Data-Driven Insights for Soccer Team Management

MLS Salary & Performance Analysis

This study examines the relationship between player salary allocation and team performance in Major League Soccer from 2018 to 2024. Using data from fbref.com and the MLS Players Association, the research applies three performance metrics — Pythagorean Expectation (PE), Expected Pythagorean Expectation (xPE), and Expected Points (xPts) — derived from actual and expected goals to quantify team success.

A Lasso regression model is fitted for each MLS team, using salary summary statistics by player position (mean, median, variance, interquartile range, and percentiles) as predictors. The models demonstrate an exceptionally high in-sample fit, with case studies of LAFC, LA Galaxy, and San Jose Earthquakes showing near-zero prediction error across all seven seasons.

Building on the regression framework, a sensitivity analysis called Player Absence Impact estimates each individual player's marginal contribution to team performance by simulating their removal from the roster. This method is applied to 2024 rosters to assess how roster changes between the 2024 and 2025 seasons affected each team's projected performance outlook. The analysis contributes to the broader sports economics literature on wage dispersion and financial efficiency in professional soccer.

Building on the JQAS study's Lasso regression framework, this section extends the Player Absence Impact analysis to current MLS rosters using salary data sourced from Spotrac (2021 - 2025 MLS seasons). Individual team reports are generated for each MLS club, providing a deeper look at how salary allocation translates to on-field value at the roster level.

Each team report applies the Player Absence Impact metric — the percent change in a team's Pythagorean Expectation (PE) when a single player is removed from the roster — to evaluate every player across multiple contract seasons. A negative percent change identifies a player as a valuable asset (the team gets worse without them), while a positive percent change flags a player as replaceable or an inefficient contract.

The reports are organized around five core analytical dimensions:

• Designated Player ROI — whether high-cost DP slots are delivering elite or marginal impact relative to their salary
• TAM & U22 Allocation Efficiency — identifying "steals" and misallocated roster budget within these spending tiers
• Homegrown Player Evaluation — assessing which academy products are ready for meaningful roster contributions
• International Slot Optimization — flagging international slots occupied by players with positive (harmful) percent change
• Positional Depth Audit — highlighting positions where no player registers a strong negative impact, revealing critical upgrade needs

A Salary vs. Value Matrix classifies every player-season into one of four quadrants — Core Asset, Hidden Gem, Overpaid, or Replaceable — based on the combination of salary level and PE percent change. Year-over-year trends are also tracked to identify whether a club's roster construction strategy is improving or regressing over time.

The Roster Efficiency Reports were created with the assistance of generative AI tools, applying the Player Absence Impact analytical framework to Spotrac salary data.

...