2014 Kentucky Derby Recap - Understanding the low Speed Figure

Preakness Stats Guide
Get ready for the Preakness Stakes with our Preakness Stats Guide.  Packed full of stats and analysis to help you include and exclude horses in the second leg of the Triple Crown.  


Kentucky Derby Recap - Understanding the low Speed Figure 
I've heard and read a lot from people who are having trouble understanding why the speed figure came back so low for California Chrome's Kentucky Derby win, even though he was well ahead of the field.  It is important to note that just because he received a low speed figure that DOES NOT mean the people who make the speed figures think California Chrome is a slow horse.  Any respectable figure maker has to let the data make the speed figure, not what he thinks the number should be based on the race visually.

As an example I'll walk you through how I landed on a par rating of 94 in the Kentucky Derby for California Chrome.  I will say that there MAY have been factors in play on Derby day that we simply do not have or cannot have data for - such as wind and the watering schedule for the track, which in conjunction with the sustained wind seemed to have fluffed it up.  Should California Chrome come back in the Preakness and run a number close to his San Felipe and Santa Anita Derby (108,106 on my par rating scale), then any serious figure maker will have to consider that there were factors involved that we were not able to consider in the making of the figure and then adjust accordingly.  Further, I have to say that I think that is a real possibility.  But, until the Preakness is run or some of the other horses from the Derby that showed a serious regression run again, we have no way of adjusting the figure beyond what the data allows.

Now back to how I came up with a par rating of a 94.  The average running time for the Kentucky Derby over the last 20 years on a fast track is 2:01.72.  That is out of 15 races where the track was labeled fast.  California Chrome ran a 2:03.66 on a track labeled as fast.  Before we consider any type of track variant that would put California Chrome's par rating at a 90.3.  To come up with a track variant I selected the other two races on the card that were run around 2 turns on the dirt and threw the 1 turn dirt races out.  The reason is because the 1 turn races had to run against the wind just one time in the stretch, unlike the 2 turn races that had to run through the stretch twice.

Race #2 was an Alw OC run at 1-1/16 mi on the dirt.  The winner received a par rating of 102.2.  The other race was #13 a Maiden Special Weight at 1-1/16mi on the dirt.  The winner of this race received a par rating of 96.4.  While I think this par rating is most significant as it is closest to the Derby, I simply cannot pick and choose which 2 turns races I want to use from the card as that would skew the data.

If you were to average all three par ratings (102.2, 90.3, 96.4) and then subtract that number from 100 (par), you'll find that the track played 3.7 points slow on the par rating scale for the 2 turn races.  Now, if you added that 3.7 back to the par ratings for each of those races, the new par ratings would be 105.9, 94 and 100.1.  That is how I came up with a 94 for the Kentucky Derby.

With all the above said, I realize that by giving the Kentucky Derby a par rating of 94 means that 12 of the 19 horses would have regressed on my par rating scale.  I don't think this is possible and it seems that there must have been other factors at play which we could not produce data for that effected the time of the race.  However, a 94 is what the data shows and until these horses run back again it is best not to adjust the figure based on speculation.  From a handicapping point of view, I would look at the figures that the horses ran in their Derby preps rather than the Kentucky Derby by itself.

Looking forward to seeing the final time in Baltimore!