THERE is a certain type of sporting conversation, usually begun in a pub after the third pint of beer, that starts with the phrase “who was the greatest…ever?” Such discussions, comparing teams and sportsmen from different eras, have the benefit of being both hypothetical and intractable, making them perfect for filling the long hours until time at the bar. 

One variant which keeps cricket fans in the pub for longer than they ought, is “who was the greatest captain the game has seen?” Cricket captains are unlike those in most other sports, because they dictate tactics on the field and are thus crucial to the success of a team. It is the captain who decides, for example, how to rotate the bowlers, where to place the field for each delivery and to what extent the team should attack or defend. So important are they that even when they contribute little to a side with bat or ball they are sometimes still included purely for their tactical nous. Mike Brearley, who skippered England between 1977-1981, never scored a Test century and had a paltry batting average of 22.88. Yet he was such a good captain that many doubt whether England would have secured their famous 1981 Ashes win without him at the helm. (He hadn't actually started the series as captain, but was called in to replace Ian Botham—the team's superstar, but a lousy leader—after it found itself 1-0 down after two Tests.)

But how can one meaningfully compare Mr Brearley to other skippers throughout the ages, such as Andrew Strauss, in charge of today's number one ranked side, England, or Bill Woodfull, captain of the great Australia side of the 1920s and 1930s, or perhaps even that 19th century behemoth, W.G. Grace?

To the possible chagrin of pub pundits everywhere, the question exercised Satyam Mukherjee, a scientist at Northwestern University in Illinois. He thought he could give a definitive answer of cricket's greatest captain by using what he calls a “complex network approach”. In a paper (which is currently under review) he used Google's PageRank algorithm to compare captains' results between 1877-2010. The advantage of Mr Mukherjee's approach is that it doesn't just rank the skippers on the number of games they won. Rather, each captain's head-to-head results are compared against all others in a matrix. In this way the algorithm can measure not just whether they won or lost, but the quality of the win. Hence, a successful captain beating an unsuccessful one does not affect either's score by much. But when that situation is reversed, there are plenty of points at stake. 

So who is the most successful captain in Test history? According to Mr Mukherjee, it is Steve Waugh (pictured), captain of Australia between 1999-2004. It is difficult to disagree with his finding. Mr Waugh holds the record for the most consecutive Test wins at 15 (the team itself was unbeaten for 16). Indeed, Australians dominate the top 10 (see table). More surprising is that Clive Lloyd, whose West Indian team of the 1970s and 1980s is generally thought to be at least the equal of Mr Waugh's, only comes in at number six. Sir Viv Richards, who succeeded Mr Lloyd, is an even lowlier 13th, trumped by far less lauded skippers such as 9th-placed M.S. Dhoni and 10th-placed Nasser Hussain (many remember Mr Hussain for being regularly thwacked by Steve Waugh's side, which makes his captaincy record appear blotted, but in fact lost him few points in this ranking). 

Mr Mukherjee also used his algorithm to work out the most successful Test playing nation of all time. Although the International Cricket Council, the game's governing body, produces its own ranking of teams, it does not disclose the methodology, and Mr Mukherjee doubts its robustness. Australia tops his list, with South Africa second and England third (for the early parts of their Test histories, teams such as India and the West Indies weren't much cop). He also calculates the best side from each decade. 

If all this has spoilt the fun of many a heated debate, then at least cricket fans can still squabble about which was the best individual team from across history. But better do it quick. Mr Mukherjee's next plan is to run his algorithm on teams from each decade. He says he suspects that the West Indies side of the 1980s will trump the recent Australians as the greatest ever. Still, it might just throw up a surprise worth arguing about.