Nobody scores at a career strike rate in the fourteenth over

A single strike rate averaged across every phase describes a player who does not exist and hides the only comparison worth making.

Strike rate is quoted like a temperature, one number per player, and it is the wrong shape for the thing it describes. Scoring rate is not a property of a batter. It is a property of a batter in a phase, against a bowling type, at a match state, and averaging over all of those produces a figure that corresponds to nothing that ever happened on a field.

Take two players who both finish a season at a hundred and thirty. One made his runs opening, against the new ball, with two fielders out and a hard ball coming on nicely. The other came in at the fall of the third wicket in the thirteenth over, against spin, with five men on the rope and a soft ball that was gripping. The second innings is dramatically more difficult, and the numbers say they are the same player.

Worse, the aggregate hides the shape of a career. A batter whose rate has been falling for two seasons in the middle overs while rising at the death looks stable in total, and what is actually happening is that he has lost the ability to manoeuvre spin and is compensating by hitting harder later. That is a decline in progress and the headline number is blind to it.

The fix is not complicated and the data exists. Report scoring rate by phase, or at minimum by over band, and stop putting a career figure on a graphic as though it means anything. Broadcasters resist because a single number fits in a box, and there is a real audience cost to a graphic with six numbers on it.

I would still take the six numbers. The audience for cricket statistics is more patient than television believes.

Where I think the phase argument gets overdone is with very short samples. Split a season finely enough and you have thirty balls in a bucket, and thirty balls tells you nothing at all, so the split has to be coarse and honest about noise.

The version that annoys me most is the comparison across eras using a raw figure, as if a rate compiled when a hundred and forty was a good total means the same thing now. It does not. The bats are different, the boundaries are different, the fields are different. The number survived. The game underneath it did not.