Sports Analytics Weekly by kubeia.io - 42/2026
- Week 42
- Week 41
- Week 40
- 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
Sports Betting Middling: Turn a 2% Edge Into 47%
pinnacle.ca - Dan Abrahams
Some people would describe me as a man of two minds. One is analytical, mathematical and serious, always looking for an edge and helping the 2% of bettors who know how to find real edges to optimize them. The second is fun and unexpected, writing parody songs and retweeting puns from Richard Munchkin on Twitter. I don’t see this dichotomy as a contradiction. While sharp sports betting is serious business, if it wasn’t fun then I’d probably have quit long ago.So, what happens when you add my two personalities to the 2% who are sharp enough to win at sports betting? As I’ll show you in this article, you get a 47% edge. Or potentially more. But how can that be? While the opportunity might be rare, the concept and the math are relatively simple.
📝 Sports Analytics
Finding The Players Traditional Football Metrics Fail to See
substack.com - Alex Marin Felices
Football player evaluation has traditionally relied on metrics such as goals, assists, chances created, and expected goals. These are useful for measuring attacking output, but they tend to favour forwards and creative players while underestimating those whose value comes from retaining possession, connecting phases, recovering the ball, or initiating transitions. Defensive midfielders and central defenders are particularly affected by this bias.
The paper proposes GoalNet, a Graph Neural Network framework designed to distribute changes in expected threat, or xT, across the players involved in creating them. Football events are represented as graphs, with players as nodes and interactions such as passes or tackles as edges. By combining spatial information, temporal context, and learned player representations, the model aims to identify important contributors who may not be directly involved in goals or assists.
🌩 Forecasting
Predictive Models Are Never Clean
argmin.net - Ben Recht
Twenty-five years ago, Leo Breiman wrote a prophetic warning to his statistical brethren:
“There are two cultures in the use of statistical modeling to reach conclusions from data. One assumes that the data are generated by a given stochastic data model. The other uses algorithmic models and treats the data mechanism as unknown. The statistical community has been committed to the almost exclusive use of data models. This commitment has led to irrelevant theory, questionable conclusions, and has kept statisticians from working on a large range of interesting current problems. Algorithmic modeling, both in theory and practice, has developed rapidly in fields outside statistics. It can be used both on large complex data sets and as a more accurate and informative alternative to data modeling on smaller data sets. If our goal as a field is to use data to solve problems, then we need to move away from exclusive dependence on data models and adopt a more diverse set of tools.”
Breiman saw that prediction was at the heart of most of his consulting projects. Rather than being asked to determine causation, he was asked to make forecasts.