How to measure a captain

It was the best of years, it was the worst of years. For Joe Root, 2021 was an annus mirabilis and horribilis at the same time. Those twin impostors, triumph and disaster, turned out to be conjoined. It wasn't just that he was often triumphant at the crease and disastrous at the helm: his captaincy alone kept contradicting itself. He ended the Ashes as the most successful captain England had ever had - and the most unsuccessful. When his team hammered India at Headingley in August, Root surpassed Michael Vaughan as the England captain with most Test victories - 27. As they left Adelaide in December, he had dislodged Alastair Cook as the England captain with most Test defeats - 23. By the end of the series, at Hobart, it was 25. Was he England's biggest winner, or their biggest loser, or both?

Actually, he was neither. Career aggregates are a poor guide to a captain's calibre: they have less to say about excellence than about longevity and fixture lists. At Sydney, he collected yet another record, outlasting Cook and becoming the first England captain to reach 60 Tests; Hobart was Root's 61st Test as captain in 55 months.

Modern sport is steeped in statistics. In cricket, it's no longer just the scores on the board - it's runs per 100 balls, percentage of dots, average against seam, average against this bowler. There's a flood of numbers coming at us all day long, and even flowing back into history. We can be told now that John Crawley's Test average against spin was 74. If the England selectors of the 1990s had known that, they might have given him a go in the subcontinent.

When it comes to batting and bowling, we are far better informed than ever before. The statisticians are getting there with fielding, too, and can show us that, in men's Tests, New Zealand have the best slips. But there is one department where the stats are still primitive: captaincy.

In the 50 years I've been following the game, nothing has changed. A captain's record pops up on the screen only when a landmark comes along. We end up applauding them for avoiding the sack. Captaincy is cricket's most cerebral element, the most analytical, and yet it is barely analysed. Well, you may say, it's impossible to pin down - it's about feel, instinct and inspiring the troops to run through a brick wall. The answer to that is: up to a point, Lord Botham.

Of course it's not all measurable, but neither is batting or bowling. Statistics still struggle to do justice to the unplayable over that spreads fear through a team, or the doughty 20 in dodgy light that sees off the new ball. The central stat remains the average, which tends to be worn like a badge, or a medal. We all know Bradman's Test batting average is 99.94, a figure so resonant it's almost a brand. We know a few others have finished on 60 (Graeme Pollock, George Headley, Herbert Sutcliffe).

Bowling averages are less famous, because of the age-old bias towards the bat, but the well-informed Wisden reader may recognise a few: ten is George Lohmann, 16 S. F. Barnes, 18 Frank Tyson, and just under 21 means Malcolm Marshall, Joel Garner or Curtly Ambrose.

In the women's game, there are some formidable figures waiting to be memorised: batting averages of 75 for Ellyse Perry and 59 for Enid Bakewell, bowling averages of 11 for Betty Wilson and 13 for Mary Duggan, who, like Garry Sobers, was both a fast left-armer and a slow one.

Averages force players to lug their career around, but they still have some use, which is why they have survived as a yardstick. We need something similar for captaincy. But what, in a captain's record, might resemble an average? It needs to be clear, clean and reasonably fair. It ought to shed light and ring true. And it should be inclusive, able to be worked out by a ten-year-old. First, let's look at percentage of matches won, with a minimum of ten Tests as captain. Here's the top ten:

MOST SUCCESSFUL TEST CAPTAINS BY WIN PERCENTAGE (P W %) 1 W. W. Armstrong (A) . . . 10 8 80.0 2 S. R. Waugh (A). . . . . . . . 57 41 71.9 3 D. G. Bradman (A). . . . . . 24 15 62.5 4 R. T. Ponting (A) . . . . . . . 77 48 62.3 5 W. G. Grace (E) . . . . . . . . 13 8 61.5 6 D. R. Jardine (E) . . . . . . . 15 9 60.0 6 F. M. M. Worrell (WI). . . 15 9 60.0 8 V. Kohli (I) . . . . . . . . . . . 68 40 58.8 8 Waqar Younis (P) . . . . . . 17 10 58.8 10 A. L. Hassett (A) . . . . . . . 24 14 58.3 10 Salim Malik (P) . . . . . . . . 12 7 58.3 Minimum: 10 Tests. All statistics are correct at February 8, 2022.

As in batting, several get close to 60, but not many climb higher. So hats off to Warwick Armstrong, the greatest of all time by this measure, even if he was a captain for only eight months in 1920 and 1921; to Steve Waugh, the greatest of the past century; and to Virat Kohli, the greatest of the past decade.

But hang on a minute - the top four are all Australians, which makes you wonder: is there a reason Aussies dominate? Yes, and it's not just that they're good at cricket. Of the ten nations and one region to have hosted more than one Test, Australia has the lowest incidence of draws, partly because all Tests there between 1882-83 and 1936-37 were played to a finish. The draw, much like hair, was big in the 1980s, before being trimmed to reasonable proportions. It accounted for 45% of all Tests in the 1980s, and only 19% in the 2010s.

Going by win percentage, you find a giant of old, such as Clive Lloyd of West Indies (48% - 36 wins in 74 Tests), being felled by a medium-sized figure from the 21st century, such as Faf du Plessis of South Africa (50% - 18 wins in 36), which doesn't seem right. So I tried putting the draw in a drawer, and looking at wins divided by losses. I raised the qualification too, to 25 Tests, because longevity does count for something, and heavier fixture lists bring more robust samples (Bradman's win/loss ratio was five, but his 24 Tests in charge came at the rate of two a year, and against only two opponents). The top ten now looked like this:

MOST SUCCESSFUL TEST CAPTAINS BY WIN/LOSS RATIO (1) (PW L W/L) 1 S. R. Waugh (A). . . . 57 41 9 4.55 2 J. M. Brearley (E). . . 31 18 4 4.50 3 I. V. A. Richards (WI) 50 27 8 3.37 4 R. Benaud (A) . . . . . 28 12 4 3.00 4 I. M. Chappell (A) . . 30 15 5 3.00 4 C. H. Lloyd (WI) . . . 74 36 12 3.00 4 R. T. Ponting (A) . . . 77 48 16 3.00 8 S. M. Pollock (SA). . 26 14 5 2.80 9 K. S. Williamson (NZ) 38 22 8 2.75 10 W. J. Cronje (SA). . . 53 27 11 2.45 Minimum: 25 Tests.

Again, there's a cluster. Several captains can manage three wins for every defeat, but only two can reach four: Waugh and Mike Brearley. This is not the last word on captaincy - Waugh had Glenn McGrath and Shane Warne, Brearley never led England against West Indies, and there's still a bias towards the land of few draws - but it sure beats win percentage. The list rings true, with Kohli toppled by Kane Williamson. And it sheds some light: who knew the only seamer would be Shaun Pollock, rather than Imran Khan?

If you stretch to a top 20, the Aussie dominance is diluted. There are seven Australians, six Englishmen, two South Africans, two West Indians, one New Zealander, one Indian and one Pakistani, who is still not the one you might expect. Root, by the way, is 41st out of 62, above Mohammad Azharuddin and Inzamam-ul-Haq, but below Rahul Dravid.

MOST SUCCESSFUL TEST CAPTAINS BY WIN/LOSS RATIO (2) (P W L W/L) 11 R. Illingworth (E) . . 31 12 5 2.40 12 M. P. Vaughan (E) . 51 26 11 2.36 13 V. Kohli (I) . . . . . . . 68 40 17 2.35 14 Javed Miandad (P) . 34 14 6 2.33 15 A. J. Strauss (E). . . . 50 24 11 2.18 16 W. M. Woodfull (A) 25 14 7 2.00 16 P. B. H. May (E) . . . 41 20 10 2.00 16 M. C. Cowdrey (E) . 27 8 4 2.00 16 M. A. Taylor (A). . . 50 26 13 2.00 20 S. P. D. Smith (A) . . 35 19 10 1.90 Minimum: 25 Tests.

So, of the 62 men who have been in charge for at least 25 Tests, only 19 have managed two wins for every defeat. If we get used to the win/loss ratio, this could become a marker, like averaging 50 with the bat or under 25 with the ball: a sign of success. Tests are not enough. This measure is going to have to work for white-ball cricket too.

MOST SUCCESSFUL ODI CAPTAINS BY WIN/LOSS RATIO (P W L W/L) 1 C. H. Lloyd (WI) . 84 64 18 3.55 2 R. T. Ponting (A)† 230 165 51 3.23 3 W. J. Cronje (SA) . 138 99 35 2.82 4 V. Kohli (I) . . . . . . 95 65 27 2.40 5 M. J. Clarke (A) . . 75 50 21 2.38 6 S. R. Waugh (A) . . 106 67 35 1.91 7 E. J. G. Morgan (E) 124 75 40 1.87 8 I. V. A. Richards (WI) 105 67 36 1.86 9 S. M. Pollock (SA)† 97 60 33 1.81 10 G. C. Smith (SA)† 150 92 51 1.80 Minimum: 50 ODIs. †Ponting captained an ICC World XI in one match, Pollock in two. Pollock also captained an Africa XI in two matches, Smith in one.

In an era when no other one-day captain stood tall, Lloyd comes top. Ricky Ponting overtakes Waugh, or builds on his success. Kohli is well ahead of Eoin Morgan. South Africa have as many entries in the top ten as Australia, and again Pollock shines.

As before, there's a cluster: 1.80 is very good going, but won't get you near the top five. The full list is 60-strong and, at No. 54, rubbing shoulders with Zimbabweans and latter-day West Indians, is an unexpected name: Sachin Tendulkar, on 0.53. Again, the list is plausible, even if Hansie Cronje's appearance is hard to stomach. If only yardsticks came with a moral compass.

MOST SUCCESSFUL T20I CAPTAINS BY WIN/LOSS RATIO (P W L W/L) 1 Asghar Afghan (Afg) 52 42 10 4.20 2 Babar Azam (P) . . . . 40 26 9 2.88 3 V. Kohli (I). . . . . . . . 50 32 16 2.00 4 F. du Plessis (SA)†. . 40 25 15 1.66 5 D. J. G. Sammy (WI) 47 28 17 1.64 6 E. J. G. Morgan (E) . 72 44 27 1.62 7 M. S. Dhoni (I). . . . . 72 42 28 1.50 8 A. J. Finch (A) . . . . . 59 32 25 1.28 9 K. S. Williamson (NZ) 56 27 26 1.03 10 Y Z K. J. Coetzer (Sco). . 41 20 20 1.00 10 W. T. S. Porterfield (Ire) 56 26 26 1.00 Minimum: 40 T20Is. †Du Plessis captained an ICC World XI in three matches.

Let's hear it for the minnows. In T20 cricket, Asghar Afghan is way out in front of the big fish and, if that's partly because he often faces even smaller fry, the same can be said of some more famous captains down the decades. For Twenty20, a format that is often 50-50, Asghar's record is phenomenal. To make this top ten, you just have to win as many as you lose, as Kyle Coetzer and William Porterfield have done, and Shahid Afridi has not. Kohli and Morgan are the only captains to shine in both T20s and ODIs, and nobody makes all three top tens.

Our yardstick also has to work for the women's game. Women's Tests are a challenge because they are so few; and three of the contenders are unbeaten. MOST SUCCESSFUL TEST CAPTAINS BY WIN/LOSS RATIO (P W L W/L) 1 L. A. Larsen (A). . . . . . . . . . . . . . . . . . 10 5 0 - 2 R. Heyhoe Flint (E) . . . . . . . . . . . . . . . 12 2 0 - 3 K. Smithies (E) . . . . . . . . . . . . . . . . . . 10 1 0 - 4 B. J. Clark (A) . . . . . . . . . . . . . . . . . . . 11 3 1 3.00 5 M. E. Hide (E) . . . . . . . . . . . . . . . . . . . 11 4 2 2.00 Minimum: 10 Tests.

Lyn Larsen is a worthy winner, but the two Englishwomen below her, admirable as they were, get too much credit for avoiding defeat. As in football, 3-1 is a better score than 1-0. But this is still a tough club to join, with Clare Connor and Charlotte Edwards among those missing out. And the solution is obvious: more women's Tests. The white-ball game works far better.

MOST SUCCESSFUL ODI CAPTAINS BY WIN/LOSS RATIO (P W L W/L) 1 M. M. Lanning (A) 66 57 8 7.12 2.10 2 B. J. Clark (A) . . . . 102 84 17 4.94 3 L. A. Larsen (A). . . 41 28 11 2.54 4 E. C. Drumm (NZ). 41 28 12 2.33 5 K. L. Rolton (A). . . 43 30 13 2.30 6 H. C. Knight (E). . . 60 40 19 2.10 7 C. M. Edwards (E) . 117 72 38 1.89 8 D. van Niekerk (SA) 50 29 19 1.52 9 M. Raj (I) . . . . . . . . 144 85 56 1.51 10 M. du Preez (SA) . . 44 23 18 1.27 Minimum: 40 ODIs.

Meg Lanning, take a bow. That score of 7.12 is beyond the wildest dreams of any other captain so far. And if Australians dominate this list, Lanning stands out even among her compatriots. I wonder if she'll do it again in T20…

MOST SUCCESSFUL T20I CAPTAINS BY WIN/LOSS RATIO (P W L W/L) 1 M. Musonda (Zim) 21 19 2 9.50 2 R. C. Belbashi (Nep) 22 17 5 3.40 3 M. M. Lanning (A) . 83 61 19 3.21 4 H. C. Knight (E). . . . 54 40 13 3.07 5 C. M. Edwards (E) . . 93 68 24 2.83 6 S. Tippoch (Tha) . . . 39 25 13 1.92 7 A. L. Watkins (NZ) . 29 19 10 1.90 8 H. Kaur (Ind) . . . . . . 65 39 23 1.69 9 S. W. Bates (NZ) . . . 64 39 24 1.62 10 J. M. Fields (A) . . . . 26 16 10 1.60 Minimum: 20 T20Is

Lanning is still there or thereabouts, but Mary-Anne Musonda is in a league of her own. On Twitter, she describes herself as a "tall glass of awesomeness", and this table backs her up. It also shows T20 doing its job, and promoting the smaller nations: Lanning is the last Aussie standing. So win/loss ratio seems to work for everything except women's Tests. It's time to see what happens if we compare captains from the same country: England, for instance.

MOST SUCCESSFUL ENGLAND TEST CAPTAINS BY WIN/LOSS RATIO (P W L W/L) 1 J. M. Brearley. . . . . . 31 18 4 4.50 2 R. Illingworth. . . . . . 31 12 5 2.40 3 M. P. Vaughan . . . . . 51 26 11 2.36 4 A. J. Strauss . . . . . . . 50 24 11 2.18 11 5 M. C. Cowdrey. . . . . 27 8 4 2.00 5 P. B. H. May . . . . . . 41 20 10 2.00 7 M. J. K. Smith . . . . . 25 5 3 1.66 8 E. R. Dexter . . . . . . . 30 9 7 1.28 9 N. Hussain . . . . . . . . 45 17 15 1.13 10 A. N. Cook . . . . . . . . 59 24 22 1.09 11 J. E. Root . . . . . . . . . 61 27 25 1.08 12 G. A. Gooch . . . . . . . 34 10 12 0.83 13 M. A. Atherton. . . . . 54 13 21 0.61 14 D. I. Gower . . . . . . . 32 5 18 0.27 Minimum: 25 Tests.

The effect is to make Brearley look even better, and to confirm Ray Illingworth, Michael Vaughan and Andrew Strauss as all-time greats. Graham Gooch is low down, but higher than his successor (Mike Atherton) and his immediate predecessor (David Gower). Captains from Essex stick close together (three between nine and 12), and Nasser Hussain gets some of the credit for Vaughan's success. And Root? On February 10, 2021, he was seventh with 26 wins, 15 losses and a score of 1.73. Then came ten defeats in 14 Tests. He suffered from bad luck, with injuries and illness and Covid bubbles, but also from bad judgment - playing four seamers on a turning pitch at Ahmedabad, disdaining a chase against New Zealand at Lord's, picking neither Stuart Broad nor James Anderson at Brisbane.

What changed, early in 2021? After the Indian tour, Ashley Giles decided England didn't need a national selector any more, and laid off Ed Smith. Selectors, perhaps even more than captains, lend themselves to win/loss ratios.

ENGLAND IN TESTS SINCE MAY 2018 (P W L W/L) Selected by Ed Smith . . . . . . . . . . . . . . . 37 21 12 1.75 Selected by Chris Silverwood . . . . . . . . . 11 1 7 0.14

Smith didn't appoint Root, but they made a good team. Since Smith's defenestration, England have been eighth in the world, behind Bangladesh (from, admittedly, a small sample). During his time, even after losing in India, England were third, ahead of Australia. The top two were New Zealand and India, in that order. If you have a win/loss ratio, you don't need a World Test Championship.

Tim de Lisle is a cricket writer for The Guardian and former editor of Wisden