Wickets per match or wickets per innings?

[Days of Grace]

I am currently in the process of updating my top 100 test batsmen and bowlers list, which I expect to post on the forum by the end of the year.

When calculating the merits of batsmen, I have settled on average, runs per innings (a measure of productivity), and strike-rate. The weighting is 4:2:1 (average : runs per innings : strike-rate).

For bowlers, I did the same thing: average, wickets per innings and strike-rate, with the same weighting. But I have my doubts as to whether wickets per innings is a better measurement than wickets per match.

For one thing, a bowler such as Imran Khan gets a much higher rating if I employ the wickets per innings measure, since there were parts of his career where he played as a specialist batsman, including his insane 1982-1986 peak. One could argue that he didn't impact matches as much as a bowler such as Dennis Lillee did, simply because he didn't bowl as much. Using wickets per match will also decrease the ratings of part-timers, who were only asked to bowl when the frontline bowlers needed to rest.

Anyone have any thoughts in regards to WPM vs WPI?

 

[TheJediBrah]

The right answer is neither.

But if you have to choose one WPI is clearly more accurate than WPM, surely

 

[Days of Grace]

Why isn't either a good measurement?

Are you saying that if a part-timer averages 20 but only takes 1 wicket per match, then they are a better bowler than someone who averages 30 at 4 wickets per match?

 

[TheJediBrah]

Why isn't either a good measurement?

Are you saying that if a part-timer averages 20 but only takes 1 wicket per match, then they are a better bowler than someone who averages 30 at 4 wickets per match?

That's an extreme example that I don't think has ever happened, within a reasonable minimum number of wickets to qualify

We've probably talked about this before but basically my line of thought is including WPM or WPI gives a less accurate rating than if you didn't, the factors affecting WPM (that aren't already accounted for in average, SR etc) are virtually all outside the bowlers control and irrelevant as to how good they are

 

[honestbharani]

WPI

 

[Line and Length]

WPM is harsh on players in weaker sides where they might lose by an innings or by very few 2nd innings wickets.

 

[Days of Grace]

That's an extreme example that I don't think has ever happened, within a reasonable minimum number of wickets to qualify

We've probably talked about this before but basically my line of thought is including WPM or WPI gives a less accurate rating than if you didn't, the factors affecting WPM (that aren't already accounted for in average, SR etc) are virtually all outside the bowlers control and irrelevant as to how good they are

I will use a less extreme example.

Bowler A is a seamer:
Average: 25
WPM: 3.5

Bowler B is a spinner:
Average: 30
WPM: 4

Spinners typically have higher averages and much higher strike-rates than fast bowlers/seamers so they need something to compensate them. They bowl longer spells so they have greater productivity, i.e. wickets per match.

 

[TheJediBrah]

Comparing seamers and spinners with the same metrics has it's own issues

If you have 2 seamers or 2 spinners with a similar split in stats the guy averaging 25 with at 3.5wpm is easily superior to the guy averaging 30 at 4wpm