Understanding Our Ranking System

The College Squash Association (CSA) rankings provide a structured approach to evaluating team performance, using different methodologies throughout the season. These rankings evolve as more data becomes available, moving from initial projections to result-based adjustments. Below is a breakdown of how the rankings are determined during each phase of the season:

🏆

Preseason Rankings

The preseason rankings serve as an initial reference point and are based on the players rostered on each team. The CSA verifies rosters, taking into account changes such as graduating seniors, transfers, and new recruits. Each player has a US Squash skill level rating. For unrated players, the CSA works with coaches to estimate ratings. Using these ratings, the CSA runs a program that simulates matches between all teams and ranks teams based on total wins in the simulation.

👥

Manual Rankings

From the start of the season through the second half of January, the CSA Rankings Committee reviews dual match results every week and manually adjusts the rankings based on the results of these matches. While the preseason rankings serve as a starting point, the committee regularly updates the rankings based on head-to-head match results. For instance, if Team B, ranked #10, beats Team A, ranked #9, then Team B will move ahead of Team A into the #9 position. This manual review continues until enough match results are available for more objective measures.

Computerized Rankings (ELO System)

From the second half of January through the end of the season, the CSA transitions to its ELO-based algorithm for team rankings. This system offers an objective evaluation of team performance, relying on the accumulation of match results throughout the season.

The ELO algorithm calculates rankings based on head-to-head matchups and results against common opponents. As the season progresses and more matches are played, the algorithm builds a comprehensive network of results, linking teams through their shared opponents. The proximity of teams in the rankings further influences the impact of individual results: a victory over a closely ranked team has a more significant effect on standings than one over a team further apart.

This system accounts for a full season's worth of wins, losses, and opponent strength, ensuring that rankings accurately reflect performance. Since it operates independently of preseason and manual rankings, the ELO methodology offers a data-driven, unbiased view of the current standings, based solely on team match outcomes. As more matches are played, the dataset improves, making the rankings even more accurate heading into the championship phase.