View Single Post
Old 18th November 2014, 01:46 PM
rich's Avatar
rich rich is offline
Senior Member
Join Date: Nov 2008
Location: Shebbear, N Devon
Posts: 1,644

From a purely statistical point of view, it is also possible that a shooter is lucky on the day and gets a score above what he or she "deserves" from their grade position. So there is a case to say, top and tail the data by removing the worst score and removing the best score as well.

To allow for equipment failure it would be fair to drop off one extra bad score at the bottom.

It's evident that frequency of updating is crucial, and I'd also like to support the time fade concept so that the more recent results carry more weight; that way any trend in performance is captured; a simple arithmetic mean obscures all these things.
Reply With Quote