I love advanced stats and what they can reveal about Hockey. Possession stats specifically catch my attention; they really give a new perspective on what it is to be an effective hockey player or team at even strength. They can tell you a lot and are proven to be a very strong predictor of future success and goal differentials.
Something that I thought interesting when it comes to these statistics was the playoffs and how well they can predict post-season success. Playoffs are such a small sample size, which inherently means they are somewhat unpredictable. That being said, I would assume that due to teams with better possession stats being more likely to win in general, that this would reveal itself strongly in the long term results of several years of playoff outcomes — let’s see if that holds up.
I took playoff data from 2011 through 2013 and matched up the games played and wins to their regular season fenwick close percent.
The first thing that stands out here is that the better possession teams have played more games. This is for two main reasons; one, the better the possession stats for a team are, the more likely they will make the playoffs and second, they are more likely to go deeper in the playoffs. Let’s take a look at this information another way, expressing winning percentages.
|Games Played 2011-2013 playoffs||Teams In Playoffs||Winning %|
|Below 50% FenClose||100||9||48%|
|Above 53% FenClose||215||16||54.9%|
|Above 53% FenClose against below 53% Fenclose||131||13 (3 played other >53% FenClose teams in first round)||57.25%|
The teams that considered very impressive possession wise — that is the ones above 53% fenwick close — have a winning percent just shy of 55%. That’s fairly impressive. Meanwhile, the teams in the “good” possession section between 50 and 52.9% actually have not fared any better than teams below the 50% possession mark; they have in fact done worse.
The fact that the teams below 50% FenClose have fared better than those in the 50–52.9% range over the past 3 playoffs surprises me. It may just be a matter of insufficient sample sizes, but still the result are not what I expected. The fact of the matter remains, however, that the great possession teams have done considerably better than others.
When the results of the 53% plus fenwick teams are separated from series’ where they played other 53% fenwick teams, the results are even more conclusive. Teams with a fenwick close above 53% have a winning percentage of 57% when facing teams below 53% FenClose.
The importance of possession is also evident in the Stanley Cup winners.
Over the past three seasons, the Stanley Cup winner has had an average regular season fenwick close of 53.46%. The runner up has averaged 53.04%. The lowest fenwick percent for a Stanley Cup winner over the last three years was Boston’s 50.6% in 2011. It’s evident, thus, that the best teams over the last three years have typically been very strong possession teams; not a particularly surprising assertion.
Even in small sample sizes, the predictive value of possession statistics holds up but to a much smaller degree than the larger sample sizes of the regular season, but still enough to conclude it has obvious value — a conclusion that begs for a larger study. However, strong possession stats did not cause as high winning percentages over the past three years as I would have expected. There are just so many ingredients that are necessary to make a successful hockey team. Isolating possession as the only factor of winning cannot entirely predict success. Is possession important? Of course it is, but it is only one piece of the puzzle, albeit a very large piece..
Hockey has so many factors that it will always be impossible to predict perfectly who is going to win. The parity in this sport allows for constant excitement and insures that even inferior teams have legitimate chances in any individual game or series. One can always predict who is most likely to win, but never who is going to. At the end of the day, that’s why this sport is so captivating; it’s why I love this league, because anything can happen.
Follow me on Twitter, @LukaRyder