Quote:
Originally Posted by classhandicapper
I'd be willing to bet that if we did a test of all the high quality publicly available speed figures there wouldn't be more than a 1% or 2% difference in win%. There "might" be more of difference in ROI because some are more widely distributed and some attract bigger bettors.
I think being comfortable with the thinking and methodology of the figure maker is almost as important as the figures.
Do you want to include pace, weight, ground loss?
Do you want races that don't fit the variant of the day broken out routinely or only in certain circumstances?
I sort of like what TimeformUS does because it gives me the final time figure and a pace adjusted figure. That allows me to see the algorithm but also allows me to make my own subjective adjustments to the final time figure if I want. There are also notations about races and days that may have been more difficult than others. That's something I wanted for years.
|
I asked about a study like that but also realized that no one has subs to multiple data vendors....so they are safe..for now
Unless a group of different data users would like to get together including a programmer to help out with the testing...wink wink
