So there’s more interest than ever in the tenth Women’s Rugby World Cup, which kicks off in England on August 23. As with any big sports event, the pundits will be picking their winners based on recent win-loss records and overall performance trends – key factors that underpin official rankings.
Rugby Vision, a suite of statistical models I developed to predict outcomes for major competitions, uses similar information but with some important differences. Notably, it is less sensitive to the outcome of any single game.
The model uses three main components: a rating system for international teams; estimation of expected outcomes for each World Cup game using those ratings; and 10,000 simulations of the tournament to account for uncertainty around expected outcomes.
Rugby Vision has outperformed betting markets and other algorithms in recent men’s Rugby World Cups. I should add, though, it is not designed to support sports betting, but rather to illustrate how statistical modelling compares with real-world results.
Picking a winner
The probability of each team advancing to each stage of the tournament is shown in the table below. England, which has dominated women’s rugby in recent years and has home advantage, is overwhelming favourite with a 70.3% chance of winning the tournament.
New Zealand, which won the previous tournament, is second favourite with an 18.8% chance of defending their title. The next most likely winners are consistent performers Canada (6.7%) and France (3.9%). There is a 99.7% chance the winner will be one of these top four teams.
In each pool, one team (from the top four) is almost assured of making the quarterfinals, one team is “likely”, and one team has a fighting chance. For example, in Pool B, Canada has a 99.4% chance of being quarter-finalists, Scotland a 69.8% probability, and Wales a 27.5% chance.
Given the tournament schedule, Canada will likely play New Zealand in one semifinal, while England will face France in another. This is a key reason why France has a 75.5% chance of reaching the semifinals, but only a 12.1% chance of playing in the final.
Behind the algorithm
The simulations are based on an algorithm used to rank teams. The algorithm used here differs from that used for the official World Rugby Rankings.
Rugby Vision rankings and rating points for the participants are shown in the table below. England (with 136.8 rating points) is ranked number one, with a large lead over New Zealand (125.4), Canada (117.5) and France (114.6). Brazil (49.5), which will make its World Cup debut, rounds out the rankings.
The Rugby Vision rankings are comparable to the official World Rugby rankings, but a key difference is that Canada is ranked ahead of New Zealand in the official standings.
The rating points can be used to predict game results. For the first round of games at the tournament, the table below shows the expected winning margin of the favourite and the probability of each team winning.
For example, in the game between Scotland and Wales, Scotland is expected to win by nine points, but there is a 29.5% chance Wales will win.
While England enters the tournament as clear favourite, rugby has never been short of surprises. Past Rugby World Cups have seen giants fall and underdogs rise. The simulations give us a guide, but they are not a crystal ball.
What is certain is that the women’s game is in excellent shape – growing fast, drawing crowds and showcasing world-class talent. This tournament is bound to be a spectacle.
The author acknowledges the assistance of Soren Winchester, whose data collection contributed to the forecasts included in this article.
Niven Winchester does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.
This article is republished from The Conversation under a Creative Commons license.
