Side track Sueprforecasting

A side track of this blog is a ruminant on predictability. Sports analytics is in a flower just now, and it is all about, of course, trying to guess at what will happen tomorrow. Considerable attention is directed at machine learning, and its potential to project the future things in the business sphere. And I am mucking about with horses, which are probably the most statistically studied of sports participant historically. Pulling bits from Tetlock's tome somewhat random as...

What Fermi understood is that by breaking down the question, we can better separate the knowable and the unknowable. So guessing—pulling a number out of the black box—isn’t eliminated. But we have brought our guessing process out into the light of day where we can inspect it. - [Excerpt From Philip E. Tetlock & Dan Gardner. “Superforecasting.” Crown, 2015-09-29. iBooks.  p. 111 This material may be protected by copyright.]

The basic handicapping method breaks the larger problem of which will win the race down to smaller problems of 1-which horses show what gradation of demonstrable class, 2-how  does their present form seem to be, 3-what speed are they likely capable of  and 4-how will this group gathered here today most likely compete in terms of early, late or stalking pace.

Running above is Skyler's Scramjet achieving a 1:10.80 for 6F on Dec 22, 2017 in 8th race at Aqueduct -- a $60,000 claiming race. The two horses chasing him are my exacta - Still Krz and Phi Beta Express who had dueled competitively with each other until they grew weary, allowing the stalking Sklyer to stride wonderfully forward under a deft ride by Dylan Davis. The $2 exacta paid a pretty $185.

I cant say I totally missed Skyler's Scramjet - Emerging from the reputable barn of Michelle Nevins, he had achieved some high numbers, was marked by TimeForm as right up there on the pace - I was diverted somewhat by a poor last out, and that he may have  experienced repeated claiming. [Claiming races are new thing to me.] In any case, I expected Phi Beta Express to  succeed on the lead, or Still Krz to come from behind  (instead of duel) and nip him. The idea was that a horse will either steal it on the front or set it up for a come from behinder - not a whole lot of effort put into thinking this one out.

So, in Fermi style, [See Note 1.] I can say in retrospect that my guess was off on how the early pace battle would enfold. And my measure of the winner's potential speed and form (readiness to run) were off too. - Romeo Racetrack.

Brings to mind old Flann O'Brien who wrote something like: When troubles seem to haunt you, and your horse has also ran. when all you have is a sack of woe, a pint of plain is your only man.


Note 1: Tetlock refers to a teaching gambit of Enrico Fermi - he would invite the class to estimate how many piano tuners there were working in Chicago. The successful among them would divide the general problem into smaller problems more easily estimated then the uber query.