But are the initial findings, that large variations in the 2-ball analysis or the over-level analysis by frequency do not favour the bowler, really that surprising? It must be so much harder for the bowler to disguise the delivery that is much quicker, or slower, than the previous one, also harder to control it — big variations are easier for the batter to spot, and likely to be easier to hit.
Thanks for pointing this out. I had considered that there is heterogeneity in how bowlers conceal their variations but obviously we do not have the data to test this out. Two things though: the conclusion that pace variation for the sake of it doesn't work is not hampered by this possibility, and secondly the fact that even small variations (which are presumably better concealed) do not necessarily cause a drop in batting numbers indicates that there is more at play.
I think, perhaps, I was thinking about pace variations in a different, “old school”, context, where the slower ball is intended to induce the batter to play early, a quicker ball to burst through the batter’s stroke — changes of pace as wicket-taking options.
But in the modern game, variations (of pace and spin) are deployed as much to defeat pre-meditated attacking options as to take wickets. It has been commented on elsewhere, that the old virtues of consistent line & length simply feed batters who are committed to hitting the ball.
More important than ball-by-ball variation might be the wider variability of pace across an over or a spell, forcing the batter to hesitate, if even only for a split second, before launching into another big hit.
Fascinating analysis.
But are the initial findings, that large variations in the 2-ball analysis or the over-level analysis by frequency do not favour the bowler, really that surprising? It must be so much harder for the bowler to disguise the delivery that is much quicker, or slower, than the previous one, also harder to control it — big variations are easier for the batter to spot, and likely to be easier to hit.
Thanks for pointing this out. I had considered that there is heterogeneity in how bowlers conceal their variations but obviously we do not have the data to test this out. Two things though: the conclusion that pace variation for the sake of it doesn't work is not hampered by this possibility, and secondly the fact that even small variations (which are presumably better concealed) do not necessarily cause a drop in batting numbers indicates that there is more at play.
I think, perhaps, I was thinking about pace variations in a different, “old school”, context, where the slower ball is intended to induce the batter to play early, a quicker ball to burst through the batter’s stroke — changes of pace as wicket-taking options.
But in the modern game, variations (of pace and spin) are deployed as much to defeat pre-meditated attacking options as to take wickets. It has been commented on elsewhere, that the old virtues of consistent line & length simply feed batters who are committed to hitting the ball.
More important than ball-by-ball variation might be the wider variability of pace across an over or a spell, forcing the batter to hesitate, if even only for a split second, before launching into another big hit.
Very interesting thought on the variance of speeds in an over being the key metric to focus on. I might write a follow-up article looking at this.