February 11, 2012
Howe and Why
Alan Ryder's 2011 NHL Review

by Robert Vollman

Alan Ryder's 2011 NHL Review came out a few weeks ago, a 90-page monster of condensed statistical analysis, and as usual it included some pretty stunning conclusions. How unimportant is defense, what part of Hockey Prospectus 2011-12 was "bollocks", which non-Vezina-nominated goalie beat out Tim Thomas as the league's best goalie, which non-Norris-nominated defensemen finished in a three-way tie as the league's best blueliner, and while Anze Kopitar was 10th, which teammate was ranked the league's 18th-best forward (hint: it wasn't Dustin Brown, Justin Williams, or even Ryan Smyth)?

Alan Ryder

Those are all good questions, but the better question is this: Who is Alan Ryder and why should anyone with even the smallest interest in statistical hockey analysis go read his 2011 NHL Review immediately?

Think of Alan Ryder as the Nicklas Lidstrom of advanced statistical hockey analysis, because he's been in the game longer than anyone else, and is still among the very best. The statistics and principles used by those who believe our field began some time between 2006 and 2009 were built on the foundation that Ryder helped build a decade ago. Not to oversell his work, but why study something second hand when you can hear it directly from one of the few statistical analysts I know who was actually alive when Kansas City had an NHL team?

We're presenting but a teaser sampler of some of Ryder's more provocative findings, to hopefully entice those who remain reluctant to go download and read his tome for themselves.

The Unimportance of Defense

Some of the more fascinating insights begin shortly after the preamble about the terms and methodology he uses, including his assertion that team defense correlates to points only half as well as offense.

"They" say that defense wins. New Jersey failed to win a playoff berth yet led the NHL in team defense. Anaheim was fourth seed in the West with the NHL's second worst defense. Boston won a Stanley Cup with a defense that was as productive as that of Edmonton (don't misunderstand my comment—the Bruins chose to invest skating energy in offense because of their goaltending).

In 2011, the correlation between defense and winning was weaker than I have seen it in some time. The correlation coefficient for points and MGD was just 35% (versus 69% for MGO and 39% for MGG).
–Alan Ryder, 2011 NHL Review, Page 24.

Based on Ryder's approach, which includes concepts like shot quality, Boston's defense is as bad as Edmonton's. Given the unusually high save percentages of their goalies, one would normally assume their defense is outstanding, especially since otherwise unimpressive goalies like Alex Auld's save percentage was .919 in Boston, higher than it was during his tours with Vancouver (.907), Florida (.888), Phoenix (.880), Ottawa (.906), Dallas (.894), NY Rangers (.904), and Montreal (.914).

Surprisingly, Tim Thomas wasn't the league's best goalie, despite his league-leading .938 save percentage and Vezina Trophy while playing behind an Oiler-level defense—that honor went to Carolina's Cam Ward.

Bollocks

We've received our fair share of criticism over at Hockey Prospectus, but we've never had our work called bollocks before.

In the past couple of years there has been a fair bit of discussion amongst hockey analysts of the possibility of "parity" (a narrow range of results) amongst NHL goaltenders. This is a very important conversation as a very narrow range implies low value added from goaltending. If parity is the case, then goaltender wages are high since parity would imply that goaltending assets are quite fungible (which implies lower wages).
–Alan Ryder, 2011 NHL Review, Page 25

If the supporting quote on page 25 looks familiar, it's from the Detroit Red Wings team essay of Hockey Prospectus 2011-12. While we defended Detroit's choice of using mediocre low-cost goalies like Chris Osgood or Jimmy Howard, Alan Ryder considers our premise to be "bollocks." Howard's performance this year seems to justify our argument, or it might be merely a fortunate and temporary coincidence.

In our defense, we don't entirely agree with Ryder's explanation of a goalie's importance on page 26—his math only adds up if you believe goalies are 100% responsible for save percentage. If true, it seems suspicious given that save percentages bounce around like ping-pong balls from season-to-season—defense and luck must have at least something to do with it.

Jarret Stoll, Superstar

While Ryder was one of the first hockey analysts to explain why settling a hockey game with a shootout correlates about as well to team skill as a coin toss, it nevertheless weighs heavily in his calculations.

While the shootout looks like a lottery, ignoring it is dangerous math. Over 10% of NHL games get resolved in this contest. That means that a goal scored or prevented after skating time is nearly as valuable as an earlier event. Think twice about Alex Tanguay's 10 shootout goals.
–Alan Ryder, 2011 NHL Review, Page 34

That's precisely why Jarret Stoll makes an interesting appearance as the 18th-best forward despite just 43 points, minus-6, and a lackluster reputation among fans. However, it's important to realize that Player Contributions measure how much a player contributed, not how good they are. Some players' contributions are entirely a consequence of their skills, others have generous helping of good fortune.

Other interesting findings on page 69 include Nikolai Kulemin ranking as the league's 10th-best left winger. Loui Eriksson? 2nd—thanks to Transitions.

Transitions

One of our favorite concepts is Transitions, which in a fashion roughly similar to Corsi or plus/minus attempts to measure how significantly a team's fortunes change (for better or for worse) when a player is on the ice, based on penalties, turnovers, faceoffs, and more.

Where Zajac wins the Gainey prize is with his defensive transition game where he led the NHL with 19 PCD(TR). This is mainly from penalty avoidance (just 7 minors). But he was +26 on turnovers and +120 on faceoffs, both of which tilt the ice away from your net. Other transition leaders were St. Louis (17), Loui Eriksson (16), Tyler Bozak (15) and Ilya Kovalchuk (13).
–Alan Ryder, 2011 NHL Review, Page 44

This kind of innovation is unique to Player Contributions, and works to quantify the value that players like Travis Zajac and Martin St. Louis bring to the game that aren't being otherwise directly recorded by other metrics.

Norris Oversights

One final intriguing finding can be found on page 46, as the Bobby Orr award for top defenseman doesn't go to any of last year's Vezina finalists, but rather a three-way tie between the unlikely trio of Alex Pietrangelo, John Carlson, and Brett Clark—not exactly the names anyone was expecting. A fascinating result, especially in the case of Brett Clark, who is rarely a preferred option to shut down top opponents in Tampa Bay.

Verdict

Some of Ryder's findings may seem peculiar at first, especially since we've only highlighted the most unexpected ones here, but make sense when you dig a little deeper. Perhaps Cam Ward truly is one of the league's most underrated netminders, Boston's defense could be overrated, and players like Jarret Stoll and Brett Clark could prove to be trade deadline steals.

Whether you agree or disagree with any of these interesting highlights, go read Alan Ryder's 2011 NHL Review, because everything he writes is explained in detail—and often proves true in the long run.

Robert Vollman is an author of Hockey Prospectus. You can contact Robert by clicking here or click here to see Robert's other articles.