Quote:
Originally Posted by westernmassbob
...Only problem is these animals ship in and out of tracks with different whip rules. That alone skews the stats and makes them highly suspect. Lord knows how confused some of these horses must be...
|
The data doesn't really support that.
I created a text file that contains four cut and pastes of sql query results that I ran showing year to date data for 2022 with the data broken out by rank within each race for the rider stats mentioned in my previous post.
Link to the text file here:
http://www.JCapper.com/Messageboard/...Riders2022.txt
1. The top section in the file is all track codes calendar year 2022 ytd (115,819 total starters from 01-01-2022 through yesterday 06-25-2022.)
Note that the rider stats breakout data by rank shows a clean downward progression. By that I mean rank=1 win rate was higher than rank=2, rank=2 higher than rank=3, rank=3 higher than rank=4, and so on, etc.
The one notable exception when you get all the way down to rank=18?
Rich Strike's Kentucky Derby.
2. The next section is calendar year 2022 ytd for three tracks I am aware of where limited whip rules were in effect: Monmouth, Golden Gate Fields, and Santa Anita. Here the breakout data shows a clean downward progression as well (except for rank=11.)
3. The next section in the file is calendar year 2022 ytd for Monmouth, Golden Gate Fields, and Santa Anita, but for only those starters who last raced at the same track code as today's race (non-shippers.)
Here, the rider stats breakout data shows an overall downward progression, but does have some noise in it. (The exception being rank=3 having a slightly higher win% than rank=2.)
4. The final section in the file is calendar year 2022 ytd for starters at Monmouth, Golden Gate Fields, and Santa Anita whose last race was at a different track code than today's race (shippers.)
Here, the rider stats breakout data for horses shipping to a track where limited whip rules are in effect shows an overall downward progression, but does have some noise in it. (The exception being ranks 8 & 9 with slightly higher win% than rank=7.)
My
lean is that the noise for non-shippers (rank=3) and shippers (ranks 8 & 9) in the MTH-GGX-SAX data is likely the result of small sample size as opposed to the horses being confused.
That said, feel free to interpret the data as you see fit.
-jp
.