Halloween may be past us, but that won’t make this weekend any less scary for Southampton fans. After conceding 9 at home last time out, they travel to Manchester City this weekend. Apply any logic, and City will surely score more than 9? The bookies certainly think Southampton have no chance, with the chance of an away implied by their odds just 3%. The Model perhaps gives less weight to the potential outlier of last weekend, and gives Southampton a healthy 9% chance of a win. City are expected to score 2.8 goals to Southampton’s 0.7, and the most likely scoreline outcome is 2-0 (12%).
The table below gives the rest of the Model’s forecasts for Round 11 of the Premier League, along with forecast probabilities estimated from current online bookmaker odds.
- Expected Goals: the forecast average number of goals the Model expects for Home and Away teams
- Outcome Probs: the model predicted % chance of either a Home or Away win, with 100 minus these two numbers being the % chance of a draw.
- Score Picks: the Most likely forecast scoreline outcome, as well as the most likely conditional on the most likely result outcome happening.
- Home wins / Draws / Away wins: the predicted % chance of various potential scoreline outcomes of the match.
- Mean odds: Estimates of online bookmakers’ average probability forecasts for the Home and Away teams to win, which could be compared with Outcome Probs.
Was this table not included in the upload or is it my browser being awkward?
This is a technical issue with the site. But you can see the tables posted on twitter: @csingletonecon or @SEconomists
Thank you. Keep up the awesome work guys!
Actual score City 2-1, that was close, but on the lower side. I would say it had something to do with the 9 changes Pep Guardiola made from the side that beat Aston Villa at the weekend. does the model account for the wholesale changes in a match?
Unfortunately the Model doesn’t account for actual selection decisions of managers… yet. But it does account for the influence of a team’s recent fixtures, for example if there was a midweek game or the team had to travel in Europe — statistically these things do seem to matter.