Tuesday, 15 February 2011

Pace: the final frontier II

EVALUATING THOROUGHBRED SPEED: A PRIMER ON PACE (cont.)

2.3 Variations in pace

Before we go on, here is something to think about regarding Secretariat's running time for the Belmont.

If the general axiom holds that a horse needs to run at even pace to achieve optimal time, it seems like we have a problem.

Secretariat ran at nothing like even pace. And his fastest sectional time came early in the race. It would be hard to think of a tougher way for a horse to run a mile and a half.

The corollary of this is troubling. Could Secretariat have run faster than 2:24 if Ron Turcotte had not made him go off so hard? It seems from this snapshot of the jockey's judment of pace that Richard Pitman might have done a better job.

But, is it really possible to contemplate that Secretariat could have run faster still? That a performance widely regarded by time experts in the US as the greatest ever might have been significantly improved upon?

I strongly believe Secretariat could not have run much faster. And the reasoning I will give - as this series burrows deeper into the subject of the equine athlete - is a clue to new ground in the study of thoroughbred pace. Until then, I would be interested to hear of any theories you may have.

3.0 Accounting for conditions in assessing final time

Let's return to looking at final time (that for the race from start to finish). On any given day at the races, it is impossible to know precisely how much a horse's final-time ability was affected by conditions.

Sure, with a large amount of data we can confidently allow for the hill at Ascot or the bends at Epsom, but what do we do about more variable and temporary effects?

There might be a wind of changing strength or which veers in different directions; it might rain during the card, or the surface may deteriorate in some other way.

Allowing for these factors is at the heart of calculating speed figures like those produced by Timeform and the Racing Post in the UK or by Andy Beyer and Jerry Brown in the US.

Different brands of figures use different methodologies and different scales, but their aim is largely the same: to derive, from time-based information, a quantitative measure of racehorse merit on a convenient scale.

3.1 A brief history of time(figures)

The various compilers of speed figures allow for the variance in conditions with the same broad approach.

They compare the data actually produced by a series of races with their expectations. They hope to produce a mathematical allowance for conditions which can be retrospectively applied to each horse's running time.

In theory, this should allow the data from one day to be compared with another.

That's as far as I want to go for now. Many people find the topic of speed figures particularly soporific.

4.0 A deeper look into pace

This is a series which takes pace as its focus. While I mine deeper into the topic, I am aware that many of you will find it useful to think about pace purely in a qualitative manner.

While many disciplines with a more eloquent language than words, many people are intimidated by mathematics. Or they simply have the type of brain that prefers a conceptual understanding to one involving data, equations or regression.

During the last part of the 20th century, and now into this one, subjects which had formerly been impenetrable to mathematical analysis suddenly became within our human grasp. This was beautifully put in a marvellous, esoteric book called Handicapping Speed by Charles Carroll.

Carroll draws from several contemporary scientific disciplines in his wide-ranging and dynamic discussion of quater-horses and thoroughbreds. Not the least of these is chaos theory:  "Art borrowed from science, even became science in some cases," he says.

Economics and weather forecasting are two examples of subjects in which statistical logic and data manipulation have greatly aided our intuitive sense of what to expect. Though both disciplines, you will quickly note, have not been without catastrophic error.

Horseracing - our own little corner of the universe, as Carroll puts it - has been more slowly affected. It is inevitable, however, that the nebulous notion of "class" in the racehorse will give way to the more quantitative measure of speed.

4.1 Future implications for horseracing of its benighted status

Still, however, formal mathematics has hardly been put to any use in our understanding of the sport. True, mathematical modelling underpins the betting success of several sophisticated operations of which even I am aware, but this approach is still a complete mystery to the mainstream.

Long may that continue, I can sense some of you saying, as you sit there with balances akin to the GDP of small South American countries on your Betfair screens.

But the subject of pace has the potential to revolutionise more than just how we bet on racing. Crucially, the biggest changes may become apparent in the way that trainers and jockeys actually operate.

The problem - or perhaps the opportunity for some - is that many racing professionals still believe that empiricism is the only route to knowledge. And this is an attitude which the reserve of the racing culture does much to procreate.

One day, we will realise the degree to which we were still living in the dark ages.

* * *

In the next part of this series, I will look into surface differences and their implications for pace, and I will flesh out the concepts referred to in the final paragraph.


Plus, for those of you serious about your racing education and technically inclined, I will be giving you the name of a book in a completely different league to any other publication in its sphere. I am sure many of you can guess.

Thanks for reading.