Originally Posted by BDL
Some good ideas my thoughts would be keeping the numbers up in the more scores counted the better the average, however dropping a % or the worst eg off days or like mentioned 80% or even 70% of the last 5 plus the remainder at 20 or 30% dependent.
All scores must be added and that includes regions were 1's are issued for anyone not handing in a card, I'd go with a warning too and on second offence disqualifying them from that series too :-).
Hope you're well fella. I think the more I look at it the more I just want the score to be reflective of current status and as accurate as we can get them without being too complex.
I accept that no system for averages is perfect unless you have an something that's looking at every single number and if that worked I'd be using it on the stock market not to see what Our Kes has scored
If a two seasons are 11 W/L and 9 GPs, then taking 80% from the last 12 scores and 20% 8 from oldest should give you something more reflective. Then it depends on how you tune out the outliers, I should also add that looking at all last years data my changes would have a greater effect on C and bottom half of B because above that people do tend to be more consistent with no outliers to speak off.
But it doesn't really answer the question asked by two regions about grade movement, Shaun's said you only get two per year for reasons obvious. So maybe the answer to that is to split the W/L & GP's in two
Then you could have 4 and get more movement.