Quote:

> > It was mentioned in another thread that the batting average of all

> > test players combined was about 35, I would like to know if this is

> > true as I have a gut feeling that it should be lower. There must be a

> > lot more bunnies like McGrath pulling the average down than batsmen

> > averaging over 50 pulling it up. If it is true I would love a link to

> > see how they worked it out.

> Sounds a bit high to me. Let's take these stats for the current Australian

> team, which I'm guessing probably has the highest average:

http://www-aus.cricket.org/link_to_database/STATS/AUS/TEST_AVS_BAT_AU...

Quote:

> Based on all matches up to and including

> Test # 1620: Pakistan v Australia at Sharjah, 3rd Test, 19/10/2002

> (includes M Waugh and Macgill)

> 1. Now adding up all the averages (414.2) and dividing by the number of

> players (13), I get:

> 31.86.

> 2. Weighting it by the number of innnings played (for no good reason) -

when

> I sum the total of the individual averages multiplied by the respective

> player's total innings (40949.94) then divide it by the total number of

> innings (1121), I get:

> 36.52

You dont multiply and divide by the total innings, you multiply and divide

by the total number of times the

players have been out. There is a difference. If a player has played 100

tests, but been out in just 60 times

(a tailender probably who didnt get to bat a lot of times), then you wanna

multiply and divide by 60 and not 100.

And thats why the averages of the top order batsman weigh more than the

averages of the tailenders.