Excitement is building ahead of the new football season, and, as ever, the wait for the real action to start is being filled with excited debate about how teams will do and what scores the opening weekend will produce. To fill the void, football economists Dr James Reade and Carl Singleton at the University of Reading have developed a computer model that is able to predict results and even scores of games before they happen. Here’s how it works.
Forecasting is a mug’s game, everyone knows this. Nonetheless, we like doing it, especially when it comes to football. How will Reading do this weekend? This season?
Given the sheer volume of information football generates in a timely fashion, it is readily collected and analysed. Statistical models are created and used to understand more about the game (e.g. when is a short corner better than a ball whipped in under the keeper’s nose?). Such models can also be used to forecast individual match results, scorelines, and even the final league table come next May.
We have created a model which estimates how many goals each team scores in a given match as a function of their own historical attacking and defending abilities, the historical abilities of their opponents, recent form, home advantage, the disruption of international breaks and European matches, and whether the match takes place on a weekend in August or a midweek evening in November.
We use this to predict what, according to the model, is the most likely score in the upcoming matches. This is an incredibly tricky task, one that usually flummoxes even the former football professionals Mark Lawrenson and Paul Merson on a weekly basis. In general, even the most likely scoreline only has a 10-15% chance of happening according to our model. So, on top of predicting the most likely scoreline, we give what the models suggests is the probability it happens.
Reading, now managed by former Derby and Swansea manager Paul Clement, kick-off the entire English football season on Friday night, welcoming rookie boss Frank Lampard’s Derby County to the Madejski. We find that the most likely final score is a narrow 1-0 win for Derby.
Does that mean we’re wrong if it finishes 2-1 to Reading instead? In a way, yes. But also no, because we can also say that there’s a 13% chance of a 1-0 win to Derby, which means there’s an 87% chance it’s not 1-0 to Derby – but that’s an imprecise forecast and not of much interest to anybody. What we are saying is that if the game on Friday night could be replayed 100 times under the exact same conditions – after each final whistle we rewound back to kick-off – then Derby would win 13 of those games 1-0.
Once we’ve predicted one set of games and the likelihood of their possible outcomes, we can carry on doing it, all the way to the end of the season. We do this using the model to simulate the entire season forward many times, updating the model estimates after each simulated game, until at the end we arrive at many iterations of the possible final league table.
The fraction of times Reading appears in the top two league positions at the end of May gives us a prediction on how likely they are to achieve automatic promotion to the Premier League. Similarly, and for all other teams in the division, we use the model and simulations to predict the likelihood of each team making the playoffs or suffering relegation.
After each week of matches over the coming season, we will update our model and our end-of-season predictions. As more information comes to light on the relative strengths of the teams both these predictions and weekly scoreline forecasts should become increasingly accurate.
See more predictions on this dedicated blog https://econscorecast.wordpress.com/
Opening Day Scoreline Predictions
Premier League[table “34” not found /]
Championship[table “36” not found /]
Final Table Predictions
Premier League[table “33” not found /]
Championship[table “32” not found /]