“One brings two”

I’ve spent too many hours watching test cricket wondering when the next wicket will arrive. Two batters, both set – how many runs can we expect them to accumulate at the crease? And, as the old adage goes, does one really bring two?

Firstly, how long does a top order partnerships tend to last? Surprisingly, the answer is not long; 25% of these partnerships make 8 runs or less.

If you define “one brings two” as a wicket before another 10 runs are scored, then the adage is true 29% of the time (first two bars on chart above) for the top order.

It’s also interesting to note how far above the average is from the median; the latter tells us that 50% of partnerships are less than or equal to 23 runs, whilst the former is skewed by big partnerships on the right of the graph.

Let’s revisit the scenario at the top; two batters are ‘in’, they’ve put on 30 together – when can we next expect a wicket? The next chart shows that 50% of the time they will add another 30 runs or less.

What’s surprising to me is that this isn’t a whole lot less than if a pair have already added 100 runs; in this instance 50% of the time they would add 36 runs or less. That is, the effect of a partnership already having been in for an additional 70 runs (100 – 30) is only worth 6 (36-30) more ‘additional expected’ runs.

A partnership is not much more ‘in’ at 100 than they were at 30; the early runs are much more valuable in this respect. Any cricket watcher or player will know this, but it’s interesting to put a number on it.

All data from Cricinfo Statsguru