A decade of toil for bowlers

Last week's column comparing the batting stats for each decade drew a fair number of responses, and many suggested a similar analysis be done for bowlers. So here goes - a look at the bowling averages in each decade, with a more detailed analysis of bowlers in the current decade.

With the batsmen making merry so much in the 2000s, it's obvious the stats for bowlers won't look as pretty. While the previous piece had considered the performances of only the batsmen in the top seven, this piece looks at bowlers against all batsmen, which explains the difference in numbers.

From the table below it's clear that bowlers haven't suffered like they have this decade for the last 60 years. The last time they conceded more runs per wicket was in the 1940s, when they averaged 35.34; in the 1990s it was 31.51, and it wasn't much higher in the three decades before that. The best decade for bowlers, though, was the 1950s, when the likes of Jim Laker, Frank Tyson, Alan Davidson and several others ensured an average of less than 29.

How much of this increase in average has to do with the presence of Bangladesh and Zimbabwe in the mix? Not a whole lot, it turns out, due to the small number of Tests that Bangladesh and Zimbabwe play each year. In matches involving only the top eight teams, the average runs per wicket is slightly higher than the overall number for this decade - 34.57, against 34. However, when the top eight have played all teams (including Zimbabwe and Bangladesh) that number drops to 32.86.

And now let's break down the numbers further and look at the number of bowlers who've averaged less than 30 - an average that's generally considered to be the benchmark for a very good bowler - in each decade. Taking a cut-off of 3000 balls bowled in each decade, 79 bowlers qualify in the overall list, of which 18 have averaged less than 30, a percentage of 22.78. The table below breaks this up further, by bowlers who've averaged sub 25 and those who've averaged between 25 and 30.

When comparing under these parameters, the difference between this decade and the previous one is more stark than the difference in averages suggests. In the 1990s, 33 out of the 66 bowlers who made the cut-off averaged less than 30. Only 10 of those had a sub-25 average, but 23 more averaged between 25 and 30, which in percentage terms is significantly better than in the 2000s. (Click here for more details.) The usual suspects are on top of the averages chart for that decade - the top six names are Ambrose, Pollock, Akram, Waqar, Donald and McGrath - but a little further down that list are more unexpected names: Steve Waugh bowled 585 overs in the decade and averaged 29.23, while Venkatapathy Raju took advantage of the designer tracks in India to average 29.79.

The 1980s was excellent for bowlers too, with more than 47% of them averaging less than 30, but even that pales when put next to the numbers from the 1950s - 29 out of 39 bowlers averaged less than 30, which also explains why the overall stats for the decade are so good.

Excluding Bangladesh and Zimbabwe from the mix, 14 out of 67 bowlers have a sub-30 average in the current decade (with a 3000-ball cut-off), a percentage which isn't that different from the overall numbers.

Of the four bowlers who are in the first list but not the second, one of them - Shane Bond - isn't there only because he doesn't make the cut-off of 3000 balls against teams other than Zimbabwe and Bangladesh. His average, though, drops off quite significantly when excluding those two teams, from 22.39 to 27.25. That's because he did cash in against those two teams - his 11 wickets against Bangladesh came at an average of 13.63, while against Zimbabwe he averaged 9.23.

Similarly, Waqar Younis' average against the eight top teams was almost five more than his overall average, but then his peak was clearly the previous decade. Chris Cairns averaged 12.23 against Bangladesh and 20.09 against Zimbabwe, but against the rest his average ballooned to more than 37.

For most of the other bowlers the numbers were fairly similar even when excluding the two weaker teams; for Glenn McGrath the average didn't change at all.

Among the feedback for last week's article was a comment that the number of batsmen averaging more than 50 was disproportionately high considering that the overall difference in average between the 1990s and the current decade is less than three runs. That's a valid observation, and indicates that many more batsmen averaged in the 40s in the previous decade than in this one. That's exactly what the table below shows, with more than 35% of the batsmen in the 1990s averaging between 40 and 49.99, compared with only 26% in the current decade. Does the quality of batsmen in the "40s list" in the two decades appear any different? I expect readers to have fairly varying views on that one.