Season Forecast
The season is over — this is the final forecast we generated, frozen before the last round (data as of 2026-05-24). We simulated 50,000 possible seasons using Elo ratings that update after every match, Poisson goal models, and Club Elo tilt.
Data correct as of
Season complete — final pre-matchday forecast
Likely Possible Unlikely
How to read this chart
- Each row is a Premier League club, sorted by predicted finishing position.
- Each column is a league position (1st to 20th). The number in each cell is the probability (%) that team finishes in that position.
- Darker cells mean higher probability. Hover or tap for exact figures.
- Zone backgrounds mark the title (1st), Champions League and Europa League (2nd-5th), and relegation (18th-20th).
Methodology
- We ran 50,000 Monte Carlo simulations of the then-remaining 1 gameweeks (10 fixtures).
- Each simulation uses "hot" Elo ratings — after every simulated match, both teams' Elo ratings update based on the result and margin. This means a team on a winning streak gets stronger within that simulation, just like real momentum.
- Match scores are sampled from a Poisson distribution. Expected goals for each team are derived from the Elo difference plus Club Elo tilt (a measure of how many goals a team's matches tend to produce).
- Elo updates are weighted by goal margin: a 3-0 win shifts ratings ~1.7× more than a 1-0 win (proportional to √margin).
- Teams are ranked using Premier League tiebreakers: points, then goal difference, then goals scored.
- Elo ratings from Club Elo. Standings from SportMonks. Updated weekly.
See also
- GW Table Possibilities → Every possible position after this weekend's fixtures.
- Run-In Tracker → Predict results for every remaining fixture and see the projected table.
- Why Club Elo? → We tested three models against 760 real matches. Here's what we found.