The short forms of cricket have been formulated with the fan in mind. Eliminating draws from the short form has been one of the features of the new formats.
To some extent, this has been done in much the same way as it is done in other sports – by creating a separate mini-game that can be used as a tie breaker. But cricket has also gone beyond any other sport I’m aware of by using some fairly complicated statistics to infer a winner from the circumstances of an incomplete match.
Breaking Tie Scores
Fewer runs are scored in short-form cricket than in first-class cricket, but there are still usually well over 100, so exact ties remain unlikely. But, as with several other sports, there are tie-breaking techniques such that even the occasional tie is not tolerated. As in some other sports that feel the need to eliminate ties, this takes the form of a separate mini-game, played at the conclusion of an otherwise tied match, to decide the winner.
Perhaps the best-known example is the practice of deciding a soccer match with penalty kicks. Short-form cricket sometimes uses a somewhat analogous procedure called a bowl-out, in which bowlers bowl at an unguarded wicket. But this practice has mostly given way to the super over, which is, in effect, a tiny new one-over game, more analogous to the other soccer tiebreaker, extra time.
Soccer’s penalty shoot-out is widely derided as no better than a coin flip, but the procedure is entrenched, now, because it does what it’s supposed to do. It breaks ties, and it does it in a way that makes for great theater. The cricket super over has the same virtues.
Inferring a Result from an Unfinished Game
A greater risk to achieving a result is that even a shortened match might be interrupted by rain or some other condition that stops play. Matches that are incomplete are not called draws, but for all intents and purposes, the outcome is the same – each side gets half of the points that would otherwise go to the winner. If the tournament is in an elimination phase, the match simply needs to be replayed.
In baseball, if a game is truncated by rain and more than half the game has been played, the victory can (subject to complicated conditions I won’t go into) be declared for whichever team was ahead when play stopped. But this straightforward rule does not transfer gracefully to cricket because in cricket it’s very hard to know, at least in close cases, which team is ahead.
Still, the need for such a method was strong enough that a number of manifestly unfair methods have been used. In the late 1990s, a pair of English statisticians named Duckworth and Lewis invented a method that has been widely adopted, and continuously refined. Duckworth and Lewis themselves have retired, and given their method into the care of an Australian statistician named Stern, so the current system is known as the Duckworth-Lewis-Stern method.
Duckworth and Lewis wrote a charming book about the development of their method. As one not shy about inventing statistical methods for assessing difficult concepts in this general domain, I recognized kindred spirits. Their breathless accounts of how their work enabled them to meet some of their personal cricketing heroes is particularly sweet.
The Duckworth-Lewis method is not simple. In its most advanced form, it is available only as embodied in proprietary software. (At one time, the software was not even available to ordinary civilians – it was licensed only to cricket governing bodies.) The mere mention of Duckworth-Lewis tends to elicit helpless shrugs from cricket fans. And yet, though there are a smattering of dissenters who claim to have come up with even better methods, most of the cricketing community seems to have made its peace with Duckworth-Lewis.
It’s hard to imagine such a thing happening in an American sport. Analytical methods have become much more mainstream since the success of the Oakland A’s recounted in Moneyball, but it would be quite another thing to incorporate a statistical procedure that almost no one understands into the rules of the sport itself.
One thought on “Super Overs and Duckworth-Lewis-Stern”