|
JCapper Message Board JCapper 101
|
| Home |
Register
Log In |
| By | JPR/UPR GAP REV | |
| bettheoverlay 4/27/2008 1:00:25 PM | I just noticed these 2 check boxes in Systems Definition. What do they do?
| |
| jeff 4/28/2008 2:13:35 PM | When horses are scratched in JCapper you now have two options: 1. All horses that are scratched are assigned a 0 (zero) for numeric ratings such as CompoundPaceFit. Each field of horses is then scored and ranked for all factors according to this model. 2. All horses that are scratched are assigned their original numerical ratings for factors like CompoundPaceFit. A gap is the difference in factor values between the top ranked horse and the second ranked horse in the field. This second model is slightly different than the first model. This second model allows for some built in "cushion" when you create UDMs. Example: Suppose you create a UDM requiring that a horse have a CompoundpaceFit that is at least 10 pts higher (a 10 pt gap) than any other horse in the field. IOW, you are looking to have your UDM idendify horses that should have a solid pace edge (matchup-wise) over their fields. Model 1. When the software evaluates CompoundPaceFit gap between horses it ignores the scores of scratched horses when it decides yes or no if a 10 pt gap exists. Model 2. When the software evaluates CompoundPaceFit gap between horses it includes the scores of scratched horses when it decides yes or no if a 10 pt gap exists. A while back, RyeSteve wrote to me pointing out that he had noticed something "strange" while checking results for each horse in a race while building a one factor UPR. His email became the impetus for me to add the checkboxes for both info models. You the user have the ability to test both models in the Data Window and decide on which works best for you. The checkboxes on the System Definitions Screen control which gap info model is applied... and whether that model is applied to JPR, UPR, or both. IMPORTANT: The program default is model 2. That's right. I used info model 2 where the factor values for scratched horses are included when I developed the table entries to drive the JPR Calculation... as well as the internal hard coded values for JRating, PRating, QRating, Alchemy, CPace, PMI, etc. Techically, info model 1, where scratched horses are completely eliminated, is what happens when human beings read a racing form and draw lines through scratched horses. But when I did the initial R&D for CPace and JRating I discovered I could get a slightly different perspective (and a better roi too) by pretending that scratched horses were still entered to run. Be VERY careful about changing info models... especially if you have spent significant time developing UDMs and/or UPR. Changing info models won't affect the internal hard coded values for JRating, PRating, QRating, Alchemy, CPace, PMI, etc. But it WILL affect JPR and UPR. And it WILL require that you rebuild your databases from scratch and spend some Data Window time adjusting the factor gaps used by your UDMs. Generally, you can get away with smaller factor gaps under info model 1 than the program's default info model 2. That said, you can change to model 1 by checking either box and hitting the Save button. Personally, I am using the default info model for JPR. My own JPR box remains unchecked. But, I AM using the other info model for my UPR. My own UPR gap revision box is checked. -jp . ~Edited by: jeff on: 4/28/2008 at: 2:13:35 PM~
|
| Reply |
|
Copyright © 2018 JCapper Software back to the JCapper Message Board www.JCapper.com |