Home

Now that I’ve finally filled the holes in my dataset and have some decent predictions, I was able to start on version two - a glicko model. The ability to generate predictions with uncertainty and work with distributions should allow for more nuanced and useful predictions. But its a big change to make, so I’m implementing it in parts. To be Agi...

Model improvement was something I wanted to tackle much earlier, but some issues with data coverage became very apparent over the summer and became a higher priority. Now that I have most of the leagues I want, I can focus a bit more in the math. Before tinkering with the computation of a team’s elo (because intuitively, a player should contribu...

How Good Can We Get With A Simple Model? The first naïve model that was tried was very simple: in every game, the home team was predicted to win, with a likelihood of 1, by a score 5 points (the home advantage). This gave us a rough idea of how our elo scores predicted match results, but was almost entirely arbitrary. This gives a lot of room f...

There are a lot of moving parts in an elo calculation. For each match, we have as elo score for each team (elot) that is calculated using elo scores for each of the players (eloi) on the teamsheet. The exact computation of this team score will be changed later, but currently its a contribution system, with each player contributing to the team’s ...

Monthly Report - May 2022 Contents Summary Prediction Accuracy Best Predictions Worst Predictions Games to Watch, Snoozes to Miss, Upsets, and Covers Closest Projections Most One-Sided Projections Biggest Upsets Best Covers Summary We had 53 rugby matches in May of 2022. ...