Over Sixties Club (Under Twenties, Too)

This feature was suggested following discussions in the CW Forum, when forum member watson posted a thread discussing American evolutionary biologist Stephen Jay Gould’s theories on baseball batting averages and its applicability to Don Bradman’s outlying average. In a nutshell, Gould posits that there are more better hitters now but also that fielders and pitchers have improved, such that what was an achievable milestone some time ago (e.g. a .400 batting average, or “four hundred hitter”) is no longer achieved, and that what was .400 before is now some lower value. That is, .400 is unlikely to be achieved again.

I should first explain what a four hundred hitter is for those of you don’t know much about baseball. In baseball, it’s considered difficult enough just to hit safely that the batting average has nothing to do with runs, rather it’s an average of how many times a batter went up to bat and how many times he was able to hit the ball and get on base. So a four hundred hitter manages to hit the ball, not get out somehow and reach base safely four times in ten attempts (for a batting average of .400, or four hundred). Whether or not runs are subsequently scored as a result of either the hit or from getting on base is not factored into the batting average.

There were two discussions which arose in this thread largely involving myself, watson, Red Hill and, appropriately, the big bambino, which I’d like to look at in more detail; the first was elicited by my posting of adjusted averages from a book by Surjit S Bhalla in 1986. Mr Bhalla’s adjustments were based on both opposition and scoring (both era and conditions) which brought down Harold Larwood’s average by 13 points – this seemed inappropriate in context and as a result I decided to run my own adjustments. Though these are based only on opposition strength, nonetheless I think they are informative (I’ll post Mr Bhalla’s adjusted averages in my next piece).

The second discussion was based on the big bambino’s suggestion that a 60-plus average has not disappeared in the same way Gould suggests that the .400 baseball average as, and as a result Bradman is even more of an outlier – this piece looks at that issue, my next piece will discuss my adjusted averages as compared to Mr Bhalla’s.

The disappearing four hundred hitter

A season batting average of .400, or four hundred, has been achieved on 28 occasions by 24 different players, though not since Ted Williams in 1941. Indeed, only two of the top 100 season’s best performances took place in the last 40 years – Rod Carew hit .388 in 1977 and Larry Walker .379 in 1999, though the latter is somewhat inflated by the friendly environments of Coors Field in the thin air of Colorado (he hit .461 at home, just .286 on the road).

Of course Gould’s .400 threshold refers to a season’s best, whereas Bradman’s 99.94 was compiled over a 20-year career. The best career batting average in baseball is .366 by Ty Cobb over a 24-year career, however the next best is Rogers Hornsby just behind Cobb on .359:-

Avg. Player
.366 Ty Cobb
.359 Rogers Hornsby
.356 Shoeless Joe Jackson
.349 Lefty O’Doul
.346 Ed Delahanty

The table above shows that, over a career, there is no outlier in baseball as there is with Bradman in cricket. Babe Ruth is often posited as such an outlier, though that is more in home runs, however Ruth has been caught and overtaken as time has gone on – that will happen of course with aggregate records.

Nonetheless, it could be that Gould’s theory holds water – if the number of players maintaining a high batting average in cricket, say 60, over a significant period has diminished, then he may be right.

The disappearing 60-plus batsman?

Below is a list of the number of matches following which a Test player has maintained a batting average of at least 60 (restricted to achieving more than 10 Tests when the average was 60+):-

Mt Player
54 Herbert Sutcliffe
49 Don Bradman
39 Walter Hammond
35 Garry Sobers
33 Everton Weekes
31 Michael Hussey
28 Javed Miandad
27 Neil Harvey
24 Jimmy Adams
23 Doug Walters
21 Denis Compton
21 Frank Worrell
18 Arthur Morris
17 George Headley
17 Thilan Samaraweera
15 Len Hutton
14 Adam Gilchrist
13 Viv Richards
12 Norman O’Neill
11 Vinod Kambli
11 Chetan Pujara
11 Lawrence Rowe
11 Mark Taylor
10 Sunil Gavaskar
10 Jonathan Trott

That’s 24 players with at least ten Tests following which their average was 60 or more. Sutcliffe’s number represents his whole career, since his average never dipped below 60, while Bradman achieved the feat for 49 of 52 matches. Other CW forum threads have discussed the hot starts of Michael Hussey and Jimmy Adams, both of whom feature prominently in the above list. There are batsmen featured who I was happy to see represented, such as Denis Compton and Doug Walters. Interestingly, no sign of Tendulkar, Sangakkara, Ponting, Lara…

Of course, we know that Bradman’s average didn’t just tip over 60, he significantly exceeded that figure, so let’s gradually increase the threshold to see who still fares well – the list below has increased the average threshold to 70 and decreased the minimum matches to five:-

Mt Player
47 Don Bradman
21 Michael Hussey
18 Herbert Sutcliffe
13 Javed Miandad
13 Jimmy Adams
12 Neil Harvey
12 Frank Worrell
11 Doug Walters
10 Vinod Kambli

We’ve now cut our original list of 24 down to nine – Bradman now has a considerable lead over Michael Hussey, who’s fast start pulls him up to second, ahead of Sutcliffe. Javed Miandad shows his class in fourth and Vinod Kambli hangs on in exalted company.

Going back to the 60-plus average, players who started well, such as Hussey, have a clear advantage over those who started slowly, such as Sobers – the more innings which have already been played, the more difficult it becomes to increase the batting average to over 60, while starting with a high average which gradually reduces can mean more total Tests over 60. To be fair to those slow starters, we can also look at over how many 10-match stretches this average was maintained, i.e. check the moving 10-Test average and note how many times it was over 60:-

Mt Player
77 Sachin Tendulkar
73 Jacques Kallis
57 Shriv Chanderpaul
56 Ricky Ponting
53 Rahul Dravid
51 Kumar Sangakkara
50 Steve Waugh
47 Brian Lara
43 Don Bradman
43 Walter Hammond
41 Javed Miandad
39 Allan Border
38 Garry Sobers
37 Ken Barrington
35 Younis Khan
33 Mohammad Yousuf
31 Len Hutton
31 AB de Villiers
30 Matthew Hayden

This list has brought the modern greats to the top of the tree, and highlights the possibly under-appreciated greatness of Steve Waugh – rarely is the great Australian mentioned in the same breath as the other batting greats around him in the above list. So why does Bradman’s moving average appear to be over 60 for only 43 matches, when he barely dropped below 100 for his 52-Test career? It’s because a moving 10-match average will began after match number 9, therefore he has 43 instances of a 10-match average to take account of – obviously in all of them his moving average was over 60.

If we now increase the average threshold to 80-plus, we see the following batting greats represented:-

Mt Player
37 Don Bradman
21 Jacques Kallis
20 Steve Waugh
15 Shriv Chanderpaul
15 Walter Hammond
14 Kumar Sagnakkara
13 Javed Miandad
12 Garry Sobers
12 Sachin Tendulkar
12 Dilip Vengsarkar
10 Mohammad Yousuf

Waugh has once again risen – I must say my own appreciation of his batting increased immensely as a result of researching this piece; at the time I suppose I was a little sceptical of the hype, but he is fully deserving of the accolades credited to him even when subject even to a statistical analysis such as this. Also Dilip Vengsarkar joins the elite group.

Finally, here is the same exercise performed with a threshold of 100-plus:-

Mt Player
27 Don Bradman
6 Kumar Sangakkara
5 Garry Sobers
5 Dilip Vengsarkar
4 Andy Flower
4 Hashan Tillakaratne
3 Ken Barrington
2 Javed Miandad
2 Adam Voges
2 Steve Waugh

We are flying in extremely rarified air now, yet Bradman has more than four times as many 10-match stretches than the next man while maintaining an average over 100, though considering that he could easily have completed his career with a three-figure average, this shouldn’t be too surprising. One again Vengsarkar enjoys a high ranking, and nice to see Zimbabwe’s Andy Flower making an appearance.

Looking at all of the above, what we can say with some certainty is that there does not appear to be a trend of reducing averages in cricket in the same way as Gould has noted in baseball – almost all of the players shown in the last three lists are what we would consider modern players. Though in fairness these modern players play far more Tests than did those of pre-World War 2 players, nonetheless those modern players listed above clearly have had no problem achieving the 60-plus threshold.

We must therefore conclude that Bradman is just, well, Bradmanesque – a true one-off. In any case, I’ll revisit this when I look at adjusted averages – possibly Bradman’s dominance will be reduced as a result of taking into account opposition strength.

Under-20s club

The very low career average of George Lohmann also had me wondering if we might on closer inspection see a trend towards higher numbers when looking at bowling averages. I decided to look at a sub-20 bowling average to determine if such a trend existed – this is a figure which is rarely achieved over a career of any significant length.

The list below shows those bowlers who have maintained a career-to-date average of sub-20 for the highest number of matches (showing number of matches with sub-20 average, wickets per match, total matches played, percentage of matches sub-20):-

The above list is restricted to bowlers with a career WPM of at least 1.00, otherwise the list would be dominated by occasional bowlers with low averages. Looking at the list above, it’s no wonder the West Indies held all before them with both Ambrose and Marshall in the side. Jim Laker is the top-rated spin bowler, and by some margin. A number of players maintained a sub-20 for the whole of their careers (the Barnes’, SF and Billy, as well as Bobby Peel, all showing as 100% above), though these bowlers played all of their Test cricket prior to World War 1.

As we did with batting, here is the list of those bowlers who maintained the most 10-match stretches with an average of under 20:-

Mt Player
44 GD McGrath
40 M Muralitharan
33 MD Marshall
32 CEL Ambrose
31 RJ Hadlee
23 AA Donald
21 AK Davidson
20 SM Pollock
20 DW Steyn
20 Waqar Younis
19 CA Walsh
18 J Garner
18 RR Lindwall
17 MA Holding
17 FS Trueman
16 J Briggs
16 JC Laker
16 SK Warne
15 Wasim Akram
15 DL Underwood
14 Shoaib Akhtar
14 AV Bedser
14 JB Statham
14 JH Wardle
13 IT Botham
13 GAR Lock
12 RGD Willis
10 TE Bailey
10 KR Miller
19 G Ulyett

What a line-up – Glenn McGrath is now our top-rated bowler in terms of duration of maintaining a sub-20, 10-match average, though many on this list are all-time greats. Again, we can see that, though this list appear to have more historical balance than our equivalent list for batting (as might be expected considering the generally lower scoring at the turn of the century) nonetheless there are many modern players shown here – just as we did in batting, we are not seeing any trend towards an unachievable threshold in bowling.

Wizards of willow and cork

Finally, let’s look at those elite all-rounders who maintained superior averages in both batting and bowling. Actually it’s necessary to increase the bowling average threshold to 25 rather than 20 to get a reasonable sample – the following list shows players who had a batting average of 60-plus and a bowling average of sub-25 simultaneously, again using the 1.00 WPM threshold and averaging over 10-match stretches so as not to penalise slow starters:-

Mt Player
18 SR Waugh
16 JH Kallis
7 SM Pollock
6 Imran Khan
4 Wasim Akram
4 Mushtaq Mohammad
4 Marlon Samuels
4 Frank Woolley
3 Warwick Armstrong
3 Stanley Jackson
3 Garry Sobers

Nice to see Frank Woolley in there, considering the recent forum discussions of an all-time great England XI, also Stanley Jackson. But once again Steve Waugh’s status as an all-time great is very much to the fore, more so even than the great Jacques Kallis.

So congratulations to the newest entry to my list of favourite all-rounders, Steve Waugh – a true wizard with both willow and cork.