Cricket: The “Best Bowling Attack” Hyperbole

Much has been made by those in the Australia cricket set up and press of late of the Australian team’s claim (via Craig McDermott and Peter Siddle principally) that it possesses the best bowling attack in the world.  This statement has come after Australia’s annihilation of England in the recently completed Ashes series.

There is no doubt that the Australia bowling attack bowled splendidly against England, led by Ryan Harris and Mitchell Johnson.  That said, I have been left to wonder over recent days as to whether the claim to fame as being the best bowling is merely hyperbole or has some roots in fact.

Obviously then it is important to consider the facts around who the best bowlers in the world currently are and then consider whether those best bowlers form the best attack.

The ICC Test Match Rankings are as good as place as any to start.  The current bowling rankings (updated to include the recently completed Ashes Tour) are:

Ranking Player Country
1 V.D. Philander South Africa
2 D.W. Steyn South Africa
3 R.J. Harris Australia
4 H.M.R.K.B. Herath Sri Lanka
5 Saeed Ajmal Pakistan
6 P.M. Siddle Australia
7 R. Ashwin India
8 M.G. Johnson Australia
9 P.P. Ojha India
10 T.A. Boult New Zealand
10 S.C.J. Broad England
12 M. Morkel South Africa
13 T.G. Southee New Zealand
14 Abdur Rehman Pakistan
15 J.M. Anderson England
16 K.A.J. Roach West Indies
17 B.W. Hilfenhaus Australia
18 N.M. Lyon Australia
19  Shakib Al Hasan Bangladesh
20 S. Shillingford West Indies

Australia has all four of its present bowling attack in the top 20 in the world at the moment, therefore, on first principals they have the best attack in the world will be the argument of some, particularly those who are ardent fans of the ICC rankings system. That said, there are many, me included, who hold little stock in the ICC system so it is necessary to dig a bit deeper in the form of the relevant bowling attacks.  Before doing so, the ICC Rankings are useful in one sense: it is possible to remove a number teams, and their attacks, from consideration here based on their lack of representation in said rankings: Bangladesh, Zimbabwe and Sri Lanka have scant representation and can be removed.

Statistically, I am of a view that an analysis of the last 12 months of test matches would be the best way to examine who has the best bowling attack.  As the tables below show, a simple statistical representation is problematic.

Most Wickets (1 Jan 2013 to date):

Player Mat Inns Overs Runs Wkts Ave SR
SCJ Broad (Eng) 15 28 512 1722 66 26.09 46.5
JM Anderson (Eng) 15 29 567.5 1768 55 32.14 61.9
DW Steyn (SA) 9 18 357.4 901 51 17.66 42
TA Boult (NZ) 12 21 429.5 1154 46 25.08 56
RJ Harris (Aus) 9 18 328.3 895 46 19.45 42.8
NM Lyon (Aus) 12 24 455.4 1556 45 34.57 60.7
PM Siddle (Aus) 15 28 489.1 1316 45 29.24 65.2
GP Swann (Eng) 10 19 440 1466 43 34.09 61.3
R Ashwin (India) 7 14 355.2 923 41 22.51 52
MG Johnson (Aus) 7 14 235.4 669 40 16.72 35.3
Saeed Ajmal (Pak) 8 14 409 1062 39 27.23 62.9
VD Philander (SA) 8 16 278 686 38 18.05 43.8
S Shillingford (WI) 6 9 261.5 802 36 22.27 43.6
TG Southee (NZ) 8 15 333.3 898 36 24.94 55.5
N Wagner (NZ) 10 18 373 1271 35 36.31 63.9
RA Jadeja (India) 5 10 255.4 573 30 19.1 51.1
HMRKB Herath (SL) 4 8 234.3 605 25 24.2 56.2
Junaid Khan (Pak) 6 11 242.1 656 23 28.52 63.1
Sohag Gazi (Ban) 6 11 290.3 807 22 36.68 79.2
ST Finn (Eng) 6 10 179.1 670 20 33.5 53.7

Test Match Results (1 Jan 2013 to date)

Team Mat Won Lost Tied Draw W/L
South Africa 9 7 1 0 1 7
Australia 15 6 7 0 2 0.85
India 8 6 1 0 1 6
England 15 5 5 0 5 1
New Zealand 12 2 4 0 6 0.5
Pakistan 8 2 5 0 1 0.4
West Indies 7 2 4 0 1 0.5
Zimbabwe 6 2 4 0 0 0.5
Bangladesh 6 1 2 0 3 0.5
Sri Lanka 4 1 1 0 2 1

The problem arises with in the obvious discrepancy between the number of tests played by the various test match playing nations. It is impossible to compare statistically the efforts of those who have played for, for example, South Africa with those who have played for England because of the significant variance between in the number of tests played by the two teams.

That being the case, a different analysis is necessary.  The efforts of a bowling attack are best examined, in my view, in the considering who often they take wickets (strike rate) and the quantum of runs scored against the attack.  The later examination is one that ought be undertaken in two parts: a per average based and then by tallying the number of innings under 200 produced by the opposition of each team.

This table sets out the results of that analysis:

Team Strikerate Innings Ave Innings <200
South Africa 50 224.27 6
Australia 57.74 268.14 6
England 57.62 281.13 3
India 56.68 254.38 5

Based on this analysis, then South Africa must have had the better bowling attack in the last 12 months mustn’t they? They have the best strike rate in terms of balls bowled per wicket and have restricted their opponents to the lowest sum of runs per innings.

Deniers of this view will say that Australia possesses more fire power and has the great ability to run through teams as evidenced by the efforts of Johnson et al against England.  Whilst it is true that Australia’s bowlers went through periods against England of taking clumps of wickets for not many runs, did they dismiss their opponents this year for less than 100 at any point? The answer to that enquiry is no whereas the South Africans did so thrice in the year.

Australia’s bowlers and coach have earned the right to be overjoyed with their efforts against England.  They should be careful though that this present attack of hyperbole as to who possesses the best bowling attack in the world does not return to bite them on the metaphorical backside.  If nothing else, one can bet that Messrs Steyn, Philander and Morkel have been watching the press coverage of Australia’s statements closely.

Sorry Sachin: Kallis is the best of the modern greats BUT not the best all-rounder (mathematically speaking)

I have spent a lot of time of late pondering the career of Jacques Kallis and his place in the game.  There has been much made of the retirement of him and Tendulkar this year and much said in the media, both formal and social, of which of the two of them is the second best player after the great Sir Donald Bradman.

If I have said it once I have said it a thousand times: it is impossible to compare players from this generation with those who played in the 1930s and earlier and that premise probably extends right up to the 80s.  It is equally difficult to compare the records of the other “great” players of Kallis’ generation with his record because of his sustained excellence across both key forms of play: batting and bowling.

Thinking about it though, if one makes an assumption that a five wicket all by a bowler is akin to a batter scoring a hundred then there is a mathematical way to compare the careers of say Tendulkar and Muralitharan to Kallis simply by adding the amount of runs scored by the respective players to number of wickets taken by each player multiplied by 20 (on my hypothesis a wicket is equal to 20 runs).  I have taken the top 5 run scorers and top 5 wicket takers in the game (noting that they are all players of the modern age given the amount of test played when compared to 80s and before) and come up with the following table:

Player Runs Wickets Combined Rating
Tendulkar 15921 46 16841 4
Ponting 13378 5 13478 6
Kallis 13289 292 19129 1
Dravid 13288 1 13308 7
Lara 11953 0 11953 8
Muralitharan 1261 800 17261 3
Warne 3154 708 17314 2
Kumble 2506 619 14886 5
McGrath 641 563 11901 9
Walsh 936 519 11316 10

It is pretty clear from this mathematic approach to considering the greats of the modern game that Kallis is head and shoulders above the rest. I apologise to all Tendulkar fans for saying this but when you look at his overall contribution across both key aspects of the game but Kallis hits him out of the park.  Indeed Tendulkar is not, statistically, even the second best of the modern generation of players.  Both of Muralitharan and Warne best him based on a combined consideration of the player’s contribution to both key skills of the game.

Having established who is the mathematically superior player between Kallis and Tendulkar it got me thinking about coming up with a way of measuring the efforts of the previous all-round greats who have preceded Kallis.  Because of the lack of test matches played during the period from the 1950s through 1980s (when the great all-rounders were playing) the methodology for comparison of the players has been necessarily tweaked to consider each players average contribution to each game in which the played.  I have not considered any player that pre-dates 1950 for this analysis and have come up with this list of all-rounders: Kallis, Sobers, Hadlee, Dev, Botham, Khan and Benaud.  I am sure I have missed some from this analysis (be happy to hear your thoughts on who).

Most pundits of late seem to box Sobers and Kallis together and then the rest in a contest to determine who the best all-rounder to play the game is.  Based on the mathematical approach I have outlined, that position seems to be short sited.  Here are the results of applying a statistical analysis to foregoing list of all-rounders:

Player Tests Runs Wickets Combined Ave / game Rating
Kallis 166 13289 292 19129 115.23 5
Sobers 93 8032 235 12732 136.90 1
Hadlee 86 3124 431 11744 136.56 2
Kapil Dev 131 5248 434 13928 106.32 7
Imran Khan 88 3807 362 11047 125.53 4
Botham 102 5200 383 12860 126.08 3
Benaud 63 2201 248 7161 113.67 6

This analysis shows that Sobers contributed most, statistically, to each game he played when compared to each of the other “great’ all-rounders.  Unsurprisingly, when you consider his bowling strike rate, Hadlee comes in a close second and, frankly, a long way back, Kallis comes in fifth.

This analysis is, of course, necessarily predicated on the pure numbers each player has “put up” and does not account, at all, for the intangibles each brings to the game.  That said, I am absolutely comfortable with the outcome of the analysis: it strikes me that Kallis is the leader of the pack when it comes to the modern greats of the game because he was a true double threat (and that is forgetting his acumen in the slips) and, as such, was constantly involved in the game.  Tendulkar, by comparison, oft spent his time in the field stationed at mid off and did not have anywhere near the work load of Kallis.

I did not have the pleasure of watching Gary Sobers play but those who did always bracket him at the second greatest player to play the game after Bradman.  I mentioned in the preamble that it is really impossible to compare players for different eras.  That said, I am prepared to conclude that Sobers must be considered in any consideration of the player next after Bradman.  The player though that sticks out for me as having been missed out in a lot of consideration of the “great” players of the game is Sir Richard Hadlee.  His contribution to each game is only minimally lower than that of Sobers and, it must be remembered, he was regularly playing a team that was a losing rather than winning.

All of this number crunching proves one thing really: Kallis has had an unbelievable career of excellence at the top level of the game for a long time.  I, for one, believe he gets the short shrift from many in comparison to Tendulkar which is just statistically wrong.

With no active player in the top 10 players I have noted above of the modern, I am left to wonder who the next “modern great” will be.  One suspects he will come from Kallis’ own team or that of Tendulkar but only time will tell.