ICC Cricket World Cup 2019 - Opta's prediction model

Posted 28th May 2019

CWC Predictor model header.png

Here we use a predictive model built by Opta’s data science team to preview the 2019 ICC Cricket World Cup.

Opta is committed to advancing sports data usage across media, professional teams, governing bodies and sport at large. Our data scientists work across sports, building analytical metrics and predictive models to bring deeper understanding to the on-field action, utilising our extensive historical database as well other data sources.

Get in touch with us here to find out how we can help you bring new insights to your audiences.

With the World Cup commencing this Thursday across English and Welsh soil, a number of questions will begin to be answered. Who will win the tournament? Who might spring a surprise? Will we see a new world-record run total or even 500 on the board?

Whilst we await definitive answers, Opta’s World Cup predictive model can guide us in the interim. Our model assesses the batting and bowling strength of each team since the 2015 World Cup and uses these to generate probabilities for each match based on the comparative strengths of the teams and the venue. The match probabilities are then used to map out the progress of each team throughout the tournament based on 100,000 possible simulated timelines.

You can read more about our model here.

The group stage

The single group format of the tournament, in which each of the ten teams plays each other once, favours consistency over the odd mercurial performance, while lessening the impact of a single abject display. This is reflected in our group stage projection, where the most highly rated teams by our model are strongly fancied to progress.

England and India are given a 76% and 75% chance of qualifying respectively, with South Africa (66%) and Australia (65%) rated as most likely to join them in the semi-finals. Should one or more of those teams lose out, perennial dark horses New Zealand (48%) and Pakistan (43%) are the most likely to usurp them.

Our model's focus on team results likely underrates the talented but inconsistent West Indies team who are given only a 3% chance of progressing. The likes of Chris Gayle and Andre Russell haven’t featured as regularly over the past four years but Opta’s player rating model sees them as major additions, while the emergence of Shimron Hetmyer more recently is another potentially devastating prospect for opposition bowlers.

The winner

India’s all-round strength in both the batting and bowling departments, where they rank second and first respectively heading into the tournament, sees them as the favourites according to our model with a 25% of winning the tournament.

Given England’s chance of progression to the semi-finals, it is perhaps a surprise to see them as second-favourites at 24% with a combination of factors contributing to our model being less confident in their chances. England’s exceptional batting is comfortably ranked first in the tournament, and is thus expected to overwhelm less fancied opponents in the group stage.

However, the likelihood is that they will face far stronger bowling attacks in the knock-out stages at venues that are historically lower-scoring grounds according to our model. 

This suggests that England’s batting edge is potentially negated and their relatively weak bowling attack becomes more of an Achilles heel. Should India or England falter, the most likely winner according to our model would be a South African team looking to excise their World Cup demons, or their serial tormentors Australia. 

A new world record?

With talk of flat pitches, big bats and short boundaries, the 2019 edition is expected to be the highest scoring World Cup to date. With 48 matches and numerous excellent batting teams matching up against relatively weak bowling attacks, there has been talk of seeing a new world record surpassing the 481 scored by England against Australia at Trent Bridge last summer. We estimate there is a 10% probability of a total of 482 or more being posted in the 1st innings of a match during the tournament. There has even been talk of a score of 500, which we estimate as having a 5% chance of occurring during the tournament, with England vs West Indies at the Ageas Bowl being the most likely match to see such a feat should the hosts bat first.

Our data science team continues to build new models, across sports, to bring deeper understanding for our clients and their audiences. Get in touch to find out how this work can bring new insights to your coverage using the form below.


Have an enquiry? Get in touch.