Exploring probabilities at #WT20 World Cup

There is one game remaining in this year’s Women’s T20 World Cup. Australia will play England in the final at North Sound on 24 November.

I have been looking at some naive probabilities about the partnerships of winning and losing teams.

Before the tournament I had recorded data from ICC T20 games in 2017-2018 (n=58). These enabled me to estimate some prior probabilities before the World Cup started.  I have used the tournament website to collect data from the World Cup in the West Indies (n=20 completed games) to give me a set of posterior outcomes.

Winning Teams

Losing Teams

At this World Cup, winning teams have established their dominance with first wicket partnerships (9 of the 20 games have partnerships > 50 runs). Losing teams have had just two 50 run + opening partnerships. Losing teams have fourth wicket partnerships as an important contribution to their run totals.

At present, the median profiles for runs scored per wicket in the tournament is:

The median winning run total is 137 (range 81 to 194) and median losing total is 99 (range 71 to 160).

Photo Credit

North Sound (CricketHer, Twitter)

What kind of game? Run scoring patterns leading into #WT20

The 2018 ICC Women’s World Twenty20 is being hosted by the West Indies. There are three games on the first day of the tournament, 9 November.

I have data from 58 Women’s ICC T20 games played in 2017-2018 prior to the 2018 World Twenty20. I will track the games in the West Indies and compare them to the median profiles from the 58 games played. 26 of these games were won by the team batting first, 32 by the team batting second.

My profiles are:


Winning Teams

Losing Teams

Naive Priors

From these data, I have identified these priors to monitor outcomes in the tournament to track when runs are scored.

My hope is that these priors give an early indication of game outcome … and enable me to explore outlier game outcomes.

Photo Credit

Frame grab, Star Sports video (Twitter)