|I never made 'par' for today's race (or a True/False indicator based on whether or not a horse has run to 'par') into one of the available factors. |
I recall doing some early research into viability of that - and my notes show that the work I did in that area back in the early 2000's using Brisnet data didn't provide enough of an roi boost to make me want to pursue that line of thinking.
Admittedly, it's been several years since I last looked into that line of thinking - but that's why that line of thinking isn't represented in JCapper.
That said, there is a tangent to that line of thinking built into JCapper that you can use...
During that early par based research, I discovered that a composite made up of speed figs earned in past running lines combined with data vendor 'pars' for those same running lines made for a decent representation of "class" and "ability."
Further, I did find situations where the composite factor I had created actually did provide an incremental roi boost (vs. ignoring that info.)
The composite Factor in JCapper that I am talking about here is Class Shift (Par vs Race Strength) and is currently listed as factor #17 on the JCapper Supported Factors Page.
To use it in a sql UDM you would reference the CLSHIFT2 field in the StarterHistory Table.
For example, the following line of sql would make a UDM flag only those horses with CLSHIFT2 greater than or equal to -12:
AND CLSHIFT2 >= -12
Hint: To get a sense of what CLSHIFT2 might mean for a given UDM, or whether or not you might want to add a CLSHIFT2 cutoff to a UDM, run the UDM through the Data Window with the data broken out by Class Shift (Using Par & Race Strength.)
~Edited by: jeff on: 10/23/2014 at: 11:36:47 AM~