Sports Analytics Weekly by kubeia.io - 39/2026
- Week 39
- Week 38
- Week 37
- Week 36
- Week 35
- Week 32
- Week 31
- Week 30
- Week 29
- Week 25
- Week 24
- Week 23
- Week 22
- Week 21
- Week 20
- Week 19
- Week 17
- Week 16
- Week 15
- Week 13
- Week 12
- Week 11
- Week 10
- Week 9
- Week 8
- Week 7
- Week 6
- Week 5
- Week 3
- Week 2
🎲 Betting
A bivariate Weibull count model for forecasting association football scores - ScienceDirect
sciencedirect.com - Georgi Boshnakov , Tarak Kharrat , Ian G. McHale
The paper presents a model for forecasting association football scores. The model uses a Weibull inter-arrival-times-based count process and a copula to produce a bivariate distribution of the numbers of goals scored by the home and away teams in a match. We test it against a variety of alternatives, including the simpler Poisson distribution-based model and an independent version of our model. The out-of-sample performance of our methodology is illustrated using, first, calibration curves, then a Kelly-type betting strategy that is applied to the pre-match win/draw/loss market and to the over–under 2.5 goals market. The new model provides an improved fit to the data relative to previous models, and results in positive returns to betting.
đź“ť Sports Analytics
How Strong Is a Team Outside Its Own League?
substack.com - Alex Marin Felices
Estimating context-specific team strength across domestic and European competitions with Bayesian uncertainty and B-CLAD.
Are Counterattacks Actually More Effective Than Positional Attacks?
substack.com - Alex Marin Felices
Across 5,700 matches in La Liga, the Premier League and Serie A, counterattacks consistently came out on top, but the more interesting story is what separates the teams that execute them well.
Introducing futi’s new advanced stats
substack.com - John Muller
When we first launched futi, we were kind of in a hurry due to some little international football tournament they decided to hold over the summer. So we did what young startups are supposed to do: focus on what makes our thing different.
For futi, that means better football models and data visuals. Just to name a few you’ve gotten to know by now:
Player ratings based on a possession value model that measures how much every touch of the ball changes both teams’ goal probabilities, scaled against a player’s data-derived role.
Data visual stories for every team and player with downloadable images like pass networks and possession value heatmaps.
Team styles and tendencies that quantify teams’ tactical approaches in different phases of play.
Deserved win percentages that simulate how often each team would be expected to win a match based on the chances they created.
Territory maps that show where teams were active, plus field tilt and better possession percentages that measure when they actually controlled the ball.
đź’° Quantitative Finance
Which Conformal Method Works for Which Alpha Signal?
substack.com - oracle
The same signal goes from Sharpe -1 to 1 depending on which conformal method you wrap it in. A practical taxonomy for matching the right method to the right signal.