Yesterday, I posted a link to my blog post, Speed Numbers: Detrimental To Your Handicapping, at Twitter, and I received the following question: “how do you feel about Brisnet, Equibase, Thograph, & Ragozin #’s.” As I was typing my reply, someone else chimed in with, “They all suffer from the same problem: human manipulation of a raw time. It’s pseudo science at best.” I could not have said it better, and I was thrilled to see there is at least one other person in the world who understands that speed numbers all suffer from one inherent problem. They attempt to take into consideration so-called track variants, and in the more than fifty years I have been handicapping horse races, I have not found a logical way to do determine if the track today could be faster or slower than it was yesterday.
Andrew Beyer, originator of the Beyer Speed Figure (BSF), utilizes par times, the average time it takes all horses to win races at any given distance at at various class levels at a particular racetrack. I can only assume that Brisnet, Equibase, et al, use similar concepts as I have never investigated how their respective numbers are compiled. In the case of BSFs, however, Beyer’s premise is highly flawed.
In his book, Picking Winners, Beyer suggested that all horses that contest races for a claiming price of $10,000 at a particular race track will run close to a par time for every distance and class level at that track. Whenever a horse entered in a $10,000 claiming race somehow beats this par time, we are supposed to believe it was the result of the track being faster than usual. What we are not supposed to believe is that when a horse is fit it is fully capable of running faster than the par time for a $10,000 claiming race.
When a horse runs slower than par Beyer wants us to believe the horse was running on a track that was somehow slower than usual. The possibility that the horse in question is not now and never will be capable of winning at the par time for the distance in question never seemed to cross his mind.
I will now provide you with a scenario and let you make up your own mind about speed figures. Assume the last three races for Horse A were each contested at six furlongs, and Horse A’s raw times for those three races were 1:10 1/5, 1:10, and 1:09 4/5. Now assume the last three races for Horse B were also contested at six furlongs, with raw times of 1:10 4/5, 1:10 3/5, and 1:11. Which horse would you wager on? To me, it is clear that Horse A consistently runs the distance of six furlongs faster than Horse B, however, by using par times, the numbers for both horses might lead you to believe they were closely matched.