We statheads have a lot of issues with the playoffs. While the regular season offers us large sample sizes and a variety of opponents, allowing us to judge most players and teams by a common standard, the playoffs are nasty, brutish and short: a few games, one key injury, and your season is done. Worse yet is the chorus of experts decrying how a team that loses a series in 7 games just “couldn’t get it done”, while the other team “showed their experience”, “took their play to another level” and “did what they needed to win”, despite the fact that one different lucky bounce would have swung the series in the other direction, and the comments would have been reversed. At best these comments are hyperbole; at worst, bull. Billy Beane one said “My s*** doesn’t work in the playoffs”, which most statisticians understand to mean that you can never “guarantee” anything in such a short series. Ergo, the playoffs are meaningless.
What if they’re not? What if there truly is some ingredient, some special sauce, that causes teams to be successful in the playoffs? After all, we know that the Ducks have it and the Sharks don’t, right? The process of finding this is, in principle, trivial: figure out, from each team’s regular season record, how well they should do in the playoffs. Then compare the expected performance level to the actual performance level, and find out which teams have met or exceeded expectations, and which have missed them.
I have performed this analysis for the 7 seasons from 2000-01 to 2007-08. There are two reasons for this: one, over a span of 8 years, a lot of the key players on teams will remain in place, making team’s playoff success from year-to-year correlated. Second, the NHL has been at 30 teams since 2000, so a lot of the factors that influence regular season to playoff success (such as the fact that 16 teams out of 30 make the playoffs, and thus the typical number of points necessary to make the cutoff) have remained constant.
As a proxy for regular season success, I used points rather than goal differential, which I usually favor. The reason is that points are what determine whether or not you make the playoffs, as well as your seeding which will influence the strength of the opponents you face. In practice, points and goal differential are very highly correlated, so the results should be the same. Since the NHL introduced the shootout to break ties in 2005-06, the average number of points per team has increased from about 86.5 to 91.5, so I used (points – average) to put everyone on a level playing field.
For playoff success, I used a simple formula: 7 points for making the playoffs, from 10 to 7 points for winning a series in 4 to 7 games, from 0 to 3 points for losing a series in 4 to 7 games, and an extra 3 points for winning the cup. I’ve then normalized this so the total number of “playoff points” given to all teams is 30, an average of 1 per team.
Here is the distribution of all 210 teams over the last 7 seasons:
There are a few obvious features: all the teams below a certain number of points have missed the playoffs, while on the right side of the graph there is a vague but present leaning to more playoff points with more regular season points, especially noticeable when looking at first-round losers who are disproportionately among the weaker teams that made the playoffs. If we attempt to pass a least-squares line with a cutoff by these points, the prediction function we get is the following:
So who are the winners and who are the losers? Some won’t be surprising, while others may be shocking:
PVA: Points vs. Average
EPERF: Expected Performance
Regular season Playoffs Playoffs Over/Under Achieve
PVA Rank PERF Rank EPERF Rank
Ducks -4.5 14 13.2 5 6.7 13 6.5
Hurricanes -12.5 18 9.2 9 5.6 16 3.6
Lightning -37.5 21 8.6 10 5.3 17 3.3
Penguins -67.5 26 7.7 13 5.1 20 2.6
Flames -6.5 16 6.9 14 5.2 19 1.7
Avalanche 86.5 5 14.2 3 12.6 5 1.6
Devils 100.5 3 15.2 2 13.8 3 1.4
Rangers -30.5 20 4.9 19 3.7 24 1.2
Canadiens -6.5 15 6.1 17 5.3 18 0.8
Sharks 64.5 6 11.9 6 11.1 6 0.8
Blue Jackets -134.5 30 0.0 30 0.0 30 0.0
Blues -12.5 17 6.8 15 6.8 11 0.0
Islanders -49.5 24 3.8 23 3.8 23 0.0
Oilers -0.5 13 6.0 18 6.0 15 0.0
Flyers 36.5 8 10.5 7 10.6 7 -0.1
Sabres 29.5 10 7.9 11 8.1 9 -0.2
Maple Leafs 34.5 9 7.8 12 8.1 10 -0.3
Kings -46.5 23 3.1 24 3.4 25 -0.3
Panthers -93.5 29 0.0 29 0.5 29 -0.5
Wild -15.5 19 4.3 20 5.0 21 -0.7
Senators 109.5 2 13.9 4 14.6 2 -0.7
Thrashers -91.5 28 0.8 28 1.8 28 -1.0
Blackhawks -91.5 27 0.9 27 2.0 27 -1.1
Capitals -54.5 25 3.1 25 4.3 22 -1.2
Coyotes -43.5 22 0.9 26 2.5 26 -1.6
Predators 0.5 12 3.8 22 6.2 14 -2.4
Bruins 3.5 11 4.2 21 6.8 12 -2.6
Canucks 53.5 7 6.5 16 9.7 8 -3.2
Red Wings 177.5 1 17.4 1 20.6 1 -3.2
Stars 99.5 4 9.3 8 13.7 4 -4.4
The Anaheim Ducks, more or less as expected, are by far the biggest playoff overachievers of the last decade, even without counting their current successful run. If we look at the top 5, they are all Cup winners or Cup finalists of the last few years, many of them surprise finalists
(Carolina in 2002, Anaheim in 2003, Calgary in 2004). In effect, given the parity of the league, ANY Cup finalist is to a certain extent a surprise finalist; as an example, the Penguins in 2008, while second in the East, only had 8 more points than the 8th-place Boston Bruins! Statistically, the Penguins were expected to play in 1.63 rounds; they were hardly “expected” to make the finals, and were probably of similar strength to this year’s New Jersey Devils, who are already gone.
So what has been the key to the Ducks’ success? Given that this analysis didn’t incorporate the 2009 playoffs, the Ducks’ postseason record consists of 3 strong playoff runs: the 2003 Final, the 2006 Conference Final and the 2007 Cup. What’s striking is that, due to the lockout and high churn of the current NHL, very few players were part of Anaheim’s 3 playoff runs; in fact, there were only 4 players from the 2003 team left when the Ducks made their 2006 run: Samuel Pahlsson, Rob Niedermeyer, Andy McDonald and Jean-Sebastien Giguere. Giguere, of course, is the key, as he was almost solely responsible for Anaheim’s breathtaking run in 2003, and was in nets again for the 2007 Cup run. Maybe Giguere is the key, the elusive “playoff goalie”? This is possible, although it was Ilya Bryzgalov in nets in 2006 and Jonas Hiller when the Ducks upset the Sharks this year.
It is possible that the 2003 run is in fact the outlier, and the Ducks’ playoff success since 2006 is the result of either their personnel (Scott Niedermeyer, Ryan Getzlaf, etc) or of the physical style they employ that is better suited for the postseason. Perhaps it is possible that
if we factor in the 2009 playoffs, the two other teams that have exceeded expectations over the last decade are this year’s Eastern Conference Finalists Penguins and Hurricanes. The Penguins have excelled based on their last 2 seasons, and indeed they may yet prove to be a top-notch playoff team should they get to the final; their young core has the ability to put together one or two more deep runs in the next few years. The Hurricanes got to the final in 2002, won in 2006 and have reached the conference final this year; the only players to be part of all 3 successes are Rod Brind’Amour and Erik Cole, although Cole missed all but 2 games of the 2006 playoffs and both have been non-factors in this year’s playoffs so far. So, Cam Ward’s playoff reputation notwithstanding, I’m lumping them in the “lucky” category for now. Defeating Pittsburgh or another run to the conference finals led by Ward and Eric Staal would change my mind.
While most teams have performed to expectations, there are 5 teams who have disappointed more than the rest: the Predators, Bruins, Canucks, Red Wings and Stars. Nashville was a strong team for 2 years, in 2006 and 2007, but in both years lost to the San Jose Sharks, even though the Sharks were a lower seed both years (that’s right: the Sharks beat a higher-seeded team twice in the last 4 years!). Vancouver has simply never put together a playoff run despite being a decent team each year; Boston hadn’t even won a playoff series until this spring despite being the #1 seed in the East twice (3 times with 2009)!
The Red Wings are in a category all their own: despite being the most successful playoff team of the last decade, they have underperformed simply because of the unrealistically high expectations they have placed on themselves by being one of the top teams of the league every single season; if the standings for all 7 seasons were combined, Detroit would have won the Presidents’ trophy by 68 points! Interestingly, the single most successful regular season team of the last decade was the 2006 Red Wings, who lost in the first round to Edmonton. Not surprisingly, those Oilers are the biggest playoff surprise over the 7 years of the analysis, while those Red Wings are the biggest disappointment. When you see the Wings playing in the final in two weeks for what seems like the umpteenth time, try telling yourself that they’ve actually performed below expectations in the playoffs.
The single worst team is the Dallas Stars. While I could chronicle the Stars’ playoff disappointments of the last decade, let’s not forget that the 2 immediately preceding years the Stars won the Cup and then returned to the Final, so it’s not like they’ve always been failures. While many will (justifiably) blame the 2004 and 2006 debacles on Marty Turco (the 1999 and 2000 Stars had Ed Belfour in nets), Turco has somewhat redeemed himself over the last 2 years.
If I had asked you at the beginning of this article: “Which team, in your opinion, has most underperformed in the playoffs since 2001?”, you undoubtedly would have answered: “The Sharks!” Those of you who were sleeping in the spring of 2007 might also have added: “The Senators!” Yet neither the Sharks nor the Senators are anywhere near the bottom of the list. The Sharks, in particular, have slightly exceeded expectations! Clearly, this shows that my analysis is bunk and I have no idea what I’m talking about, since everyone knows the Sharks are chokers. Seriously, with the 2009 debacle excepted (check out my colleague Timo Seppa’s article on this for more about the Sharks’ choke job), the Sharks have actually performed up to expectations over the last few years. A team that obtains between 100 and 110 points in the regular season will, on average, last 2 rounds in the playoffs. Many will lose in the first round, like the Bruins and Predators of years past, while a select few will make a run at the Cup. People focus on the Sharks because they’ve always been a good team, but the fact is with the parity of the NHL, you’re never that good. Maybe that’s the lesson from all this: no team is good enough that a run to the Final is guaranteed, so if your team makes it, enjoy it: you may not see it again. Unless you root for the Ducks.
Tom Awad is an author of Hockey Prospectus.
You can contact Tom by clicking here or click here to see Tom's other articles.