Can Analytics Predict the Playoffs? — Part 1
One of the biggest reasons that puck possession numbers have become so prevalent in today’s NHL is because of their ability to predict playoff success.
Over the past decade or so, teams that finish the season with high 5-on-5 Corsi For percentages find themselves going deep into the postseason, especially the teams that finish near the top of the league.
Here’s a quick and simple chart, showing the Stanley Cup winners and runners-up of the salary cap era, as well as their 5-on-5 score adjusted Corsi For percentage (rank in parentheses). The 2005-06 season is kept off due to the high number of power plays we saw in the league that year (less time at 5-on-5 means less importance for puck possession metrics).
As we can see, the winning team has been an elite possession team in almost every season; seven of the nine winners were top five in Corsi For percentage, while the two teams that fell outside of the top five still rated high by possession metrics.
The 2010-11 Boston Bruins finished 10th in Corsi For percentage, which is good, but not great. They had a massive advantage over the rest of the league when it came to their goaltender, however, as Tim Thomas posted a 0.938 save percentage during the regular season, and a 0.940 save percentage during the playoffs, transforming a good Bruins team into the Stanley Cup Champions.
The 2008-09 Pittsburgh Penguins may have finished 19th in the league that year, but they had a coaching change late in the season, firing Michel Therrien and hiring Dan Bylsma. Under Bylsma, the team was 6th in the league with a 53.6 percent Corsi rating.
Even among the runners-up, six of the nine teams finished the year in the top 10 in Corsi For percentage.
Overall, among the 18 teams that have competed for the Stanley Cup in the past nine years, 14 of them (15, if you count the 2008-09 Penguins) finished the regular season in the top ten for score adjusted Corsi For percentage.
Looking at puck possession metrics instead of the league standings is generally going to give you a good idea of which teams are ripe for playoff success. The key reason for this is because puck luck and random variance play a significant role in determining the outcome of games during the regular season, while shot attempt metrics remain relatively stable.