At Wiser Fans, we love hockey. Growing up in Montreal where hockey lies at the heart of the city, we would spend countless hours watching our beloved Habs compete against teams from across the league. We’d park ourselves in front of the television and watch, in breathtaking awe, the smooth, nifty hands of Alex Kovalev and relentless determination of Saku Koivu.
However, over the years, our experience as fans has evolved significantly. With the emergence of Fantasy Sports, we are no longer simply spectators watching from afar. More than ever, we are now involved and invested. We set lineups for our leagues and we put money on the line. We have skin in the game. We want more than just our home team to win. We want to win.
In the constant search to gain an edge, we set out to develop a platform that gives passionate hockey fans deeper insight into the sport we love. Inspired by a desire to win our Fantasy Leagues, and a late night viewing of Moneyball, we set out to answer one simple question: in a world overflowing with statistics how can we use numbers to our advantage?
In our journey to answer this question, we found both power and utility in hockey analytics – but something was missing. The data provided significant insight, but is often complex and difficult to digest. The data was presented in Shakespearen pentameter but we wanted Dr. Seuss pros.
Thus led to the birth of Wiser Fans – a platform committed to bringing you the fans, superior insights from analytics in a user friendly, digestible format. The one stop shop to give you the analytical edge without having to scour the internet and bust out your calculator.
We began by compiling various analytic stats for all teams over several of the past years. The next step was to determine what we would consider a "good" team. We concluded that, all else being equal, a team that made the playoffs in the year in question was considered to be a good team - we referred to them as being a playoff team. We then examined the different analytical categories and identified how the success in each of these was correlated with being a playoff team. Of course in hockey, the game is played on the ice, not behind some Wiser Fan analyst’s screen, and therefore, no one stat or combination of stat can truly predict the quality of the team. However, we focused on the stats that had the largest impact on whether a team was a playoff team (ex: did the team that was among the highest in Goals For % finish as playoff team).
We ran regression analyses, compared R-squared figures, looked at coefficients (you know the usual statistical tools) and identified the stats that showed the largest correlation with a team being considered a playoff team. We then created a model to assign weights to these stats based on the strength of the correlation between success in the stat category and being a playoff team. After countless reviews and tests, the team rankings product was born.
Current rankings based on results from our proprietary model. Follow these to find the top teams.
PDO is the sum of a team’s shooting percentage and save percentage. A PDO of above 100 usually suggests that a team is a benefiting from a certain amount of “luck”. A PDO of below 100 suggests a team’s “luck” should improve.
Daily moving average of the change in the team’s rank over the past seven days. Use this to get a better idea of the direction the team is trending in.
We would like to say that the process to develop the player rankings differed, but ultimately we applied the same logic...
Gather as many analytic stats as possible
Examine the correlation of each stat to the player finishing in the top percentile in scoring
Create a model and assign weights based on the strength of the correlation
So, you’re looking at these rankings and saying “ok, cool, but what does this all mean?” Well, the purpose of these rankings is to take a different approach to evaluating the top teams in the league. Forget about the top goal scorers and the teams with the most points in the standings. Analytics provides additional data sets that help evaluate performances that are otherwise not possible. The top team in the standings sometimes does not play like the top team in the NHL and may have benefited from a few lucky bounces or a lighter schedule. A top goal scorer may have been riding a hot streak or be benefiting from an unsustainable shooting percentage. Analytical data provides a larger sample size of stats and helps identify fundamental strengths in the league. These rankings are not meant to be predictive, but simply indicative of the current view of the teams and players from a hockey analytics perspective.