What's On My Mind
by David Keating

April 1, 2014

Swordmetrics – The Home Court Advantage in College Squash

If baseball can have the Society for American Baseball Research, or sabermetrics, squash could develop swordmetrics.  Perhaps someone will start a Squash Organization for Research and Data.

However you look at it, in the past there has been a paucity of data on squash compared to baseball or many other sports.  Sometimes data vanishes, which happened in Canada recently so that junior championship results from the last few years are currently unavailable online.

That’s not to say no one is working on more robust data and skill ratings.  Clearly some are and US Squash appears to be devoting substantial efforts to improve data collection and analysis.  Exhibit A is the new skill ratings published by US Squash.  Exhibit B hasn’t arrived yet, but US Squash is reportedly working on a new scorekeeping app that might capture a huge amount of data currently unavailable and allow for a revolution in data and player analysis.

The US Squash website says its “new rating algorithm is iterative, calculating the results for players many times relative to one another in order to bring all players to the appropriate rating level. It also has the ability to weight play, both by placing more importance on the most recently played matches and by weighting the importance of certain types of matches more significantly over others. The new system uses players with many match results (and therefore generally more accurate ratings) to refine the ratings of opponents who have fewer results in the US Squash system.”  It was developed by Elder Research.

While the old algorithm, which was based on ELO chess rankings, was transparent, it didn’t work very well, especially for rapidly improving players such as juniors.  Their skill ratings often lagged well behind their actual skills.  The new skill ratings don’t seem to have this problem.  While it’s not clear exactly how the black box works, the skill ratings seem more accurate than before.  The increase in accuracy, especially for high school player skill ratings where players often had few match results that were meaningful, is impressive.

Today I’ll take a look at the home court advantage in college squash.  Does it exist and how much is it worth?

The short answer appears to be yes, at least for the top eight nationally ranked college squash men’s teams over the last three years.

There are many reasons why the home court advantage theory intuitively makes sense.  Teams that play away matches are playing on unfamiliar courts and sometimes play in front of a large number of hostile fans.  They may have taken a long bus ride to get to the match and may be stiff from the ride, short on sleep or both.  Oftentimes teams will play three matches on a single road trip, so they could be tired compared to their opponents.  It was no surprise that Trinity’s home match winning streak, for example, lasted longer than its away streak.

I studied the top eight teams over the last three seasons of college squash.  While these teams only play one regular season match with an opponent, the national championship often presents a second match between similarly ranked teams on different courts.  This second match provides one way to determine whether a home court advantage exists.

In the last three years, two national champions won at home – Princeton and Harvard.  Trinity won its title on neutral courts. 

During the last three seasons, there have been 29 team pairs that have played twice in one season.  Of those matches, a stunning 14 of them showed a home court advantage compared to the match played either on neutral courts or the other team’s home courts.  Twelve matches produced the same outcome.  There were only three match pairs where a team had a better result on neutral or away courts than at home.

How much is the home court advantage worth?  Over the last three seasons, it ranged between 0.45 of an individual match win in 2012-2013 to 0.91 in 2011-2012.  The average over the three seasons is 0.66.  That’s a substantial advantage.

Studying data takes a long time and I freely admit there could be more accurate ways to assess whether there is a home court advantage and whether the effect is as strong for weaker teams.  For example, instead of analyzing the team match results, one could study the results between the same players in a single season or perhaps study the team game or point totals rather than the match wins and losses. 

But my rudimentary look at the squash data will have to suffice for now.  As Erasmus said, in the land of the blind, the one-eyed man is king. 


What's On My Mind is a column by rotating authors.
Contact DailySquashReport@gmail.com






Back To Main