Devils and Shot Quality

Are the Devil's a good possession team? If you don't know the answer to this question I'd like to welcome you to the ILWT community as this must be your first article. There have been loads of questions about how the Devils could have managed to underperform their possession statistics so drastically. Last week, I introduced a new stat site, Progressive Hockey, and their important consideration in their statistics, Shot Quality. This article will be a concisely posited stance on the implication of the Devils shot quality statistics.

Point 1: The Devils Underperformed their Shot Quality Numbers

Shot Quality does a great job of putting possession statistics into perspective. It's worth mentioning at this point that possession statistics aren't actually possession statistics in the literal sense. Though they are strongly correlated to possession, they are really shot differential statistics as they measure shots rather than possession time. So when we consider the fact that Corsi and Fenwick are, in reality, shot differential statistics, one way of making those more accurate is by judging the types of shots that were taken. There are those games where the Devils outshoot a team 35-16 but all on Salvador/Volchenkov shots from the point which are something short of dangerous. The other team meanwhile gets a couple breakaways and a few 2-in-1s and convert those opportunities like they should.

It is at this point in the article that we bring up how the Devils fared last year when we adjust them for shot quality. According to Progressive Hockey (Progressive Hockey: Team Stats), when adjusted for shot quality, the Devils had the 4th highest Expected Goal Differential Percentage in the league behind only LA the Shark and the Blackhawks. Also worth noting is that we finished exactly one spot above the Rangers.

With an actual GF% of 48%, the Devils were in the lower half of the league. More worthy of conversation however is that they were bottom of the league in residual goal differential (difference between expected and actual). Their goal differential was a league low 7% lower than their ExpGF%.

Point 2: New Research Implies Shooter Skill Affects SH%

Shot quality does account for some of the randomness of shooting percentage, but according to Matt Pfeffer's research, quality of shooter can account for 3-5% of the shooting percentage. In his research the largest differences came from Stamkos, Kovalchuk, and Tanguay. The first two are widely regarded as among the best shooters in the world and Tanguay has a career shooting percentage of 20%. You can't score once in every 5 shots in an entire career just by chance.

Point 3: The Difference between Marty and Cory was Massive

Just as shot quality affects shooters statistics, it can have the reverse effect on goalie stats. If a goalie is repeatedly facing high quality shots then his save percentage is going to be way lower than it should be. In light of that Progressive Hockey has included Quality-adjusted Sv%. Cory Schneider was 5th among goalies with at least 30 starts at 0.917 behind only Lunqvist, Varlomov, Price, and Rask. Furthermore, Brodeurs Qual-Adjusted Sv% was 0.892. He outperformed Reto Berro, Kevin Poulin, Devan Dubnyk, and no one else.

Conclusion: The Devils Might Need Goal Scorers, but Schneider Starting is Way More Important

When I got to point #2 I though I was going to have to contradict my own opinion in light of new information. After finding some work done that shows how players' shooting ability can affect the shooting percentage and that we monumentally underachieved our expectation, the new goal scorer additions will most likely prove to be necessary. However, after establishing point #3, it is still pretty clear to me that Schneider starting, plus the regression of shooting percentages should make us a playoff team and new additions will just be icing on the cake.

Your Thoughts

Plenty of people disagreed last time I said the Devils didn't need scorers so I invite you back out to point out new errors in this thought process. Do you disagree that Martys Qual-Adjusted statistics were more important than the Devils lack of shooter-influenced Sh%? How do you feel about Shot Quality adjustments to statistics? What are your opinions on life the universe and everything?