© 2013 Michael Parkatti History3

Relative Strength of Cup-Winning Teams Since 1968

I recently started a new series of posts tabulating Corsi data for notable games in Oilers history, beginning with their 1987 Game 7 Stanley Cup Finals victory over the Flyers.  It was a fun little exercise, and provided me a legitimate excuse to revel in the past — a favourite pastime of any Oiler fan.  The game exceeded even my nostalgic expectations, with the Oilers claiming 72% of all shot attempts on the way to a 3-1 victory.  But what I couldn’t stop thinking about was this — was this Oiler team of the 80s some kind of super-machine from the past that the advanced stat community today could not fathom?  Were they a 70% shot attempt team all the time?  It’s pretty well-known that that team was able to sustain some ridiculous percentages over the decade, but what about sustaining ridiculous shot rates?  What kind of monster is this?  How would they compare to championship teams of the past and future?

Well, there’s really no way to go back to estimate shot attempts — the NHL didn’t start tracking that stuff at a team level until about 1998, and didn’t start tracking them at a player level until 2007.  There’s also no way to find even strength differentials, as this stat was also a relatively new invention. So I decided to compile plain old shot on net differential.  I wasn’t aware of a place that had total team shot attempts over a season earlier than 1998, so I decided to add up game sheets, one at a time.  For this, I used a wonderful site called the Hockey Summary Project.

My methodology was pretty simple.  Go down the list of every Stanley Cup winner since initial NHL expansion in 1967-68, and add up the shot differential of every game they played in their playoff runs before winning the Cup.  Even though we could look at more modern measures like 5×5 Corsi for modern winners, we want to find a measure that is apples to apples across eras and statistical availability.  But I think overall shot differential is still a reasonable proxy for team strength — this past 2013 season, for instance, overall team shot percentage had a 96.9% correlation with 5×5 shot percentage and a 94.0% correlation with 5×5 Fenwick percentage.

Now, the first obvious question would be: is there era bias?  Have Stanley Cup winners over time shown a distinct pattern of increasing or diminishing strength over their playoff runs that would bring into question an even comparison between them?  As it turns out, not really — here is a graph of each Cup Winner’s playoff shot percentage since 1968:


Adding a trendline suggests that there is a slight positive bias, but it is very weak — you’d add about 0.01% each passing season to each Cup winner’s expected shot percentage in the playoffs.  It seems that in every era it is possible for both strong and weak shot percentage teams to win the Cup.

Now, to the fun part.  Here’s the full list of 44 Stanley Cup winning teams ranked in order of their playoff overall shot percentage:


First off, what you’ll notice is that the top-ranked Edmonton team was the 1986-87 team that I just watched this past weekend, placing 10th out of the 44 teams at 54.9%.  The high-flying 1983-84 team was ranked 12th, while the 1993-94 Rangers team that was essentially another Oilers team (heh) was ranked 13th.  Waaaaaay down at Rank 43 out of 44 is the venerable 1990 Oilers Cup team with 47.5% shot percentage during their playoff run.

The top team is the 2008 Detroit Red Wings team who are way out in front at 60.7%.  Fittingly enough, the 2nd place team was the Red Wings edition from 11 years earlier (1997) with 58.8%.  3rd place is taken by the 2000 Devils team who perfected the art of defence — in one game in the 2nd round of the playoffs versus the Leafs, they held Toronto to SIX shots against over an entire game.  6.  Incredible.  I won’t talk about who came 4th.

You’ll notice that the late 70’s Montreal juggernaut holds 3 spots within the top 10 teams ever, placing 5th, 8th, and 9th. The Beliveau-led Habs from 1968 ended up 6th.

The best Islanders team from their run in the early 80s was the 1982 edition, who captured 54.3% of shots on net, ranking 14th.  The best the early 70s  Bruins were able to do with Bobby Orr at the wheel was the 1970 edition, with a shot rate of 52.9% or 21st in this group.  This actually turned out to be the median shot percentage among all Cup-winning teams.

The Lemieux-led Penguins were one of the poorer-placing teams on this list, with the 1991 and 1992 teams placing 40th and 39th, respectively, both under 50%.

What do the bottom teams have in common?  Well, the worst 4 teams all had Conn Smythe-winning goaltenders. 6 of the bottom 10 teams had Conn Smythe winning goaltenders.  The eyes of Conn Smythe voters were not deceived — their goalies did a lot to secure these teams’ victories.  Of the top 10 teams, only one had a Conn Smythe winning goaltender (which hilariously was Mike Vernon in 1997, an overrated goaltender who was pulled multiple times that year in front of one of the most dominant teams ever).

The lesson in this is the same in any application of shot metrics — outshooting an opponent is important, but not what decides victory.  A goaltender can get hot for an entire playoffs (see: Ken Dryden 1971) and steal his team a Cup.  There are also instances of teams who had crappy goaltending (early 90s Flames, late 90s/early 00s Blues), who were denied shots at the Cup.

What’s interesting to me is this: that 1987 Oilers team that had 54.9% shot differential in the playoffs were only a 49.5% shot differential team in the regular season (yes, I added up all 80 games).  I ran a quick t-test to test the underlying sample means, and there is approximately a 0.1% probability that these two teams had the same game-by-game shot differential means.  These were literally two statistically different teams taking the ice, one coasting their way to a President’s Trophy, the second putting their skates to the throats of the best teams in the league.  At this point in time, it looked like the Oilers really were learning how to “take it up a notch” for the playoffs.

Post a Comment

Your email is never published nor shared. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>