Howard was ahead of his time

[Tim Y]

I have read almost every book that was suggested as a corollary to understanding the essence of the methodology from the Follow Up. Many, after reading them the first time around, didn't seem relevant to horse racing, so over the years I have re-read many of them in the later context of understanding the methodology better.

Sartin forever suggested all of us forget the horsey talk of class structure, the effect of the trainer and rider, and ONCE you put all of that into the dust bin where it belongs. you have ONLY the program left. Most people, and he mentioned this repeatedly, FIGHT in retaining these horsey ideas and never are able to just let go and let the program do what it is designed to do.

Put alcohol, water and vegetable oil in a glass. Shake it up and the let it sit. What happens: it self organizes into three layers of differing density. A race is very much the same way: enter a bunch of lines and it organizes into the race structure all by itself.

I have followed up on the school of chaocomplexity at the Santa Fe Institute that Howard Sartin enlightened me to so many years ago. They have found that in what appear to be chaotic systems, an emerging self organizing order over time.

What is the program other than a vehicle to find that emerging order? Throw in 10 lines and see what the self organizing system (the program) spits out. Don't try to overlay the illusion of some other "order" i.e. class structure the effect of a work out on racing or a trainer.

Use horse races to allow that order to appear. That's it. Use what Sartin predicted would work all those years ago and FORGET THE REST!

[For The Lead]

Originally Posted by Tim Y

Sartin forever suggested all of us forget the horsey talk of class structure

Nice article, Tim. However, with respect to class structure, you will always get an argument from me. Moreover, I can support my argument with facts based on actual race results. In that regard, I would venture a guess that I have more information available to me than Howard did. I seriously doubt that anyone had "millions" of horses race results at their fingertips back in the '80's, with the exception of perhaps the Daily Racing Form.

Can a cheap claimer beat an allowance horse? You can bet it can, but that is a "trick" question. And in order to understand the "trick", one must first have a working knowledge of horse racing and it's class structure, whether one likes it or not.

The "trick" is this.
Once a horse breaks its' maiden, the most logical next race is one for non winners of a race other than maiden or claiming (NW1X) or non winners of two races lifetime(NW2L). Both of these types of races are available in the allowance category, however, only one puts all the entrants on a level playing field and that is non winners of two races lifetime. They are on level ground because all the horses have won only one race, their maiden victory.
In the allowance category, most people would view this as one allowance horse beating other allowance horses. This is an error.

Then there is the allowance race for non winners of a race other than maiden or claiming(NW1X). By the way, this is more common in the allowance category than a NW2L condition. In this race, you can have the same allowance horses that have only won their maiden races and just keep running in this condition because their owners do not want to drop them down into the claiming ranks for fear of losing them through the claim box. These horses can be "1 for 10", "1 for 20" or whatever. Are these legitimate allowance horses? NO! And they should be placed in claiming races and allowed to find a level at which they are capable of competing and even win. It is in these "allowance" races where a claimer can beat an allowance horse, because the "1 for ?" allowance horses in these races are really NOT allowance horses. They are "1 for ?" plugs. A claiming horse, who is also eligible for this race, could have 10 wins, 20, wins or 30 wins as long as they are all claiming race wins and, of course, its' maiden win. Now I ask, which is the better horse, the "1 for ?" who just keeps running in these races and never winning or the claimer that demonstrates the ability to beat fields of horses who also have the ability to beat fields of horses? Personally, I'll take a horse that has demonstrated the ability to beat other horses over a horse that has demonstrated it CAN NOT!

So, in the above case, you can put in a line for an "allowance" horse and a line for a "claiming" horse and if the line for the "claiming" horse shows it can beat the "allowance" horse, then it probably will. However, if you try and convince me that one of these claimers could have beat SECRETARIAT, good luck! I don't think even YOU believe that!!

Does this mean that the "highest class horse" wins every race? Of course not. But facts are facts and the facts are these; under the current class structure of horse racing, the majority of races are won by horses that are either dropping down in class or remaining at the same level as their last race. And as you would say, "PERIOD"!

[Bill V.]

Thank you so much
Your post is very helpful.

[Tim Y]

One can overlay a perceived order that has no cause and effect and accept it as so. Shown over and over again in many disciplines.

Horses run...within the bounds of a match up. Ability versus ability no some superficial subjective evaluation selling pecking order.

[Tim Y]

These are some of the psychological traps people enter when trying to understand the whole by piecing together the parts and then, of coure, ONLY looking at data that CONFIRMS what they were looking for in the first place.

Of these the ones most commonly associated with the MANY fallacies of horse racing is the Texas Sharpshooter fallacy.

http://en.wikipedia.org/wiki/Texas_sharpshooter_fallacy
http://en.wikipedia.org/wiki/Confirmation_bias

In 1897, Tolstoy wrote:

I know that most men, including those at ease with problems of the greatest complexity, can seldom accept the simplest and most obvious truth if it be such as would oblige them to admit the falsity of conclusions which they have proudly taught to others, and which they have woven, thread by thread, into the fabrics of their life.

http://en.wikipedia.org/wiki/Availability_heuristic

[Tim Y]

I recently saw a very interesting chart showing the fallacy of the database using historical data to predict current data.

The example was a graph showing the spectrum of investment return, from highest to lowest, of stock analysts for a five year period.

The 2nd example was a graph of the SAME analysts, placed in the same order as the first graph, showing THIS TIME their five year placing (up for positive and down for negative) in that next five years. The graph was all over the spectrum with no correlation to the first rankings.

If these analysts were so good at their job, is should have been a mirror of the first graph. It was pure noise with NOT correlation. RANDOMNESS

[For The Lead]

Tim,

Neither you nor I structured these class levels, "horse racing" did, we just have to live with it and deal with it in the best way we can.

[For The Lead]

Originally Posted by Tim Y

I recently saw a very interesting chart showing the fallacy of the database using historical data to predict current data.

I have 12+ years of race results that tell me horses that are either returning to the same class level as their last race or a lower class level than their last race win 65% of ALL races.

Funny how this information remains stable from one year to the next. Perhaps that is because there are such large numbers in each and every year. (40,000+ winners in each year) I didn't bother to do the math, but my guess is that the standard deviation would be + or - 1%. This means that at the end of any given year, one could have predicted the the results for the coming year and been accurate within 1% either way.

[Tim Y]

Originally Posted by For The Lead

I have 12+ years of race results that tell me horses that are either returning to the same class level as their last race or a lower class level than their last race win 65% of ALL races.

Funny how this information remains stable from one year to the next. Perhaps that is because there are such large numbers in each and every year. (40,000+ winners in each year) I didn't bother to do the math, but my guess is that the standard deviation would be + or - 1%. This means that at the end of any given year, one could have predicted the the results for the coming year and been accurate within 1% either way.

Texas sharpshooters fallacy. It is alive and well. Superimpose a belief system over the reason that woman's hosiery run and you can well believe it. Correlation maybe, NOT cause and effect.

p. 173 The Drunkard's Walk : "When we look closely, we find many of the assumptions of moderns society based on shared illusions. ...I shall approach the issue from the opposite direction and examine how events whose patterns appear to have a definite cause may actually be the product of chance."



Occurs much more often that we think.

[Tim Y]

Originally Posted by For The Lead

Tim,

Neither you nor I structured these class levels, "horse racing" did, we just have to live with it and deal with it in the best way we can.

That system is, and was, designed to buy and sell. It never was defined to establish what horses can do what.

These are the direct outcome and remnant of "selling races" which are still alive and well in Europe. Read a little history of the game beyond the last decade and you will realize that.
http://www.flatstats.co.uk/horse_race_information.html