Batting Hazard Rates in Test Cricket
Understanding when batters are most likely to lose their wicket.
If you’ve ever studied for actuarial exams, the little orange book of mortality tables is like an old friend. It contains the answers to some of life’s great questions, like the odds of a man on his 92nd birthday making it to 931.
I created the equivalent set of hazard rates across a Test cricket innings. Here’s how you read them: given that a batsman has made N runs (the x-axis), what is the probability of them being dismissed before the next run (the y-axis)?
There’s a clear “excess mortality” bump between 90 and 120, owing first to the nervious nineties, and then to the proclivity to throw your wicket away after reaching 100 through either boredom or fatigue. This table lists the players with the highest conversion rate from 90 to 120, given a minimum of fifteen opportunities.
| Player | Innings of 90+ | % where reached 120 |
|---|---|---|
| DG Bradman | 27 | 88.9% |
| JE Root | 21 | 85.7% |
| WR Hammond | 20 | 80.0% |
| CA Pujara | 18 | 77.8% |
| Younis Khan | 32 | 75.0% |
| KC Sangakkara | 39 | 74.4% |
| GC Smith | 28 | 71.4% |
| PA de Silva | 20 | 70.0% |
| BC Lara | 40 | 70.0% |
| GS Sobers | 26 | 69.2% |
In a word, ruthless. These cricketers aren’t slaves to the base-10 number system.
The other end of the spectrum seems to comprise opening batsmen, romantic dashers or occasionally, both.
| Player | Innings of 90+ | % where reached 120 |
|---|---|---|
| KD Walters | 15 | 20.0% |
| PBH May | 16 | 25.0% |
| AI Kallicharran | 17 | 29.4% |
| RR Sarwan | 17 | 29.4% |
| DR Martyn | 15 | 33.3% |
| MJ Slater | 23 | 34.8% |
| MC Cowdrey | 24 | 37.5% |
| ME Waugh | 20 | 40.0% |
| TW Graveney | 15 | 40.0% |
| G Boycott | 24 | 41.7% |
I used Python’s BeautifulSoup, pandas and Plotly libraries for the web scraping, data manipulation and visualisations respectively. Check out the Jupyter notebook here.
1 It’s only 80.3%. Don’t buy green bananas.