Quote:
Originally Posted by classhandicapper
Here's another "potential" issue that comes up with regression analysis.
Let's say I'm ranking the horses based on multiple factors and these are the speed figures for two races.
100 - 1
90 - 2
89 - 3
85 - 4
79 - 5
100 -1
99 - 2
98 - 3
85 - 4
79 - 5
Ranks don't do a good job of getting at the gap between the top rated horse and the rest.
|
The more commonly used factors that you use, the more likely you are going to be on the favorite. It's not a good way to pick for value.