Hrmph. I was thoroughly unimpressed with the numbers I received from looking at the takeaway numbers and win/loss splits in my last post, so I've decided to delve into the world of Corsi in this one to get a more definitive view of how much two-way pivots mean to a Stanley Cup winner. Feel free to correct or dispute me, because I've got a very basic understanding of what the various Corsi numbers actually mean.
For those who aren't familiar with Corsi numbers, I'll try to define the ones I will be using. Corsi itself is a different measure of plus/minus based on how many shots a team/line/player directs on the opponent's net opposed to how many they allow on their own. So, if Detroit's line of Datsyuk, Zetterberg, and Holmstrom fires 7 shots at the Pittsburgh net while on the ice and only allows 4, they have a Corsi +/- of +3. It was believed by Jim Corsi (former Sabres goalie coach) that a team will always have a better chance of winning in the long run if they can garner more shots than their opponent.
As for the aspects I will be using: Corsi Relative (CR) is on-ice player Corsi rating minus the off-ice Corsi rating. This rating analyzes how good a player is relative to his teammates. Here's an excerpt from Justin Azevedo at Matchsticks and Gasoline about how to interpret Corsi Relative:
Corsi REL is a relatively simplistic way of measuring how effective a player is in driving possession relative to the rest of his team. At its most basic level, the stat reflects a player's raw [even strength] Corsi relative to the raw [even strength] Corsi to the rest of the team when he is not on the ice. For example, Robyn Regehr's Corsi REL last year was -11.4, which means ~11 shots more per 60 minutes of even strength ice time were directed at the Flames' net while he was on the ice then when he wasn't. Now, that may seem really terrible (in fact, it was in the bottom third of the league last year-lumped in with superstars like Deryk Engelland, Cody McCormick and Bryan Allen) but when you take into account a couple of other stats, a number that low is to be expected.
Corsi Relative Quality of Competition (CRQoC) is the average Corsi Relative of opposing players weighted by head to head ice-time. In short, how good of competition did a player face relative to how good that player/line was, and how often they were matched up against each other. And, in combination with CRQoC, I'll also look at Corsi On (CO), a measure of a player's Corsi Rating while on the ice.I'll start off with the 2007-2008 season, and analyze the final four teams to see how much a respective two-way center, or lack thereof, added to or detracted from their team. I'll also be taking a look only at the final four teams - in this case being Detroit, Dallas, Pittsburgh, and Flyers (links to each team's stats linked).
With a blunt look at those three statistics, it's obvious that Zetterberg and Datsyuk were the two best players in the playoffs. Pittsburgh had only one player with an even CO, Philadelphia had 3, and Dallas had 7 while everyone on the Red Wings roster with over 10 games played was a plus in Corsi rating. Datsyuk comes in fifth in CO on Detroit at 27.48 (Zetterberg led with 29.90). CR shows a similar story, with Datsyuk ranking third behind linemates in Zetterberg and Holmstrom. It gets interesting when you compare the CRQoC with these previous two numbers. Pavs and Hank didn't face as good of competition as others on the team, in part because they are some of the best competition. If you look higher up the CRQoC ranks and compare that to the player's CO, you'll notice that almost no one did as well as the Eurotwins with a similar amount of competition. Relative to teammates, the only other forwards whose CR was above 10 when facing the type of competition Z and Pavs did (about .600 CRQoC) are Jeff Carter (score!), Scottie Upshall, Mike Knuble. Zetterberg technically wasn't a center back then, but...score a few for the two-way pivot.
Taking a look at the '08-'09 playoffs final four, which featured Detroit and Pittsburgh again along with Carolina and Chicago, reveals some interesting things. As far as the Red Wings go, Datsyuk was the our best player, leading in both CO (26.20) and CR (16.9), although he faced some pretty abysmal competition (-.600 CRQoC). The only forward on the Wings who excelled when facing some good competition was Marian Hossa. Pittsburgh's best forward was Tyler Kennedy, who faced constantly faced great competition competition (2.305 CRQoC) and still put up the second highest CR and CO on the team. Carolina had a few extremes. Matt Cullen faced the other team's best almost every time (4.107 CRQoC) and had middling results, while Eric Staal played fourth liners and third pairs (-3.778 CRQoC) to accumulate his numbers. It shouldn't be surprising that two-way grinder Chad Larose had some of the higher CO and CR numbers on the team while still playing against able opponents. Chicago's own two-way man in Jonathan Toews paced the team in both CR and CO but with a lower CRQoC. No Blackhawks who faced competition above the mean had positive Corsi numbers, relative or not. It's hard to say how Pittsburgh was able to win the Cup, based on how effective the Wings were in several Corsi categories, but that's why you play the games.
The '09-'10 playoff look won't feature the Wings, unfortunately. That year's final four were Chicago, Philadelphia, Montreal, and San Jose. Not surprisingly, Jonathan Toews placed the highest amongst those who played above-average competition on the Hawks, with Pat Kane as a surprising second. Also high in CO and CR were Hossa, Sharp, and Byfuglien. The Hawks won with a defensively responsible team who were still able to out-shoot their opponent - simple as that (also, tell me if that formula sounds familiar?). Another surprising statistic is how far the Flyers were able to get on poor possession play. Only five players were positive in CO, with their most effective forward against competition being - surprise, surprise - Ville Leino. James Van Riemsdyk was the Flyers best forward, relative to teammates. Montreal was even worse, with PK Subban as the runaway best player and only one above zero in CO. Comparing CRQoC and CR, Scott Gomez and Brian Gionta were Montreal's most effective forwards. As far as the Sharks go, Logan Couture, Manny Malhotra, Torrey Mitchell, and Joe Pavelski were the most important forwards in driving possession for the Sharks. Much less-than-stellar two-way play from stars Marleau and Thornton seems to have been a contributor to a conference finals exit.
And finally, this past season's final four were Boston, Vancouver, Tampa Bay, and San Jose. A lot of players on Boston played very well, so it's tough to pinpoint if anyone was more instrumental than another. However, centers Brad Marchand and Patrice Bergeron faced the toughest competition on Boston and both still managed solid CO and CR numbers. The Bruins first line of Lucic, Krejci, and Horton were much more efficient, although they played against much weaker opponents. Vancouver's best forward against good opponents was Jannik Hansen, Alex Burrows was very good against slightly above-average competition, and two-way stalwart Ryan Kesler was also a solid contributor. The Sedins were the best players on the team, and their CRQoC numbers may also be skewed a bit much like the Eurotwins (especially since Burrows plays with them and had an almost 1 unit difference). Tampa is a very interesting team, with Marc-Andre Bergeron sporting far and away the best CO I've seen at 53.3. Tampa's top three forwards were Steve Downie, Sean Bergenheim, and Teddy Purcell. I would have thought St. Louis and Lecavalier would have been top players even playing against great competition as they did, but they were well below where they should be. Joe Thornton played much better for the Sharks, and was their best forward along with Devin Setoguchi and Patrick Marleau, even though they all faced great opponents. Unfortunately, they were the only three good forwards against good competition, as Couture, Pavelski, and Mitchell all fell off from the year before.
And with that somewhat small sample size, I think I've discerned what I expected when I undertook this endeavour. Players like Datsyuk, Zetterberg, Toews, Hossa, Bergeron, and other two-way players without quite as much skill - guys like Cullen, Larose, Kennedy, Kelly - were important pieces of Stanley Cup winners or contenders. Another trait was, unsurprisingly, that the whole team seemed to be contributing in some fashion. Teams who have a deep, defensively responsible forward group where at least nine men are shooting more than they allow tend to have a good chance at winning the cup.
This doesn't bode so well for the current iteration of the Wings, however, as only Datsyuk has set himself apart from the pack. Yet somehow, the Wings continue to roll and are only two points away from the best record in hockey. Here's hoping that Babcock can provide the motivation, and the guys can find that extra gear in the playoffs and make a push for #5 for 5.