View Single Post
  #20  
Old 28th August 2006, 02:33 PM
jfc jfc is offline
Member
 
Join Date: Jan 1970
Location: Sydney
Posts: 402
Default

Quote:
Originally Posted by wesmip1
jfc,


To determine the bias you divide the number of wins (or places) by the number of runs in the barrier.




Actually I do compute that ratio (= "fair share" for argument's sake) in all my tests. It is obviously more meaningful than Strike Rate.

Consider these tests based on my original sample comparing barriers 1&2 (inner) versus 9+ (wide):

Constant Field Size of 10, Any Price:

10.7% Inner S/R
9.4% Wide S/R

Or as Fair Share: 107% versus 94%.

Suggesting the wide have a negative bias:


But with Constant SP of precisely 2/1 and any Field Size:

27.1% Inner S/R
29.7% Outer S/R

Or as Fair Share: 186% versus a whopping 342%.


Damning evidence that for favourites the Market overpenalises wide barriers.
Reply With Quote