Could you find out what Everton Weekes would average using the same metric that you used?
I am more interested to see if Wally Hammond would average more or less than Pujara by that metric
Here's a sampling of some old timey types:
Len Hutton - 49.47
Garry Sobers - 57.29
Everton Weekes - 56.35
Ken Barrington - 55.09
Jack Hobbs - 27.99
Herbert Sutcliffe - 42.97
Wally Hammond - 43.62
I don't want to piece together the whole formula, as it would be kind of time consuming as it's spread throughout multiple cells of an Excel sheet, but the way it works is this:
Start with an adjusted average for these batsmen based on the methodology of power regression from ICC Ranking Average and Era Adjusted Average, described in this post:
Its amazing how consistent a band test averages stay in over history. When I watch the modern player block the ball to the boundary through the covers, or straight, with his modern super bat. I think, Mark Richardson was born for this era and would average 55 instead of 45. But, he probably...
www.cricketweb.net
For old timers you subtract their debut year from 1978 and fit the entire history of cricket before that point on a normal distribution, calculating their z-score in relation to where they are in the history of cricket, and using a normal distribution z-score table to determine their percentage modifier. You multiply that average then by the percentage modifier.
So yeah, that's kind of how it works, and I think it works pretty well given when the biggest growth period of skill in the game was. Accepting the overall premise, I do think there is a bit of a flaw in that using the debut year penalizes much older players who had longer careers and were actually more modern than their debut year would indicate. So in particular, I'd say there is an effect where the 2 decade careers of the three H's above lead to a somewhat suppressed average due to overaggressive modifier (especially 23 year career of Hobbs), in comparison to the more typical 12 year career of Sutcliffe which isn't as heavily modified. The effect shouldn't be huge though (maybe add a maximum of 2 runs to Hutton and Hammond, 4 runs to Hobbs). I could fix this by going with something like median career year to calculate the modifier, but I cbf any more.