|
I have experimented
You can use bris files. i.e. I will take Win percentage % Sprint or Rte % for Jck and take Tn% plus best percentage on whatever the angle Bris sees as a positive for the trainer pertaining to the records in these types of races. I will then add these percentages together and you can take their previous post position in their last race and subtract that percentage to today's post. It will either give a positive boost to the their post or nothing or it will downgrade it.
Then I will take all those percentages and divide by the number of horses in the race and will a show whose has an advantage or who may improve.
I have done this with spreadsheet handicapping in terms of a rating number, there are many out there who disagree but a jockey who has a good sprint percentage and knows how to get a horse to break properly. I only do this with dirt sprints. or 1 turn miles. You can use equibase as well if using other past performance data. It is just another statistic but it will help even up some of the performance percentages particularly if a jockey is new working with this trainer or limited number of amounts. It can coincide with a jockey switch or a trainer switch or trainer's original jockey was booked.
I have also done things like in many experimental manners as I am looking at developing my own software. One of the things you can do with these percentages is to take a look at the best early fraction of a paceline and add that into your position. Just thoughts, there will be the pundits on here that do not take it into account but the best trainer(despite his suspicious records and suspensions due to drugs) for getting his horses to break fast from the gate, always claims it is one of the most important factors in teaching a horse and to get the horse into proper position and clear. If Bob B. is wrong, well I do not know then I am wrong so to me those percentages can be used as part of anyone's handicapping arsenal. I guarantee all software developers use this in part of their developed algorithms when they code their machine learning.
|
