Articles Home : Articles : Biases, Systems and Independent Trials
Biases, Systems and Independent Trials

As you may know, I'm not a believer in track biases beyond weather and/or track structure kinds and even then I don't fall into the sequence of independent trials trap, as you will see. I've seen no evidence of a "subtle" track bias in all my years of race gazing. Even if I thought I saw one (I don't drink at the track anymore), other factors in the race could stand in as an explanation once the fog lifted. I guess parapsychology folks run into the same deal.

Steve Klein's Bias Ratings that he rolled out for us in his new book, The Power of Early Speed, started all of this rigmarole. So today I'm going to try to explain to you the reasons that Klein's ratings are not only naive but dangerous to your bankroll. So please indulge me for a bit as I try to show you how we got to this point. I'll briefly touch on thinking modes, systems theory, and probability theory as it relates to the Gambler's Fallacy. I'll try to make it painless, even fun. Some of what follows is silly but underneath there's some very serious stuff.

Thinking Modes:

We are linear thinkers for the most part and we learned the easy way; we were programmed, conditioned by institutions, from the Ivy Leagues to the Ghetto Leagues. The main character in this charade was an ole crusty pajama-wearing thinker named Aristotle who assured us that we live in a Yes and No world - it caught on, big time, bigger than the pet rock. There aren't many gray areas in Aristotelian thinking; it's simple and that's popular. The charade turned nasty when Descartes, who was obsessive compulsive about chopping things up into parts, entered the programming curriculum. He gave us the notion that we are in "here" and everything else is "out there", which is a handy belief when you need to blame something so you feel a little better; he was a "parts" man. Yes/No, Me/That attempted to explain our complicated world. It did all right for good while. You might say Aristotle and Descartes were the jamming Reductionism Brothers. They had everybody so wrapped up in their ear-worm that guys like Chuang Tzu and Loa Tzu (systemic thinkers) were delegated to the institutions' creepy poetry classes.

This Aristotelian-Cartesian thinking habit is now in upgrade mode because our world has become much more complex than back in the day. So has horse racing; ever seen a 1940s Daily Racing Form? Bela Banathy of the Society for General Systems Research states that a reductionist type of viewpoint is not able to understand "wholeness" which emerges from the mutual interaction of parts, where the part gets its meaning from the whole and by its interaction with all the other parts of the whole. The properties of the whole cannot be seen from the viewpoint of the parts. The Reductionism Bros would be giggling their heads off or foaming at the mouth over that one. True believers in subtle track bias ratings are doing a little giggling themselves right about now, I'm sure.

Well, in the first part of the 20th century we entered the quantum age with the Copenhagen Theory, Heisenberg's Uncertainly Principle, and later on with General Systems Theory. Chuang Tzu and Loa Tzu might have their day yet. As my favorite pop-philosopher, Robert Anton Wilson, says, "We have absolutely no guarantee that equally important and obvious facts are not screened out of our perceptions by our current dogmas". There's a doggie out there and his name is Bias, I think.

So today we realize that the ole (Banathy once again) reductionist method has to be complemented with synthesis and with expansion, aimed at understanding larger and larger wholes in which our systems of interest are embedded. In other words, it just ain't that simple.

Systems Theory:

The model that works for me, as I seek to understand a horse race, is General Systems Theory - or you might say, systemic thinking. Dr. David S. Walonick of StatPac defines GST: System theory is basically concerned with problems of relationships, of structures, and of interdependence, rather than with the constant attributes of objects, and if I may add, the parts. That's important as you will see.

So we can extrapolate a bit and say a horse race is a system made up of numerous components/factors. These factors are the horses themselves, running styles and the dynamics of the match-up of those running styles, form cycles, class, pace ability, pace of the race velocity, final time ability, jockeys, trainers, track structure, the weather, and, yes, the track itself, the dirt....and many more which might even include the starting gate and its crew - down in Florida there's even been some talk about the tides getting in on the act. The race/system is complex. The components/factors are interdependent. See where I'm going with this?

Probability Theory Informs:

Since the outcome of one race/system does not affect the outcome of another race/system, probability theory informs us that races are independent events/trials. This is important because it plugs into the Gambler's Fallacy, which will help us put this thing to bed.

Let's summarize what we have so far:
Smartasses Aristotle and Descartes were the primary characters who gave us our habit of linear thinking thanks to the insistence of institutions - the separate parts of our yes and no world, the Klein approach. Systemic thinking, the Tzu Bros, wouldn't be caught dead on a straight line A_____________Z. As our world became more complex we needed new ways to explain the complexity, which evolved into quantum physics and the philosophical implications thereof. Was that a wave or a particle and how did it get from "here" to "there" instantly? The Yes/No, Me/That kind of thinking won't get 'er done on that one. Anyway, this "wholeness" way of thinking helps us understand complexity as systems and the interconnected and interdependent factors of those systems. Probability theory informs us that a horse race is an independent event/trial. That is to say that the outcome of one event does not affect the outcome of another - like flipping a coin. That $5,000 claiming race has no affect on the outcome of the next Allowance race. Each pick six ticket race you mark lives in its own world, you might say.

We're just about there........

Let's, now, define Gambler's Fallacy: The gambler's fallacy (Roney & Trick, 2003) is the propensity to perceive a series of independent trials as a continuing sequence or as a whole block. Wow! An example of this would be if a series of coin flips turned up heads several times in a row and one expected the next independent outcome to be different due to previous outcomes. Those who believe in the gambler's fallacy are more likely to continue to gamble/make bias ratings even if they do not have the funds to do so or an edge on the game. They perceive the independent trials as a whole and believe that their luck/bias will change; given previous outcomes, a particular outcome is more likely, because everything should balance out. The Gambler's Fallacy presents us with two extremes: It can ruin your day or it can lead to pathological gambling - or even worse pathological dirt watching.

And now we have arrived at the end where thinking modes, systems, and probability theory converge along with a punch line:

Mr. Klein would have you believe that complex systems of independent events -remember all the components in a race/system and the variation of strengths and weaknesses of those dynamic handicapping factors that we all struggle with and analyze everyday? - will help you win future independent events, and has provided a simplistic results chart system, expressed as a number, to do just that! No consideration is given to the many interdependent handicapping factors/components of the system of which the dirt is only a part. Say what!? Descartes would love him. A high school probability teacher might make him stay after school.

Since we know races are independent events, you might say he flipped the coin ten times for a "race day", seven times it landed on heads and called it a heads bias - and don't forget to record it in your notebook. He tried the same thing with Breeder's Cup trends: Independent events of yesteryear can help us win money with independent events of today, he says. How many other folks sell "stats" of this kind in our industry. Could you name a big one?

Here's the punch line:

The Klein Bias Ratings should be understood to be the Klein Gambler's Fallacy Ratings. Even if none of this were true, it still behooves a player to consider ALL components and the strengths and weaknesses of those components in THAT race/system, THAT independent event with an unbiased opinion upfront so as to not fall into the big bias, confirmatory bias. It's the biased dirt he says that affected the day's results (sequence of independent trials), and he wants you to believe, like the gambler's fallacy, you're due (horse ran against a bias so, well, you know) in some future independent event.

"The way the track was playing" is akin to saying the way the coin flips were playing. Whatever happened in one race (please remember all the many components/factors), and that includes the dirt, has no affect on the outcome of the next. This is tricky stuff; we can easily be fooled into believing the fallacy. Write it on card and keep it handy: The gambler's fallacy is the propensity to perceive a series of independent trials as a continuing sequence or as a whole block.

Even if we could draw a trendline, expressed as a number like Klein did with his bias ratings, across independent and multifaceted events, that trendline must represent the multitude of handicapping factors - and that would consist of quantifying factors like form cycles, which is a difficult proposition - if we're to come to some kind of comprehensive understanding of the race outcome so as to help us predict future outcomes. Anything less would create a distorted picture. Once again, the Yes/No, Me/That kind of thinking won't get ‘er done.

So as you turn the page of your PPs to the next race, you are confronted with another independent event. There's never been one exactly like it - which means the handicapping factors under consideration have never been weighted exactly the same. Fresh start to work your magic.

The Klein Bias Ratings are the most blatant example of the Gambler's Fallacy I've seen in years, and it has the DRF Press, America's Turf Authority, stamp of approval on it. And that's a cause for concern. But I'm sure somebody was just out to lunch when the decision went down to proceed on this one. It happens.

That was fun and I hope you enjoyed it and I appreciate the indulgence. If we can't make racing fun then there's no point in doing it.

But before we end this thing, I'm going to present to you, drum roll, the Giles Pace Velocity Bias Ratings. I'll use the information from the charts in the book, the one at Churchill Downs, May 27, ‘05. (For entertainment purposes only. Do not try at home or track)
Here's how to calculate the Giles Pace Velocity Bias Ratings

1. Generate the pace of the race velocity rating for all dirt races for the day. Some will be - (slow) and some will be + (fast)
2. Add those together (remember the positive and negative numbers)
3. Multiply the result by 10
0 (zero) represents a bias free pace of the race velocity day
A plus (+) result represents a biased day on the fast side
And of course, a minus (-) result represents a slow day bias
When the rating exceeds +50 (fast bias) that's good and that means that was really a tough day.
When the rating falls below -30 (slow bias) that day was really easy on the horses.

When horses come back and the rating tells you the day was hard, really fast like +60 or something, then you'll know to upgrade his chances in today's race. If the rating is in the minus range then you'll know the horses had a very easy day of it and so you can downgrade its chances in today's race.

Here are the pace velocity ratings for the eight races at Churchill:
-1
+1
-1
0
+6
-6
+3
-1
Result: +1 x 10 = +10

That day wasn't much of a fast biased day; the horses were playing close to par for the day.

There you have it. Buy some notebooks at Walmart or somewhere and you'll be ready to kill the races.

Do you see the foolishness of all this bias rating business across independent events?

All right, we've had some fun but now it's time to get serious......If The Power of Early Speed happened to be a car it would be recalled for defective parts. As you have seen, Klein's bias ratings are profoundly false, and naive. Will the Daily Racing Form make a retraction for this blunder, unfortunate misinformation? No. Status Quo is a tough nut to crack. Therefore, DRF is now involved in disinformation. So be it.