Zenyatta's Apple Blossom, Speed Figures, and Performance Ratings

This post is more or less a review of last weeks stakes results but I wanted to include a discussion of a couple of different articles that have come out over the last couple of days discussing Zenyatta's performance in the Apple Blossom in connection with speed figures and performance ratings. 

First things first, below are last week's graded stakes winners and their respective speed figures, as provided by the DRF.

*Click on the race name to be taken to the race chart at DRF.com

The Apple Blossom figure stands out as one of the slowest speed figs that Zenyatta has run in the last two years, and over at the DRF, publisher Steven Crist has a blog entry indicating that it was the lowest Apple Blossom figure in twenty-one years.  Crist notes the following in that entry:

The 6th race was the only other route on the card, but falls right into line with the day's other races, and it's simply impossible to give the Apple Blossom a stronger final-time figure...

Obviously Zenyatta is capable of running faster (her previous 11 Beyers were all bigger, including two 108's and a 112 in the BC Classic.) As a deep closer who is ridden positionally (i.e. drop back to last regardless of the fractions), she is going to run weak final-time figures sometimes, especially when the pace is slow and the competition can be disdainfully dispatched with a single furlong of top effort.    

I think the second part of that quote was the most critical portion of the entire post because it's, in my opinion, a critical flaw of speed figures - they don't tell you how the race was run or the ease or difficulty of the trip.  They are also locked into producing a number based on a small sample size of similarly run races on that same day (ie - routes, sprints, etc.).  Taken at face value, Crist notes that Zenyatta's figure can't be higher because of how another race was run earlier in the day at the same track; one other race helped to decide how fast Zenyatta ran.  That's not to suggest that I believe Zenyatta's figure is wrong, however I think it does indicate a limitation of its usefulness. 

The abscence of knowing "how a race was run" based off of a speed figure is not as big of a problem if you only follow top level racing where players can easily watch the performances of each horse throughout the year.  It's a much more difficult proposition if you don't have the luxury of watching races, all the races, of the top contenders in a specific race.  The visual aspect of how a horse runs can tell us plenty about that horse's abilities and the relative depth of the field they were facing.  In the case of the Apple Blossom, it was clear to anyone that saw the race that this mare barely had to break a sweat to win, and while the final time may have been slow, it's clear she can run much, much faster.

If Zenyatta wasn't Zenyatta, but instead just "Jane Claimer" at your local track, and she just ran the slowest speed figure in two years at her highest level of competition, what would our conclusion be about her performance?  If we just went by the figure we might erroneously conclude that she's vulnerable or on a downward cycle because without knowing the context of the race (the pace, the abilities of the other contenders, etc.) we'd likely be led astray in our analysis.  Suppose that in Zenyatta's next race we see one or two other fillies that ran Beyers of 96 or 97 in their last race, would we conclude that these fillies are more likely winners than Zenyatta?  Of course not.  But we might do that for a more unknown horse.

Speed figures are an invaluable part of the handicapping process, and at the highest levels of the game they are a pretty reliable indicator of thoroughbred ability.  But at other levels they are just a piece of the puzzle and without being placed in proper context (class) they can be lead a player in the wrong direction.

On another part of the spectrum are performance ratings, or ratings not based entirely on the final time but instead a combination of factors.  Over at the Racing Post (UK) there is an excellent weekly column that analyzes the top performances every week based off of the Racing Post Ratings.  This week's installment takes a look at Zenyatta, who earned a RPR of 123+ for her win in the Apple Blossom last Friday.

Race ratings, either from the Racing Post or Timeform, are assigned to a horse based on factors such as time, class, weight, and depth of competition.  RPRs can change once they are assigned if the quality of the field turns out to be much stronger than originally thought.  

A simple guide as to what kind of performance an RPR is describing is this handy chart that you can find at the Daily Racing Form.

Going by this guide, Zenyatta's win in the Apple Blossom equated to a good Group 2 level performance.  The "+" sign after the rating indicates that the performance may actually have been better than it appeared, which makes sense considering that Zenyatta was asked for very little run as she passed her outclassed rivals in the stretch. This number gives us similar information that her 95 Beyer does, in that it was a good, but not great performance.  It also indicates that she could produce a much higher rating in her next race.  

Another factor in this equation is the concept of pace since speed figures are heavily dependent upon the early fractional splits; a fast early pace will generally produce a fast final time while a slow early pace will typically produce a slower final time.  Sure, we see race shapes where the leaders go very slow in the beginning and them come home very quickly, producing a quick overall clocking.  Of course, we see that most often in turf racing where speed figures (and pace figures) are much less reliable than in dirt racing.  But generally, big speed figures in races on dirt come in races where the early pace is very, very fast. 

Performance ratings also incorporate how the race was run, but there's a little more leeway to assign a higher rating and the ratings of the other horses in the field do not have to be tied together.  Take note of this passage from Crist's piece:

The 6th race was the only other route on the card, but falls right into line with the day's other races, and it's simply impossible to give the Apple Blossom a stronger final-time figure. The runner-up, 35-1 Taptam, gets an 88 for the race, consistent with her career tops of 88, 87 and 85 in three of her last five starts.    

In speed figure analysis, if Zenyatta was given a higher figure then the horses behind her would also get a higher figure, which as Crist notes, would have not been consistent with the figures that those horses earn from race to race.  The figure is limited not just by how the individual horse performs (Zenyatta) but also as to how the other horses perform in the key race and in their previous races.  That's an important fact to remember about speed figures.

A horse like Zenyatta, or essentially any horse that is able to win off of a hot or a cold pace, is (in my opinion) a horse that should have their speed figures taken with a grain of salt.  The ability to win under a variety of pace conditions will likely produce a wide variety of speed figures, and the conditions under which those figures are earned seem much more critical than the raw number itself.