Making the rankings system a little less un-scientific

Making the rankings system a little less un-scientific

Brian Mason (Athletic Communications Assistant)

For some reason, it’s human nature to attempt to quantify things that don't necessarily lend themselves to being quantified.

Whether it’s barbeque, quarterbacks or fighter pilots, we’re always trying to figure out who the best is.
 
In college sports, it’s the weekly debate about who makes it into the all-important Top 25.

Rank something, and it’s sure to be debated. (Case in point: There’s no way “Gimme Shelter” is only the 38th greatest song of all time!) It’s the principle that has cable networks filling their programming schedules with an endless stream of shows devoted to ranking reality show moments or counting down the 40 most awesome truck stops in America.

Swimming is no different from myriad other college sports in that someone has undertaken the task of attempting to place the best teams in their proper order, and someone else has taken the time to politely disagree.

It’s not an easy task. How do you begin to quantify the quality of collegiate swimming teams? Head-to-head matchups mean little since there are so few among the top teams, and most teams' worth isn't apparent until the end of the season when their athletes taper, shave and put it all on the line at conference and NCAA championship meets.

Ranking the top swimming teams in the country is certainly an unscientific process, but a group of former swimmers from Princeton has recently injected a little more science into it.

The group, which started www.swim-rankings.com, has set out to "apply objective analysis to college swimming, a sport that is inherently objective, by empirically rankings its teams using a custom-made computer script."

Essentially, they've employed a computer program that takes in data (from the USA Swimming top times list) on a number of teams and simulates dual meets between every team, using lineups created for each team based on the data. The program then determines the outcome of each meet and goes on to rank the teams based on overall point differential.

In theory, then, the teams that win the most, and by the most points, will be ranked highest.

So, what does it mean for the Badgers?

While the Wisconsin women’s team is ranked 17th in the latest CSCAA poll, released Jan. 16, the Badgers find themselves at No. 10 in the new computer rankings.

The UW men enjoy an even bigger bump, as they are only receiving votes in the CSCAA poll but are ranked No.15 by the computers.

Is the new system perfect? Of course not. Is it a more accurate measure than the current poll? Logic would say that it is.

Consider the current system for ranking the top teams.

Like many rankings, those that lay out the 25 "best" teams in college swimming are often disputed. The College Swimming Coaches Association of America (CSCAA) Top 25 poll has been the standard for determining teams' comparative strength for years, but its rankings rely on the same subjective approach as is used in other college sports like football or basketball.

While asking coaches to vote on the nation's top teams is a far-from-perfect system even for sports that receive plenty of national media attention, it is inherently more flawed when applied to swimming.

Unlike football or basketball, in which voting coaches can keep up with the goings-on of major programs on a daily basis thanks to live TV broadcasts of games, the coaches comprising the CSCAA's voting panel are limited to seeing only the teams their own programs go up against in meets during the season.

Essentially, the USA Swimming times database provides a foundation for the voting panel, which then extrapolates teams' worth without the help of any additional context for the performances they're looking at on paper.

Also, unlike the major coaches' polls in football (61 voters) and basketball (31 voters), the CSCAA poll is comprised of the votes of just nine coaches. That's representation for just 6 percent of the 141 NCAA institutions offering Division I men's swimming and a scant 4.5 percent of the 198 women's programs.

Compare that to the 51 percent of FBS football teams represented in the USA Today Coaches Poll.

With such a small sample size, it takes only miniscule changes in the coaches’ ballots to make a large impact on the poll. Then there’s the fact that the voting panel is not released publicly and that the poll is extremely hard for the average fan to find, as the CSCAA Web site has apparently been taken offline.

Still, it’s not the CSCAA’s fault that no one else has come up with a better way of developing rankings. Until now, that is.

The guys behind www.swim-rankings.com have done just that and are even engaging feedback in the hope of further improving their system.

It may not provide all the answers, but the new system has at least begun the conversation.


Appears in Women's Swimming Blog

Tagged with Swimming, Men's Swimming Blog, Women's Swimming Blog

Bookmark and Share
ON WISCONSIN
Get to Goodman
  • Loading Tweets...
    1 second ago