BRIDGEPORT, Pa. — To quote a movie from the ’80s – “There can be only one” – or so the argument goes.
A few weeks into the 2017 men’s varsity season, the answer to the question of the Top Five teams in the nation is unanimous with defending National Collegiate Athletic Association (NCAA) Champion the University of California, Stanford University, the University of California-Los Angeles, University of the Pacific and the University of Southern California rating among the Top Five in the FOSH’s Algorithm for Rate-Referenced Iterative Analysis of Game Activity (ARRIAGA) Top 20 Rankings, the USA Water Polo Freeman Top 25 Rankings and the Collegiate Water Polo Association (CWPA) National Collegiate Top 20 Poll.
However, the order of the Top Five is different among the three groups of rankings – with the FOSH’s ARRIAGA disagreeing with both the Freeman and CWPA/National Collegiate polls on the No. 1 team in the country.
Through games of September 24, the Freeman Top 25 and National Top 20 both concur on the Top Three teams being UCLA, USC and California. The two disagree on the final two positions in the Top Five as the National Top 20 has it at Stanford and then Pacific, while Freeman places Pacific in front of Stanford.
The FOSH/ARRIAGA poll disagrees with the official National Top 20 and Freeman Rankings at No. 1 as Stanford comes in at No. 1. All three agree with USC and California at No. 2 and 3, respectively, while the FOSH places National Top 20 and Freeman top ranked UCLA at No. 4. Pacific, similar to the National Top 20, also comes in at No. 5 in the ARRIAGA rankings.
So what causes the differences between the three sets of rankings?
Part of the issue is based on how the numbers are calculated – human judgment, experience and perception versus machine computation based on preset factors.
The National Top 20 Poll is compiled each week based on the opinions of coaches from around the country considering the compiled list of scores. To take any degree of across the board bias out of the analysis, the Top 20 is voted on by a panel of coaches from the Mid-Atlantic Water Polo Conference (MAWPC), Northeast Water Polo Conference (NEWPC), Golden Coast Conference, Western Water Polo Association (WWPA), Mountain Pacific Sports Federation (MPSF), Southern California Intercollegiate Athletic Conference (SCIAC) and the National Association of Intercollegiate Athletics (NAIA).
“Due to water polo being a national sport – and with issues such as budgets and missed class time, preventing a East Coast and West Coast teams from competing on a weekly basis outside season-opening or Spring Break trips – comparisons between two teams will always be difficult,” noted a National Top 20 pollster who are kept anonymous to prevent coercion and bias.
“Computers do not see the teams play – and coaches can sway calculations by running up the score or keeping contests close.”
Comparatively, both the ARRIAGA and Freeman polls are derived based upon algorithms and mathematical formulas without the human element (other than the individual who derived and wrote the code).
“Our ranking is less about whether you won or lost, and more about how different your result was from what it ‘should have been’, for every game – not just the ones against big-name opponents,” said FOSH co-founder Dave Danese.
“Results like UCLA’s three-goal win over LMU [Loyola Marymount University], when our calculations had them winning by 11, are a lot more damaging than results like Stanford’s close losses to Cal and UCLA – three teams we all know are really good and are separated by a very thin margin.”
However, how do computers adjust for human factors?
“This is all built on the assumption that every possession from every game, no matter who the opponent is, gives us valuable information for rating teams. Obviously this means we can’t account for things like mass substitutions in blowouts, but that’s the trade-off,” added Danese.
“Our system is 100% objective, with no human judgment other than the judgment that went into writing a couple of simple math equations to try to describe what happens in a water polo match at a fundamental level, without arbitrary coefficients or ‘fudge factors’. Systems with more human judgment are more flexible, which is a good thing, but they’re also subject to biases in judgment. In the end, each system has its strengths and weaknesses, but at least historically, our system has been pretty good at figuring out which teams are going to end up in the national final, even when it’s not that obvious from the eyeball test.”
The Freeman Rankings appear to operate in a similar fashion, per the explanation on CalPreps.com:
“It systematically sorts through all the results for the season (season-to-date results if we’re dealing with an in-progress season). It takes each result and compares it to what “should” have happened given the ratings of the teams. It knows that if A played C, A should have handled them fairly easily. If A lost that game, or even squeaked by with a narrow victory, its rating is hurt, while C’s is helped. The system keeps checking through all the results for every team.”
For now there are three polls – but in the end all three will end with the same team at No. 1 when the NCAA Champion is crowned the week after Thanksgiving.
