Pushing the (sealed) envelope: An alternative to the coin toss

Originally published in 2014

Anton Chigurh, the villain of Cormac McCarthy’s No Country for Old Men, plays a simple game with those unlucky enough to cross his path. One toss of a coin decides their fate: heads they live, tails they die. His brutality is fearsome indeed: a potent cocktail of death and recreation bearing all the hallmarks of a psychopath.

The ICC has come in for criticism of all sorts during its 105-year existence, but no one has so far accused its officials of such cold-blooded malice. And yet there are pitches around the world on which captains, waiting in suspended animation as the coin falls, must see in the shadow of the watching match referee the spectre of Chigurh. Lose the toss, lose the game. No losing captain in today’s media-savvy era would ever go so far as to say that the toss decided the match; plenty, of course, would want to.

It is worth examining the effect of the toss in the context of a proposed alternative. Under a “sealed bid” system – to be discussed later – the winning captain has choice of innings but must concede a pre-declared number of runs to the opposition.

The toss has, over the entire history of Test cricket, provided a slight but clear advantage to the winner. Of the 1,397 matches that have produced a winner, 737 of those (53 per cent) were won by the team winning the toss, and 660 (47 per cent) by the team losing it. (This is a statistically significant ratio at the 95 per cent confidence level, suggesting the perceived advantage is unlikely to be the result of natural variance alone.)

He believes fielding first is worth at least an early wicket, which against Australia is somewhere in the region of 45 runs. He also knows that, since 1970, the toss has been worth a mammoth 71 runs on average at Lord’s, with the captain calling correctly having won 59 per cent more Test matches. Most of them, though, batted first and Clarke has never gone any higher than 25 in the bid. For Cook, it would be folly to give away more runs than he needs to. In the end, Cook bids 42 runs, keen to beat away a bid of 40 from Clarke, or a possible bid of 41 designed to beat his own 40 bid.

In the other dressing-room, Clarke has already made up his mind. Maybe it’s a bowling morning, but there are five days to this Test, and all this statistical stuff is pure ivory-tower nonsense. He knows Cook is likely to want to bowl first given the conditions and the result at Cardiff. Clarke bids zero – a strong statement of intent and faith in his batsmen. It has the added benefit of making Cook look foolish for giving away all of those runs for free, when he could have won with a bid of just 1 run. First blood to Australia, who effectively start the game 42 for 0.

The toss of a coin to decide the start of the game is more than just traditional – it is poetic, and a great leveller. No matter how large the gap between teams, the match starts with an act that brings the captains together in a moment of equal chance and, as the coin falls, the audience watches on in hope. But in an era where many wonder where Test cricket will be in 20 years’ time, we need intrigue and debate, not poetry and tradition. Sealed bids are a rare way of making our great game better while courting the controversy that the sport needs to survive. Perhaps, in 100 years’ time, this new method of starting the game will be seen as just as traditional as the coin toss is today.

Anton Chigurh’s game of chance is ill-suited to cricket, which is, at its heart, chess on grass; a test of skill. The world of modern cricket is no country for old men: let the traditional coin-flip die.

Figures correct up until June 30, 2014

October 21 2019: We have been made aware of a similar proposal – the NBmethod – made by Nagesh Bharadwaj prior to publication of this article. While the author was unaware of this previous proposal, we wish to credit Nagesh’s work. Click here to read his proposal.