Day: 2015/06/08

A Day at the Races

Many of my posts focus on what I consider horse-racing in the law school context (LSAT tea-leaf reading, applicant counts, etc). Because I’m a fair-weather fan of thoroughbred racing, I figured it’s time to indulge myself and write about it for real in deference to American Pharoah’s Triple Crown victory at the Belmont on Saturday.

Personally, I didn’t think I’d see a Triple Crown winner in my lifetime given the disappointing losses I’ve witnessed, like Real Quiet in 1998 and Charismatic in 1999. Maybe it’s changes in breeding practices, increased frequency of fresher horses challenging the leader at the Belmont, or bad luck.

Nevertheless, I thought it’d be fun to look at the winners since 1931, when the track distances and the spacing between the three races were finally standardized. Years with vertical black lines denote Triple Crown winners.

Winners' Fastest Times

(Forgot to mention, the units are minutes, the source is Wikipedia.)


(1) Each of the races took longer on average before 1950. The phenomenon is especially pronounced for the Kentucky Derby and the Preakness. For example, American Pharoah’s 2015 Preakness victory was about one second faster than the pre-1950 average but three seconds slower than 1950-forward average. The difference: There was a rain storm before the race this year, softening the track and dampening the atmosphere.

(2) Secretariat (1973) was a monster. It helps visually that the 1970 Belmont was also dumped on. I have no idea how often a horse like that comes along, but it’s rare.

(3) Consequently, it’s surprising that Seattle Slew (1977, and also a beast in his own right) and Affirmed (1978) occurred back to back, much less at all.

I’d need to look at the dispersions of the results for each year and each race and figure out how to account for weather, which I don’t feel like doing, so the results here suffer from substantial sample bias. Three races per year with different horses running them in different conditions can’t really tell you much. (Bad) luck factors heavily in these races, and assuming that something unobvious changed in 1950, there aren’t many good predictors for which horse will go three for three.

I guess that’s why people bet on it.