Article - Correlating Works to Performance

December 31, 2008
Steve Thygersen

A bullet work, that's really good, isn't it? That has been open to debate for some time, so we thought we would run a little
experiment to see if there really is any correlation between fast work times and race day performance. Initially we
cataloged the top 6 furlong works and have now expanded the base to include 5 and 7 furlong works as well. We chose
not to include 3 furlong and half mile works because just about any horse can throw excellent short works. We will also
exclude mile works as they are fairly rare, generally a half dozen per season.

We will use works from Hollywood Park and Santa Anita as our source, and then will track the race performances of
those runners with top works during the last couple of days at Hollywood and the full Winter/Spring meet at Santa Anita.
The parameters for "top works" are that they were the fastest work of the day at the distance or are among the top three
works at the distance for that day. Normally we will set 15 as the minimum number of runners for 5 and 6 furlongs, or
any 7 furlong work times in less than 1:27. Here are some of our preconceptions based solely on experience prior to
starting the experiment:

1. That fast works are a good indication of the fitness, readiness and talent of those readying for their maiden debut.

2. That fast works are a good indication of the fitness and readiness of those coming off of layoffs

3. The longer the work, the greater the validity.

The problem is (like all things handicapping), is it true? Only one way to find out, and that is to run an experiment, and
that is what we are doing. In order to make it readily available to everybody, we have posted it on Google Docs.

The URL is:

http://spreadsheets.google.com/ccc?key=pIaalb5z28p9wyzO8GdaaXQ&hl=en