Plans built only from the evidence.

Right-handed batters: average of 34.8, 3.4 above their average of 31.5 on all balls bowled (+11%) — allowing for sample size, that change is in the bottom 21% of comparable players (their median change: -1%) — 559 balls bowled.

Sample: 559 balls · medium confidence

Middle (11-40): strike rate of 82.5, 41.3 above their strike rate of 41.2 on all balls bowled (+100%) — allowing for sample size, that change is in the bottom 3% of comparable players (their median change: +12%) — 330 balls bowled.

Sample: 330 balls · low confidence

West Indies: average of 35.6, 4.1 above their average of 31.5 on all balls bowled (+13%) — allowing for sample size, that change is in the bottom 27% of comparable players (their median change: -4%) — 280 balls bowled.

Sample: 280 balls · low confidence

At home: average of 35.6, 4.1 above their average of 31.5 on all balls bowled (+13%) — allowing for sample size, that change is in the bottom 28% of comparable players (their median change: -3%) — 280 balls bowled.

Sample: 280 balls · low confidence

Powerplay (1-10): strike rate of 31.6, 9.6 below their strike rate of 41.2 on all balls bowled (-23%) — allowing for sample size, that change is in the top 3% of comparable players (their median change: +15%) — 505 balls bowled.

Sample: 505 balls · medium confidence

Powerplay (1-10): economy of 3.74, 0.84 below their economy of 4.59 on all balls bowled (-18%) — allowing for sample size, that change is in the top 20% of comparable players (their median change: -6%) — 505 balls bowled.

Sample: 505 balls · medium confidence

Left-handed batters: average of 27.8, 3.7 below their average of 31.5 on all balls bowled (-12%) — allowing for sample size, that change is in the top 30% of comparable players (their median change: +2%) — 388 balls bowled.

Sample: 388 balls · medium confidence