Charts and Scientific Games

Tweet 

Kentucky Derby 137 Chart by Race Position
(click on chart to enlarge)



I was messing around today and put together a quick chart to see who was making the biggest moves both forward and backward at the end of the race. It shows the chart positions at the six points of call during Saturday's Derby.  Interesting stuff, look at the late runs by Master of Hounds and Santiva to get to 5th and 6th respectively, both could prove dangerous on Belmont day. The pace wasn't right for Dialed In or Twice the Appeal who both show the hockey stick curvature upwards with their late moves.  



Speaking of statistics, a math guy got a hold of my spreadsheet and went all standard deviation on me.  While I was at the Derby I received an email and copy of my spreadsheet file with the following message:

"Here is my update of the excellent spreadsheet. I have attempted to analyze the speed data and post position data to come up with some kind of number reflecting speed, consistency and post position.

In the sheet I inserted a tab which shows the following, I took the three latest Beyer #’s and weighted them 100% for the most recent, 80 percent for the second and 60% for the third and divided by 3 to come up with a number. (Master of Hounds had two Timeform #’s. In my research it is the general rule to deduct 12-14 from the Timeform to equate to a Beyer, I took 15 just to be sure). Then I took the Standard Deviation of the three Beyer #’s to come up with a measure of consistency. Then I attempted to establish a Trend factor – that is, if the horse consistently improved speed over the last three races, it got a Trend of 3, if it ran slower in each succeeding race it got a 0, if it ran faster the last race but had run slower in the 2nd race from the first it got a 2.

So, if a horse has a 3 Trend, I added 25% of the Standard Deviation to the score, a 2 did not add the Standard Deviation and a score of 1 or 0 and I subtracted 25% of the Standard Deviation. This rewards for consistency of improvement and penalizes for a horse that has consistently ran slower.

Then, I took that score and added 25% of the Percentage in the Money from Post Position, to give a bonus for horses that have a statistically significant positive post position. That makes up my final number, Weighted Speed Consistency plus Post Bonus.

As Franie Bagodonuts (name changed to protect the innocent) reminded me, this ignores lots of factors you horse guys know and use, like impact of the poly track etc…but it is an attempt to use the data from the original spreadsheet to help make a decision about the horses in the race. I’d be interested in your comments or thoughts on any obvious flaws etc.  So, there you have it. Many thanks to Geno for the original spreadsheet and hope he doesn’t mind my tinkering with it."

So his final figures came up with the following ranking:
1 - Archarcharch
2 - Mucho Macho Man
3 - Nehro
4 - Twinspired
5 - Master of Hounds
10 - Animal Kingdom

Not too shabby actually, with three of the top five, and Archarcharch should be discounted because of his post draw, he was a real threat before drawing the rail. So no, I don't mind that you changed my spreadsheet and I may have you do a little math on it next year for kicks.  Hope you find it as interesting as I did.