Finally, An Objective Assessment of OUA Men’s Basketball Players


This past weekend, Ontario University Athletics (OUA) announced the winners of the Men’s Basketball Player of the Year (POY) award.  The LUSAG would like to congratulate Philip Scrubb of the Carleton Ravens (East Division) and Lien Phillip of the Windsor Lancers (West Division).

OUA administers a voting process among coaches to select its POY recipients.  How accurate are the opinions of these basketball coaches at selecting the best player in the league?


In 2010, a correspondent for Bleacher Report outlined a new formula for predicting the most valuable player in the NBA. This formula, called the MVP Rating, outperformed the conventional assessments of a basketball player’s on-court value by examining that player’s offensive statistical production (OSP) and his team’s number of wins.  A full breakdown of the formula can be found in the Bleacher Report article.

Just how accurate was this new formula?  The MVP Rating formula correctly predicted 21 of the previous 33 NBA MVPs (64%).  Furthermore, actual NBA MVPs coincided with the top two players produced by the MVP Rating formula 27 times (82%).

Being more than passive fans of OUA, we noticed a glaring lack of POY trackers for OUA men’s basketball.  Employing the MVP Rating method, we looked at the past 10 seasons (including 2012-13) of men’s basketball statistics from  The results were as follows:


In the past 10 seasons there have been 20 POY awards given to OUA men’s basketball players (1 to an OUA East player and 1 to an OUA West player per season).  The MVP Rating formula correctly predicted 14 of these selections (70%) and was correct within one player 15 times (75%).

It’s important to realize that an MVP predictor formula need not be right 100% of the time.  This is because the actual winners are chosen by a panel of coaches that combine both objective statistics and their subjective opinion to select the MVP.  The outcome of an ideal MVP formula will most likely differ from actual award winners when giving a definitive and objective selection.

Curious to see how the OUA’s top players measured up against Philip Scrubb and Lien Phillip?

Below is a chart and table outlining the OUA’s top twenty players in terms of MVP Rating.  The bolded names in the table represent the highest-ranking players in each division (both Josh Collins and Lien Phillip are bolded because their ratings were so close that selecting either player as the POY would not be considered a discrepancy).



* Wins = the number of projected wins in a 21 game season.  This season, teams in the East division played 20 games, while the West division teams played 21.  Without the adjustment, the MVP Ratings, calculated as the player’s portion of the team’s OSP multiplied by the number of wins, would be skewed in favor of West teams.


The biggest discrepancy between the player with the highest MVP Rating and the recipient of the POY award occurred in 2008-09.


Josh’s MVP Rating was 27% higher than Stuart’s, yet Stuart was awarded the POY award for the East division.  It’s difficult to say why this happened.  It could be that in this particular year, the coaches subconsciously overvalued wins in their selection of the league’s POY (Carleton had 21 wins, while Ottawa had 19).

Josh Gibson-Bascombe, if you’re feeling as bitter as we are about you being robbed of the POY award in 2009, we would be happy to support you in an appeal to the OUA.  To many, using advanced statistics to argue the recipient of a university basketball POY from four years ago may sound ridiculous, but we at the LUSAG believe in justice.

The NBA. Where Zipfian Distribution Happens

George Zipf, the late American linguist, recently celebrated his would-be 111th birthday.  In honor of this occasion, let us all take the time to remember Zipf’s law, one of the most freakishly accurate regularities in economics.

Zipf’s law explains frequencies for a shockingly high number of data sets.  It states that the frequency of any observation is inversely proportional to its rank in a frequency table.  To put it another way, in a data set where zipfian distribution applies, there will be a few observations that occur very frequently, a medium number of observations occurring with medium frequency and many observations that occur much less frequently.  Here’s a simplified version of what Zipf’s law looks like in a frequency table.


This phenomenon can be best explained through one of its notable applications.  Zipf noticed that in the English language, there seemed to be a pattern for how often words appear. For example, the word “the” is the most common word in the English language and “of” is the second most common.  “Of” appears half as many times as “the.”  It follows that the third most common word appears 1/3 as often as “the” and the 99th most common word would therefore be used 1/99 as often as “the”.  This pattern holds nearly perfect for the entire English language as well as all other spoken languages.

Zipf’s law also holds true for a variety of different data including the ranking of U.S. city populations, revenue distribution of different companies, and even library book checkout patterns.  It is interesting that Zipf’s law is not a law at all, but rather a statistical model for ranking many (but not all) data sets.  Read more about Zipf’s law in this New York Times article written in 2010.

Curious to see how Zipf’s law applies to Big 4 North American sports?  We were too.

At first, we were disappointed to find that it does not apply to attendance, market value, ticket price, or historical wins for teams in any of the Big 4 leagues.  It wasn’t until we realized that Zipf’s law is a model that explains human preference on a macro scale that we found a match.  Everyone speaking the English language chooses to say words more often than others.  Population distribution among cities reflects the decisions of an entire nation’s inhabitants for where to live.

Professional sport attendance does not meet Zipf’s criteria.  On any given night people on the west coast can’t choose to attend a sport event on the east coast, and vice versa.  Their choices are limited geographically.

So, what in sports resembles the distribution of a large population’s preferences?

Social Media.  Let’s take a look.

The following graphs and tables compare the actual number of ‘Likes’ and ‘Follows’ for NBA teams (in blue) with the ‘Likes’ and ‘Follows’ predicted by Zipf’s law (in red).








For you statistics nerds, the variation of values predicted by Zipf’s law explained 95% and 98% of the variation in actual ‘Likes’ and ‘Follows’, respectively.  If you’re interested, take a look at the full regression summary statistics for Facebook and Twitter.

Anyone can choose to ‘Like’ a team on Facebook or ‘Follow’ it on Twitter regardless of where they live.  Social media followings for sports teams are a perfect example of human preferences at work, even though it’s still difficult to explain the zipfian distribution.  I think the only mystery greater than Zipf’s law is why people continue to ‘Like’ and ‘Follow’ the Los Angeles Lakers after they started the season a disappointing 25-29.

‘Fan’ Battle of Ontario

Since Ottawa’s entry into the NHL in 1992, there has been a rivalry between the Ottawa Senators and the Toronto Maple Leafs often referred to as the Battle of Ontario.  Playing in the same division, these teams face each other six times per season and in these matchups, it is not unusual to see season-ending injuries, team-wide fights, and fanatic fans for each team.  Ottawa currently leads the rivalry (they are 49-34-3-5 9 (.598) in all games against Toronto).  The two teams have met in the playoffs four times in the five seasons leading up to the 2004-05 lockout.

The only thing stifling this great rivalry is the fact that the Leafs have not made the playoffs since 2004-05.  While it’s clear that over the last seven seasons, the Senators have been winning the Battle of Ontario on the ice, the teams also compete for the attention and interest of Ontario hockey fans off the ice.

Inspired by a blog written by the Harvard Sports Analytics Collective, we have conducted a study to produce a geographical representation of this rivalry.  We wanted to see where Ontarians stand on the Battle of Ontario and find out which team comes out on top in terms of fan interest.  It will also be interesting to see if the four NHL teams on the province’s outskirts (the Red Wings, Sabres, Canadiens, and Jets) have been able to establish a foothold in the market for Ontario hockey fan interest.

For the purpose of this study, fan interest in an NHL team was measured as the number of ‘likes’ the team has on Facebook.  The ‘Create an Ad’ Facebook feature was used to collect the data, as it allows an ad buyer to assess the market size, in terms of Facebook population, of a particular city and the market size for a city’s population that ‘Like’ a particular page on Facebook.  For example, using this method, we found that there are 5,600 people that live in Sudbury, ON that ‘like’ the Toronto Maple Leafs on Facebook.

This methodology was used to calculate the number of ‘likes’ for the Toronto Maple Leafs, Ottawa Senators, Montreal Canadiens, Detroit Red Wings, Buffalo Sabres, and Winnipeg Jets in all 583 Ontario census subdivisions, as defined by Statistics Canada.  From there, the proportions of fan interest in the teams across Ontario were calculated after aggregating the subdivisions into 49 census divisions and plotted on a map of Ontario.  Each census division was awarded to the team with the highest proportion of Facebook ‘likes’ in that territory.

Map #1 shows fan territories when all six teams are compared. As you can see, the Toronto Maple Leafs dominate 42 of the 49 territories, proving their victory over the Senators and the other teams in the ‘Fan’ Battle of Ontario.

Map #1: Ontario Fan Territories- All 6 Teams


Map #2 shows a 1-on-1 comparison between the Maple Leafs and the Senators using a color grading scale.  The 49 Ontario divisions were colored according to the proportion of ‘likes’ for both teams. While several regions surrounding Ottawa were colored in favor of the Senators, the high levels of deep blue further indicate the Maple Leafs as the clear winner of the ‘Fan’ Battle of Ontario.  In fact, 30 of 49 census divisions were dominated by 80%+ Maple Leaf fans (80%+ of all hockey fans that liked one or more of the Leafs, Senators, Canadiens, Red Wings, Sabres, or Jets Facebook pages).

Map #2: Ontario Fan Territories- Toronto vs. Ottawa


The Montreal Canadiens are the only team that competes with the Maple Leafs for fan interest across the province.  While it’s obvious in Map #3 that the majority of the census divisions are won by the Maple Leafs, you can’t ignore Montreal’s job of diluting the blue divisions.  Only 2 of the 49 census divisions were dominated by 80%+ Maple Leaf fans.

Map #3: Ontario Fan Territories- Toronto vs. Montreal


Now let’s look at the ‘Fan’ Battle of Ontario in a different context- by per cent of the Ontario Facebook population that ‘Like’ a team’s Facebook page.


It is astonishing to discover that more than 1 in every 20 people on Facebook in Ontario like the Toronto Maple Leafs’ page.  Frontenac, Ontario marks the most highly concentrated area of Maple Leafs fans with 13.2%, or nearly 1 in every 7 people on Facebook ‘liking’ the Maple Leafs Facebook page.

There you have it.  It turns out the Battle of Ontario off the ice wasn’t much of a battle at all.  For many, this study only confirms what they already know- that the Toronto Maple Leafs are the most popular team in Ontario.  If you’re going to take anything new from this article, let it be this: if you’re a diehard Senators fan travelling outside Ottawa, leave your Daniel Alfredsson jersey at home… you will be met with great prejudice. Goes Live