Using statistics to test cricketing superstitions

15 June 2017

Oliver Stevenson
Oliver Stevenson

Oliver Stevenson has used statistics to find out whether there is any truth to the cricketing superstition known as the ‘nervous 90s’. This refers to the belief that players’ batting deteriorates as they near the milestone of scoring 100 runs – but Oliver’s recently completed masters thesis has found little evidence to support that contention.

“That’s not to say players don’t get nervous in the 90s,” says Oliver, a keen cricket player and follower. “Rather, for most players, nerves do not significantly affect batting ability while on scores between 90 and 99.”

Oliver played cricket competitively between the ages of 6 and 18, but now “it’s mostly social stuff with mates such as Twenty20 and indoor cricket – and watching international and domestic cricket”.

It’s the variable nature of the game that appeals, he says: “One day the game can be heavily dictated by the batsmen and the very next day by the bowlers. The contest between bat and ball is always fascinating – a bowler can bowl the exact same ball to a batsman a number of times, with each delivery producing a unique result. Additionally, cricket is one of few sports where external factors such as the pitch and weather can have a massive bearing on how the match is played out.”

Oliver wanted to combine his love of sport and statistics in his studies, and when Dr Brendon Brewer, a keen cricket fan, posted a notice that he was looking for a student to investigate the ‘nervous 90s’, Oliver jumped at the chance.

Oliver’s masters work has interested Cricket New Zealand to the extent that he will be doing a PhD in collaboration with the national body, with Brendon again his supervisor. The topic has not yet been settled, but will be around advanced player performance and contribution metrics, using statistics in areas of the game that are underutilised.

In general, says Oliver, cricket offers a lot of opportunities for statisticians. “Firstly, the large number of statistics collected during each game means we have a lot of raw data to work with,” he explains. “Secondly, the closed nature of the skills of batting and bowling ties in nicely with some of our statistical assumptions. Unlike sports such as rugby and football, this generally allows for the large amounts of data collected each match to be treated independently of the specific match scenario.”

Oliver, who is from Auckland, did a Bachelor of Science majoring in Statistics and minoring in Psychology at the University of Otago, and in 2015 completed a Bachelor of Science (Honours) in Statistics in the Department of Statistics.