outlined by Andy Beyer himself in his books, nowhere more eloquently than in "Beyer On Speed". I really think that in the right hands understanding the Beyers is the light and the way!
DKD
Are you putting me on? Please explain.
FGS
No, I'm very serious about the Beyers. I'm also aware that many handicappers cannot get past the value of a Beyer and imagine what it could be. As a mathematician, I naturally love numbers and I know how the Beyers are made; as a handicapper, I must imagine that a horse might improve by 6+ points given favorable circumstances. But as a handicapper it's my job to imagine; not Andy Beyer's job. I want his figures to reflect every horse's efforts--no more and no less.
DKD
What about declining Beyers?
FGS
What about them? It's the handicapper who must learn that a Beyer earned last week may change today. That's what makes this game fascinating. And believe me, if you examine a horse's recent Beyers, and keep asking yourself about that horse's form today, right now, you will find yourself imagining what its next, future Beyer may be!
DKD
Do you really think that approach leads to winning?
FGS
Yes. There's no other way! The Beyers reflect all the mathematical adjustments needed: Distance-to-Distance, Track-to-Track, and Daily Track Variants. By the way, who could possibly do all this work better than the DRF? Handicappers should sharpen their imaginations and go beyond the data.
DKD
And WIRED 4-SPEED embodies what you've discussed here?
FGS
Yes. Imagining makes the difference between winning or losing. If you don't go beyond the data, you pick what everyone else sees. As the WIRED 4-SPEED programmer I had to translate, into computer code, handicapping art and science. And I needn't tell you that it was the art that proved challenging.
DKD
Is there anything else you would like to discuss?
FGS
On horseracing?
DKD
Yes.
FGS
No, not today.