This is the companion article to the one on most valuable batters published last month. It follows similar analytical ideas, with minor variations. I will repeat the idea behind this pair of articles again, so that readers are not confused. Early next year, I intend to present a pair of articles about the best Test batters and best Test bowlers across teams. Those will be based on comprehensive, complex, context-centric analyses. This article and the one before it are curtain-raisers, using non-contextual analysis and base measures.
In the forthcoming best-bowlers article, context will be everything. In addition to the five basic measures enumerated below, I will look at the quality of batters dismissed, when they were dismissed, overall batting strength, type of pitch, support received, innings/match status, location, results, peer comparisons (within and without), Player-of-the-Match performances etc.
My apologies in advance for the batting-centric examples to follow, but if you think Don Bradman should be ahead by a mile on the batters list, remember that he has accumulated only 45% of the highest run tally, which is a key measure in this analysis. If you think that VVS Laxman should be a shoo-in, compare his average with that of Virender Sehwag. If you think that Hanif Mohammad is placed very low, do not forget that he scored only 3915 runs at 44. And so on.
This explanation has been presented in view of the responses I received for the previous article. May I specifically request readers not to question the importance given to the key measure, the career wickets aggregate? That is a deliberate decision, as far as this analysis, quite limited in its scope, is concerned.
Based on the comments received for the last article, and my own reading of the measures, I have made the following minor tweaks to the bowling methodology, as compared to the batting one.
1. I have reduced the relative weight for the Wickets measure by a fourth. This is fairer to bowlers who played in earlier eras. The weight is now 3.0 instead of 4.0.
2. I have changed the Bowling Average measure significantly. The Batting Average measure is a single unified one. My Weighted Batting Average is an improvement on that, but ultimately follows the same principle - that of runs per innings. On the other hand, the bowling average is a composite of two distinctly different measures - the bowling strike rate and bowling accuracy.
3. In mathematical terms, Bowling Average = Runs/Wickets. This can be represented by the alternative equation, Bowling Average = (Runs/Balls) x (Balls/Wickets). The balls component gets cancelled out but has clear relevance when the two measures are considered independently.
4. So I have split the bowling average into its two components. Among these two measures, the strike rate is far more significant, since taking wickets at a higher frequency is critical. Hence the weight for the strike rate is quite high. The weight for wickets (3.0) is lower than the combined weights for strike rate and accuracy (3.5).
I have used the five measures below, the values for which are basic for each bowler and easily obtainable in the public domain. A maximum total of ten points are allotted across these measures for each bowler. I have shared all the data for this in an Excel sheet so that anyone interested can play around with differing weights. I considered, and discarded quickly, the idea of having differing weights for different teams. It would not have made a difference to the final placements, which are based on relative weights, anyhow.
3.0 points: Career Wickets
This is a measure of how a bowler performs his main role: taking wickets. Everything else is secondary. This is a longevity measure but I have worked on the basis that capturing wickets is the main responsibility of any bowler. Everything else is secondary. The maximum points are allotted for 850 wickets. (Current high value: 800 by Muthiah Muralidaran).
2.5 points: Bowling Strike Rate
This is a true performance measure and gets a high weight. The maximum allotment is 35 BpW. (Current best value: 39 by Kagiso Rabada)
1.0 points: Bowling Accuracy
This too is a performance measure but a secondary one. A tight bowling accuracy of 1.50 runs per over gets the maximum allotment. (Current best value: 1.67 for Tom Goddard)
2.0 points: Wickets per Test
Another performance measure, one that is subtly different from the average. The maximum value here is 7.5 wickets per Test. (Current high value: 7.0 by Sydney Barnes)
1.5 points: % of Team Wickets
A reflection of the bowler's contribution to the team. This is the percentage of all the wickets taken by the team, including run-outs and other non-bowler dismissals. The full value is 40.0% share of team wickets. (Current high value: 38.6% for Muralidaran)
For Australia, the qualifying mark for a bowler to get into the mix is 125 Test wickets, which is the tally for around 30-35 Tests. For England, the qualifying mark is 150 Test wickets, which translates to about 35-40 Tests. For Zimbabwe, the qualifying mark is 70 Test wickets. For the other seven teams, a bowler has to take 100 or more Test wickets. It pains me to leave out George Lohmann. However, reducing the cutoff to 100 wickets for England would dilute the selections a lot, and special treatment for this great bowler would be unfair. Suffice to say that he stands very tall indeed. In fact, he would have been in the top five, despite a career haul of only 112 wickets.
For Australia, 29 bowlers qualify.
A banker to start with. Shane Warne leads the elite quintet comfortably with excellent all-round numbers. Glenn McGrath, whose average is an outstanding 21.6, comes in in second place. No one should be surprised by the presence of that great bowler, Dennis Lillee, in third place, helped by a very good average of 23.9. Nathan Lyon is fourth, with his huge tally of wickets. Clarrie Grimmett is a bit of a surprise in fifth position. One reason for his inclusion is that he has a relatively low average of 24.2, for a spinner. Pat Cummins just misses out, but could well move into the top five as he plays more Tests. The average for these five bowlers is a high 481 wickets apiece, though at at a middling bowling average.
On to England next, where 25 bowlers make the rather high cut-off of 150 wickets.
Sydney Barnes, with performance figures that put him in the top two on almost all of the measures, makes up for a low haul of 189 wickets to top the table. Jimmy Anderson is in second place, his 700-plus wickets helping to make up for some mid-level performance numbers. Stuart Broad, who bowled with Anderson in Tests, is comfortably in third position. Both Anderson and Broad have taken only slightly more than 20% of the wickets England took during their time, indicating that they received good support. Fred Trueman, his outstanding average of 21.5 making up for a relatively low value of 307 wickets, is in fourth place. Ian Botham, the allrounder par excellence, completes the top five. Alec Bedser misses out by less than a tenth of a point. England's top five have an overall average that almost matches Australia's.
South Africa, hampered by an enforced absence of a quarter century from international cricket, can only find 17 bowlers who reached the cutoff.
As expected, Dale Steyn leads. Over 400 wickets at a sub-23 average is the reason why. Kagiso Rabada, who has the best strike rate among the 170 bowlers selected in this exercise, is in second place, followed by the great Allan Donald. Then comes that magnificent allrounder, Shaun Pollock. The top four all have averages at or below the 23 mark. Makhaya Ntini completes the five. Morne Morkel just misses out. Their average is a tally of nearly 400 wickets but at a very good bowling average, under 24.
A group of 24 West Indies bowlers, mostly of the faster variety, like with South Africa, reached the three-figure mark on wicket aggregate.
That all-time-great bowler Malcolm Marshall just pips Courtney Walsh to top place. The others in the top five are Curtly Ambrose, Joel Garner, and Michael Holding. Together they form a group of five bowlers whose average career haul is 362 wickets at an average of 22.34. That average is the best among the ten groups of bowlers featured here.
Not having played many five-Test series, New Zealand only have 15 bowlers who make it inside the qualification enclosure.
Richard Hadlee is on top, ahead by multiple laps. Trent Boult could have closed the distance to Hadlee significantly if he had not quit red-ball cricket when he was at his peak. Tim Southee and Neil Wagner follow next, and the still-active Matt Henry completing the featured five. Chris Cairns just misses out. The average tally is 300-plus wickets at a reasonable average.
With their many five-Test series through their team's near 100-year existence, no fewer than 24 Indian bowlers qualify.
An engrossingly close fight between two great spinners from the south of India sees R Ashwin just edge out Anil Kumble, the five-run difference in average more than making up for Ashwin's smaller number of wickets. No one should be surprised to see Jasprit Bumrah, with just above 200 wickets, upstage those with nearly double his tally. His remarkable sub-20 bowling average is the jewel in the crown.
The two other 400-plus-wicket warriors from India, Harbhajan Singh and Kapil Dev, complete the quintet. I am glad that the incomparable Kapil Dev made it to the top. Ravindra Jadeja, in sixth place, is almost certain to go past Kapil in the next few Tests. Next are Bhagwat Chandrasekhar and Bishen Singh Bedi. A quartet of spinners dominating the top six is as much an ode to spin as the West Indian group was to pace bowling. The wickets tally is the highest of all the teams here, but at a relatively high average, for obvious reasons.
Twenty Pakistan bowlers reach the cutoff despite their relatively late entry into international cricket.
Waqar Younis just about pips Wasim Akram and Imran Khan. These three bowlers are within 0.08 points of each other. It is as if they are tied together with an invisible chord. A slight change in weights would put Akram or Imran on top. They are that close. A better fast-bowling trio is impossible to find - the variety is amazing. Then come two spin bowling greats with not much to distinguish them. Yasir Shah has captured more wickets but at a greater cost. The unlucky bowler to miss out is Mohammed Asif, despite a low aggregate of wickets. Some great Pakistani bowlers follow. The average for these five is 300-plus wickets at a very good average.
With Muthiah Muralidaran taking nearly 40% of all wickets taken by Sri Lanka, it is no wonder that only nine Sri Lankan bowlers make the grade.
Muralidaran leads by the proverbial mile, not just among Sri Lankan bowlers but all bowlers. Not only did he take 800 wickets, he also had a career average below 23, a pace bowler's average, which makes him arguably the greatest Test bowler ever.
Rangana Herath is comfortably ahead of Prabath Jayasuriya at No. 2, although the latter could close the gap in years to come. Chaminda Vaas is the lone pace bowler to break the spin hegemony, completed by Dilruwan Perera. Lasith Malinga just misses out, having made the cut by the skin of his teeth. The average tally is 374 - mainly because of Muralidaran's wickets. And at a reasonable cost.
Zimbabwe have a small contingent of three Zimbabwe bowlers, despite the lowering of the cutoffs, led by Heath Streak. He took as many wickets as the next three bowlers together did. Ray Price and Paul Strang come in next.
Taijul Islam, the unassuming left-arm spinner, leads the quartet of Bangladeshi bowlers who make the cutoff. He is still active and is almost certain to become Bangladesh's highest wicket-taker. Shakib Al Hasan is second, closely followed by Mehidy Hasan Miraz, who ought to move to second place soon.
The complete data for all qualifying bowlers is in this Excel file.
Seven current players (Lyon, Rabada, Henry, Bumrah, Jayasuriya, Taijul, and Mehidy) are in the 47 bowlers featured in the top fives in this article. (Jadeja could get in, but he might also retire soon.) That is a fairly decent representation. Overall there is fairly good representation across the eras. The high performance measures of the older-generation bowlers compensates for their lower wicket-aggregate values, to a certain extent.
Below are the five bowlers who have secured the most points, across the teams. Possibly a pointer to the best Test bowler analysis. Of course, context will play a very important role in that analysis. Note how far ahead of the closest competitor Muralidaran is. I would be very surprised if he and Warne are not in the top three of that forthcoming analysis.
- 8.43 Muttiah Muralidaran
- 7.24 Shane Warne
- 6.94 Sydney Barnes
- 6.85 R Ashwin
- 6.76 Glenn McGrath
Potpourri
This time's collection, a fairly long one, is a smattering of extreme averages in different areas, mostly featuring little-known players.
- Kurtis Patterson: 2 Tests, 144.00 ave, 1 hundred
How did Patterson play two Tests, average 144.0 , and never get selected again? Unfortunately, that is the hard truth when one is an Australian batter.
- Afaq Hussain: 2 Tests, 66 runs, 0 dismissals.
Hussain played four innings for Pakistan, was never dismissed, accumulated 66 runs, and then was never picked again. He does not have a batting average, to boot.
- Roshan Jurangpathy: 2 Tests, 4 innings, 1 run, 0.25 ave
Poor Jurangpathy played two Tests for Sri Lanka against India, scored a single, got out four times, had an amazing average of 0.25, and not unexpectedly, was not selected again.
- Naeem Islam: 95.4-8-303-1
Naeem played eight Tests for Bangladesh, bowled nearly 100 overs, and took a single wicket for over 300 runs. That wicket came in his first over in Test cricket, after which he bowled 94 more overs for no wicket. Not to forget that he was primarily a batter.
- Anwar Hossain: 58.0-5-307-0
Anwar Hossain Monir of Bangladesh performed even more poorly than Islam. He played three Tests, bowled 58 overs for 307 runs, and conceded 95, 110, and 102 runs for no wickets in his three Tests.
- England: 517 for 1 vs Australia
In Brisbane in 2010, England, facing a first innings deficit of 221, went berserk. Andrew Strauss and Alastair Cook added 188 runs for the first wicket, and then Cook and Jonathan Trott added an unbeaten 329 runs for the second. Then England declared. That is an amazing 517 runs per wicket.
- England: 849 all out
In 1930, England, led by an Andy Sandham triple-century, scored 849 runs, and were bowled out. So they averaged nearly 85 per wicket - the highest in a completed innings.
- New Zealand: 26 all out
In 1955, in Auckland, New Zealand scored 26 in their completed innings, averaging 2.6 runs per wickets - the lowest ever.
Talking Cricket Group
Any reader who wishes to join my general-purpose cricket-ideas-exchange group of this name can email me a request for inclusion, providing their name, place of residence, and what they do.
Email me your comments and I will respond. This email id is to be used only for sending in comments. Please note that readers whose emails are derogatory to the author or any player will be permanently blocked from sending in any feedback in future.