The Night Watchman

Many a time, batting first or chasing is discussed in the forums and yet never are any real consensuses reached. Some pitches are said to be better to bat first, others bowl first, but such word of mouth is often not to be trusted. So, this article will examine a number of factors related to the toss and examine if that’s how real cricket works.

I’ll start by telling you about the data I’ll be using. I will use every competitive senior one day game my team (Greenfielders) have played. They include 12 cup games,  24 games from West Indies division 3.4 and 68 games from division 2.1. That means absolutely no selection has been made, it includes bots and games where I am hopelessly outclassed. To compare with real life, a source of real life data is needed and thanks to the wonders of cricinfo’s statsguru I can access that (so many thanks to them).

Since the toss should be random, lets examine if it is… of the 92 games recorded, I have won the toss on 44 occasions, losing on 48. Assuming the toss is a genuine 50/50, there is a 38% chance of 44 or less wins, far from insignificant, so this result suggests the toss does appear to have probability of 0.5. Some have theorised that a better captain will win the toss more, in my early seasons, my captain was almost always Shakir Jaber, he was an expert captain with high experience, so likely would win many more tosses than 50% were this theory correct, there is no evidence of this within my data, until Jaber’s sale, I won 26/58 (44.8%) tosses, since I have won 18/34 (53%), that change isn’t even significant at the 90% level. [Editor's Note: For those of you who are unfamiliar with statistics this is just a way of testing.]

But, of more interest, is the importance of the toss. In real ODIs there are 1415 instances of the team winning the toss then going on to win the game, 1369 lose despite winning the toss, that’s 50.8% win win the toss is won, far from a significant advantage (20% chance of that result, or more extreme, assuming the actual probability is 50%). So how does FTP fare?

To answer that question, we need to consider both teams in our games, so as to maintain a 50% chance of a win ignoring the toss (I have no ties, so for every win I get, my opponent loses and vice-versa). So for teams winning the toss then winning there are 45 instances for winning the toss and still losing, 47 instances. That’s a VERY slight bias towards the winning toss actually being a hindrance (there is only an 8.3% chance of the more balanced result). So, a big thumbs up for FTP there.

So, does the toss matter? from the data above, it would appear not, however there are effects from tactics that cannot be accounted for in these stats. But what about pitches? Is it true that its more important to win the toss on particular pitches? The table below shows the results of this division:

The most striking result is that on the slow pitch, where 11/13 (85%) of games were won by the toss winner, there is only a 2.2% chance of this or more extreme occurring, suggesting that this is a significant result. The effect on a flat pitch is less reliable due to the proliferation of bots using this pitch (the toss has almost no effect here as the bot will stand essentially no chance whatever). The uneven pitch appears to be the pitch to lose the toss on, but this is not a significant result (27% chance of this or more extreme a result).

So, there we have it. In this article I have concluded that overall, the toss has little to no statistical effect on the result of a game, but that winning a toss on a slow pitch could swing the odds in your favour. Many would have different tactics if batting first rather than second given the chance and will plan their orders to suit one. What this article has not done, is disprove any advantage in that approach. I’ve also seen no significant effect from the use of a better captain gaining more toss wins.

[Editor's Thought: It would be interesting to see whether picking to bat first or chase has any effect on each pitch as from reading this I feel it is probably manager's choice after winning the toss which removes any advantage that could be given by winning the toss]