Tuesday, 1 March 2011

Equine Flow Psychology II

"Horse racing analysis turned from an art to a science in the 1960s and 70s along with such unrelated, former academic arts such as sociology and anthropology.

"It was the spirit of the times: physicists reduced from atoms to quarks; biologists reduced from cells to electrochemistry. So, handicappers reduced from hot tips to speed figures and track variants.

"Handicappers borrowed from science, and scientists became handicappers."

The above quote is from Handicapping Speed by Charles Carroll. The writer uses the term "handicapper" in the US sense, as we would say "punter" in Britain and Ireland.

It still represents the most exciting thing I have ever read about horse racing. At once, it changed the direction of my life. It gave me the belief that "our own little corner of the universe", as Carroll puts it, was a worthwhile study and a justifiable source of intellectual fascination.

But, the new-found zeal to apply scientific rigour to racing soon ran into trouble. The closer that racehorse performance was analysed, the more difficult it became to understand.

Of course, the appreciation of racing as a pastime does not require microscopic examination. Even the most basic punter believes he has a working understanding of the sport. 

And, if things confound them? It isn't his fault. It's larceny.

Those described by Carroll who burrowed deeper into racing were inevitably drawn to the subject of speed. First, to race times themselves, then to figures evaluating the merit of their achievement and finally to sectional times.

The trouble for some of us is that we don't know where to stop. The study of anything becomes a fascination of its own and the obsession can be increasingly with the abstract, the theoretical, framed by the predictable past.

The trouble with reality is that it is dirty, messy. Fundamental rules don't always work in real-world settings, especially when data is extracted from a chaotic environment like a horse race in the attempt to predict future events.

And that's exactly what I found about sectional times. It is one thing understanding complex events of the past, but it is very much another projecting their future impact.

When I came across horses who had run their races in a hopelessly inefficient manner, I thought I knew that they had to do better next time.

And many of them did. The fundamental law of energy use - that optimal time depends on ideal pace - seemed to hold in so many cases. Pleasingly, horses I studied - and even bet on - did indeed improve their form when running closer to an even tempo.

But the exceptions were always there. And, just as scientists have always done, I tried to ignore these outliers as meaningless "noise", to dismiss them from the sample as if they had never happened.

That was until I read about Chaos Theory. I learned that when scientists finally stopped trying to bend intractable data to their inflexible will, they noticed that the exceptions did indeed "prove the rule". 

In other words, it was possible to learn more about the physical relationship you were studying by looking at the examples which did not conform rather than those which did.

And, when I looked at the noise in my data - the racehorses who ran inefficiently, who needed to run inefficiently - I made the biggest leap in breaking down the complexity which so confused me.

In trying to understand racehorses, I was making a mistake in trying to hold them all to the same ideal. I was looking at them as if they were insentient beings of uniform physical identity, as if you would racing cars.

I had assumed that energy supply was constant until the tank ran dry. But, this is a big mistake. Racehorses are not like that at all.

Instead, my hours and hours of work led me to a new understanding of the racehorse. I began to appreciate the relationship between how fast a horse runs and how they want or need to do so. 

Where before I had considered the psychological element of performance to be woolly and unimportant, I saw it was of prime importance to the science of sectionals. It wasn't some nebulous concept of just those who love the horse as an animal and like to remind us that "they aren't machines".

No, it is inexorably bound up with interpreting their measurable physical capabilities. It can be inferred from data and the data itself cannot be understood properly without it. 

It is why they run fast, it is concerned with the psychological mechanisms that empower optimum performance.

You can call it what you choose. I call it Equine Flow Psychology. And, of the many performances which show it best, why not use one which we have already discussed? Why not use the greatest of them all?

Secretariat. Every discussion about the 1973 Belmont Stakes has been concerned with the brilliance of his performance, the startling nature of his 31-length victory, the unparalleled achievement which was a mile and a half in 2:24.

But, aside from every well-chosen phrase which was generated by the occasion, aside from the way it released a deep-seated electric charge into the soul of sporting America, is one rather large, inconvenient truth.

Secretariat put up the greatest measured performance in the history of racing - one that even he did not approach again - while running inefficiently:

23 3/5, 22 3/5, 23 3/5, 24 2/5, 24 4/5, 25 !!

Every precept of horse racing analysis, every concept of tactics which we hold and use to understand the way races are run, every insight about pace we have borrowed from human athletics, all rely on the sensitive relationship between ideal pace and final time.

Before and after Roger Bannister and the Four-Minute Mile, to every world record attempt on the track or in cycling, swimming or rowing, pace is bound tightly to the achievement. Yes, the sprinter does it one way and the stayer another - as we have discussed in Pace: the final frontier - but there is still an absolute necessity to adhere to its importance.

Everything we know experientially about the middle-distance racehorse screams that the same rules apply. We could look at a sample of 10,000 fast times and see that the vast majority of them feature even pace, give or take a click or two.

But nobody seems that perturbed that Secretariat's 2:24 does not. In the narrative of his historic feat, the nonlinearity described by his sectionals has been subsumed by the concept that his greatness overcame all - including our understanding of energy use, apparently. (See Pace: the final frontier I for a full description of Secretariat's sectionals.)

If Secretariat could run 2:24 in the Belmont while going off much too fast, it should follow that he could have run significantly faster if Ron Turcotte had rationed his energy more evenly.

Yet, instinctively it is very hard to believe that he could. No horse has come that close to his record in the 37 renewals since, after all. 
  
And, it is not just Secretariat's Belmont which seems like a rupture of the bonds between even pace and optimal final time. If there are tens of thousands of horses who seem to "behave" according to our expectations, there are hundreds who do not, who put up their best performances while all over the place on the graph of pace against time.

From them, we can learn a lesson of great importance to the understanding of the running horse, of pace and its relationship with energy, of how we project future performance and predict the order of finish in a race. More of that in part III.