Hi everyone!
Welcome to the first edition of my EPL Predictions newsletter, where each week, I'll post detailed forecasts for each game.
As a general rule of thumb, each issue will contain an assessment of the previous week's forecasts, plus naive and biased predictions for upcoming games.
Since my model has been built upon a Bayesian framework (you can learn more about the general assumptions of the model here), I'm able to bias my predictions based on what my gut tells me will happen in each game.
The bias is introduced into my model by assigning the number of goals I'm expecting each team to score and the confidence I have in my assumptions. On the other hand, naive forecasts are solely based on historical data.
Let's get to it.
Naive Predictions
The naive predictions are computed using historical data only, and it's the forecast system that I use to simulate the end-of-season probabilities for the Premier League.
The following table showcases each game's forecasts under this model.
Biased Predictions
In this section, I present my biased forecasts, which, as I mentioned at the beginning of this post are based on both my hunch and historical data.
For each match, I'll present a detailed viz with probabilities as well as the prior parameters I'm assigning to each game – that is, the number of goals I expect each team to score (it doesn't have to be a round number) and the confidence I have in my assumption (from 0 to 100%).
The following table shows the summary of my biased forecasts.
Crystal Palace vs. Arsenal
Prior parameters:
- Crystal Palace expected goals and confidence level: 0.75; 35%.
- Arsenal expected goals and confidence level: 1.5; 40%.
Fulham vs. Liverpool
Prior parameters:
- Fulham expected goals and confidence level: 0.25; 60%.
- Liverpool expected goals and confidence level: 2.5; 50%.
AFC Bournemouth vs. Aston Villa
Prior parameters:
- AFC Bournemouth expected goals and confidence level: 0.75; 10%.
- Aston Villa expected goals and confidence level: 1.25; 10%.
Leeds United vs. Wolverhampton
Prior parameters:
- Leeds United expected goals and confidence level: 1.25; 10%.
- Wolverhampton expected goals and confidence level: 1.0; 10%.
Leicester City vs. Brentford
Prior parameters:
- Leicester City expected goals and confidence level: 1.0; 10%.
- Brentford expected goals and confidence level: 1.0; 10%.
Newcastle United vs. Nottingham Forest
Prior parameters:
- Newcastle United expected goals and confidence level: 2.15; 40%.
- Nottingham Forest expected goals and confidence level: 0.1; 10%.
Tottenham Hotspur vs. Southampton
Prior parameters:
- Tottenham Hotspur expected goals and confidence level: 2.75; 20%.
- Southampton expected goals and confidence level: 1.0; 20%.
Everton vs. Chelsea
Prior parameters:
- Everton expected goals and confidence level: 0.2; 45%.
- Chelsea expected goals and confidence level: 1.0; 35%.
Manchester United vs. Brighton & Hove Albion
Prior parameters:
- Manchester United expected goals and confidence level: 1.5; 10%.
- Brighton & Hove Albion expected goals and confidence level: 1.0; 15%.
West Ham United vs. Manchester City
Prior parameters:
- West Ham United expected goals and confidence level: 0.25; 25%.
- Manchester City expected goals and confidence level: 2.5; 20%.
Hope you enjoyed today's issue! Let's see how we fare in this season's first gameweek, I'll post the model's error in next week's issue.
If you like my work, please help me by subscribing to my website and sharing my posts.
Catch you later! 👋
