The weekend is over, and another weekend full of European soccer in the books.
📝 Each Tuesday going forward, we would like to take stock of SFM’s predictions that we shared in our preview the previous Friday – and add potentially some interesting cases of the weekend that we missed to anticipate.
👉 Taking stock is not only crucial to understand the model’s performance in real time in a true out-of-sample setting, but is also hopefully just quite fun and interesting to read.
👉 Special thanks to one of our most loyal followers for really pushing us to do so!
This first post of the series will rather be of qualitative nature only, but in future iterations, we strive to provide a more principled and quantitative evaluation of SFM’s performance.
So, let’s get to it...
📝 Recap of Friday’s Preview
- Erling Haaland – Manchester City
- Probability of Scoring at least one goal:
- ~ 63%
- Probability of Scoring two goals:
- ~ 20% (❗️ Friday’s special feature)
- Starting XI:
- yes
- Goals Scored:
- 2
- Pierre-Emerick Aubameyang – OM:
👉 Well, the Ligue 1 was on pause over the weekend. So, the prediction for Aubameyang will be put to the test in the game against FC Nantes on January 4th.
- Harry Kane – FC Bayern München:
- Probability of Scoring at least one goal:
- ~ 48%
- Starting XI:
- yes
- Goals Scored:
- 1
- Robert Lewandowski – FC Barcelona:
- Probability of Scoring at least one goal:
- ~ 41%
- Starting XI:
- no
- Substituted in:
- Minute 62’
- Goals Scored:
- 0
- Kylian Mbappe (**) – Real Madrid:
- Probability of Scoring at least one goal:
- ~ 37%
- Starting XI:
- yes
- Goals Scored:
- 1
SFM Coverage
That’s it for this weekend. If you want to find out more, the SFM covers the #premierleague, the #bundesliga, #laliga, #seriea, and #ligue1. All available at:https://soccerfactormodel.com
Further SFM Resources
Link to the Paper: Unveiling True Talent.
Link to the Dashboard: Skill- & Performance-Above-Replacement.
#socceranalytics #pymc #sportsanalytics #bayesian #statistics #soccer #football #prediction #forecasting #sfm #factormodels #alpha