Skip to content

Latest commit

 

History

History
28 lines (20 loc) · 3.35 KB

README.md

File metadata and controls

28 lines (20 loc) · 3.35 KB

Club_Soccer_Season_Projections

Club_Soccer_Season_Projections utilizes Expected Goals (xG) data and squad quality ratings within an elo-based Monte Carlo model to predict the final standings in 7 major European soccer leagues and the UEFA Champions League. These projections are generated through 10,000 simulations for each competition.

Overview

Each league's summary includes estimated final positions, points, goal differences, title chances, chances of qualifying for the UEFA Champions League, and relegation possibilities.

These simulations are calculated using a combination of actual league results and simulated matches based on Elo ratings. Elo Ratings are weighted based on three different metrics, as shown in the following table:

Weight Short Description Long Description
50% Club Elo Club Elo ratings are derived from real match results using the formula provided by Club Elo.
25% Expected Goals Expected Goals (xG) ratings utilize a Poisson Distribution model to predict the points exchanged in a match between two teams. These statistics focus on the quality of scoring chances created during a match and do not consider the actual match results. The Expected Goals data is sourced from fbref.com based on Opta's estimation of expected goals.
25% Squad Quality Squad Quality Metrics are determined by the quality of players within a club. Ratings on a scale of 0 to 100, provided by footballtransfers.com, are converted into an Elo rating. This rating is combined with other metrics to assess a team's overall strength on paper, considering any boosts from incoming transfers.

Expect regular updates, typically on a weekly basis, though the frequency may vary depending on the match calendar.

Simulations Include Results For: