Having built a model to forecast the future progression of coronavirus in different geographical locations, and calibrated assumptions to match the UK experience, I regularly check the forecasts against reported figures. In the UK there are several different sources for coronavirus related data. The headline figures (those with greatest media focus) are published alongside the…
A mathematical case for masks
Small changes to coronavirus infection rates make a big difference to death rates. At a time when governments across the globe consider which policy tools can best reactivate economic activity – it’s worth evaluating how even small policy tweaks can help keep the death rate down. In my blog “How to relax lockdown without a…
How to relax lockdown without a second peak
Recent data demonstrates the UK is likely past the peak of new coronavirus infections. The lockdown has done its job. Now the UK, and other countries, are contemplating strategies to relax lockdowns and revive economic activity. But it’s a tricky balancing act – because weakening restrictions too much risks a second peak of infections and…
Building a pandemic model from scratch
Introduction The coronavirus pandemic has thrust forecasting models into the mainstream spotlight. They’re now recognised as essential tools to guide government policies that could prevent or cause many thousands of deaths. Whilst it’s reassuring that such grave political decisions are influenced by models built by experts, there’s a lack of transparency about assumptions and potential…
2019/2020 Premier League Projection
Here’s my 2019/2020 season Premier League projection I use a simple shot based model to assess teams, it’s explained here – but to summarise, it works out each team’s attacking and defensive strength by applying factors to goals, shots and shots on target, to produce a rating I call “adjusted goals”. Adjusted goals for and…
2018/19 Premier League Projection review
The table below shows a comparison of my model’s pre-season premier league points projection against the betting market (using Sporting Index’s pre-season mid-points price). My model performed well for the top 6, but overall the Sporting Index price performed better.
Leicester City v Manchester United match modelling
Experimental presentation of match likelihoods using adjusted goals model
2018/19 Premier League Projection
Here’s my 2018/19 Premier League projection I use a simple shot based model to assess teams, it’s explained here – but to summarise, it works out each team’s attacking and defensive strength by applying factors to goals, shots and shots on target, to produce a rating I call “adjusted goals”. Adjusted goals for and against…
World cup 2018 preview 3 – name the finalists likelihood
Using my simulation model, these are the likelihoods of 2 given teams meeting in the final. The structure of the knock-out rounds means that there’s a lower likelihood of Germany v Spain or France v Brazil finals as they’re more likely to meet before the final (if they win their groups). Note: the probabilities are…
2018 World Cup preview 2 – Projected outcomes
Here’s my full projection for the 2018 FIFA World Cup This is a result of 5000 simulations, using my International ratings. I’ve put some notes about the methodology behind these below. International football is difficult to analyse. Unlike domestic league football, where vast quantities of data exist to assess team and player performance in a…
World Cup 2018 preview 1 – how goal patterns repeat in international tournaments
International football is a strange beast, for the period in-between tournaments it’s an irritating punctuation –offering sporadic qualifiers or sterile friendly matches – unwelcome to players, managers and fans alike. Yet international tournament football has a capacity to engage and unify unlike any other form of the game (or probably any other sport). International football…
goal projection 