I was having a great discussion with Kyle Alexander and Dan Stewart on twitter today. Kyle was talking about a comment by the coaches of the Tampa Bay Lightning that cited Goals Against per Game as evidence of the team’s defensive troubles. His point was that the team was by other measures much better defensively.

TB gives up the 2nd fewest shot attempts/60 (only Detroit is more stingy) and the 6th fewest SOG/game. Team defense? 100% a “top 10” unit. — Kyle Alexander (@kalexanderRC) February 15, 2015

The problem? Ben Bishop’s .9142 5v5 SV%. 38th out of 54 in the NHL for goalies w/ <500 minutes TOI. He’s been below average. — Kyle Alexander (@kalexanderRC) February 15, 2015

Because everything’s really about goaltending if you try hard enough, I happily hijacked Kyle’s point and we started talking about how the defense affected Bishop instead of talking about how Bishop was affecting the defense. Hahahah. See what I did there?

The issue for me, as someone who watches this team through the lens of goaltending, is that there are a lot of defensive errors that lead to rapid and unpredictable changes in situation for the goalies. The defense is a young defense and there are a lot of turnovers in the defensive zone or just over the blue line and there are a lot of breakaways and (although I never got around to talking about this one) there are a lot of dropped coverages. In essence, there are a lot of shots that are tougher for a goalie to anticipate and that has an effect on Bishop’s performance.

Kyle knows the Lightning’s scoring chances much more thoroughly than I do, as he manually tracks them. And he’s correct when he says that “Tampa outchances it’s opposition by a hefty margin on a nightly basis.” They’re definitely a very good possession team.

From the goalie’s perspective, though, the difference between what the Lightning are able to do in the offensive zone and what they’re doing in the defensive zone is stark. Using War On Ice’s team scoring chances, the Lightning are the top team in the league in 5v5 SCF% (or the percentage of scoring chances that are for the Lightning.) They’re 3rd best in the league at generating Scoring Chances For per 60 at 5v5. But they’re only 11th best in allowing Scoring Chances Against per 60.

Now, 11th isn’t 15th. The Bolts are certainly better at limiting scoring chances than they are at limiting goals against. But it’s definitely not 3rd, either. The bulk of the Lightning’s dominance in scoring chances is coming 180 feet away from Ben Bishop. And that does matter to a goalie. It is important context.

In fact, Ben Bishop’s 5v5 Adjusted Save Percentage (a number at War On Ice that accounts for shot locations) is significantly higher than his unadjusted 5v5 Save Percentage. Only two goalies with 500+ 5v5 minutes see a bigger change when accounting for location. Only three goalies move up in the rankings by a bigger margin. He has played more than twice as many games as any of them.

Name Gm Sv% AdSv% Diff. SV% Rank Adj Sv% Rank Diff Alex.Stalock 14 90.53 92.48 1.95 45 34 11 Anton.Khudobin 20 90.25 92.06 1.81 49 39 10 Ben.Bishop 44 91.3 92.62 1.32 38 30 8 Chad.Johnson 15 88.58 89.73 1.15 54 53 1 James.Reimer 24 90.51 91.61 1.1 46 45 1 Niklas.Svedberg 13 91.84 92.93 1.09 32 23 9 Karri.Ramo 19 91.42 92.49 1.07 37 33 4 Niklas.Backstrom 19 88.6 89.56 0.96 53 54 -1 Marc-Andre.Fleury 43 92.39 93.34 0.95 19 13 6 Corey.Crawford 37 92.21 93.15 0.94 23 17 6 Jaroslav.Halak 42 92.3 93.22 0.92 21 15 6 Mike.Smith 41 89.97 90.82 0.85 51 49 2 Jhonas.Enroth 38 91.5 92.34 0.84 36 37 -1 Ondrej.Pavelec 35 92.11 92.94 0.83 27 22 5 Viktor.Fasth 25 89.47 90.3 0.83 52 52 0 Ray.Emery 23 91.27 92.03 0.76 39 40 -1 Petr.Mrazek 20 92.72 93.47 0.75 16 11 5 Eddie.Lack 17 92.11 92.86 0.75 28 26 2 Semyon.Varlamov 36 92.17 92.92 0.75 25 24 1 Kari.Lehtonen 47 92.03 92.75 0.72 29 28 1 Cam.Talbot 17 91.67 92.39 0.72 33 35 -2 Jonathan.Quick 47 92.19 92.9 0.71 24 25 -1 Dustin.Tokarski 12 91.05 91.76 0.71 41 43 -2 Ben.Scrivens 38 90.4 91.11 0.71 48 47 1 Jonathan.Bernier 39 92.73 93.43 0.7 15 12 3 Jake.Allen 25 91.58 92.26 0.68 35 38 -3 Jonas.Hiller 38 92.34 93.02 0.68 20 20 0 Curtis.McElhinney 22 91.24 91.9 0.66 40 42 -2 Cam.Ward 36 90.99 91.63 0.64 42 44 -2 Antti.Niemi 43 91.91 92.5 0.59 30 32 -2 Michal.Neuvirth 19 92.57 93.12 0.55 18 18 0 Ryan.Miller 42 91.86 92.37 0.51 31 36 -5 Thomas.Greiss 13 92.16 92.6 0.44 26 31 -5 Frederik.Andersen 44 92.24 92.67 0.43 22 29 -7 Al.Montoya 16 90.21 90.63 0.42 50 51 -1 Cory.Schneider 48 93.52 93.93 0.41 8 7 1 Jimmy.Howard 34 92.62 93.03 0.41 17 19 -2 Michael.Hutchinson 27 93.52 93.92 0.4 9 8 1 Braden.Holtby 48 92.76 93.15 0.39 14 16 -2 Sergei.Bobrovsky 33 91.62 91.98 0.36 34 41 -7 Steve.Mason 34 93.83 94.17 0.34 6 5 1 Robin.Lehner 24 90.99 91.32 0.33 43 46 -3 Carey.Price 44 94.13 94.44 0.31 4 3 1 Devan.Dubnyk 32 93.52 93.81 0.29 10 9 1 Antti.Raanta 13 94.26 94.53 0.27 3 2 1 Carter.Hutton 14 90.84 91.08 0.24 44 48 -4 Calvin.Pickard 15 95.43 95.62 0.19 1 1 0 Darcy.Kuemper 29 90.49 90.67 0.18 47 50 -3 Tuukka.Rask 46 92.84 92.97 0.13 13 21 -8 Brian.Elliott 28 93.97 94.07 0.1 5 6 -1 Henrik.Lundqvist 39 93.27 93.28 0.01 11 14 -3 Craig.Anderson 31 93.66 93.6 -0.06 7 10 -3 Roberto.Luongo 44 92.97 92.75 -0.22 12 27 -15 Pekka.Rinne 42 94.47 94.24 -0.23 2 4 -2

Adjusted Save Percentage doesn’t take into account all aspects of shot difficulty, of course. We don’t have any big data measure that really captures what makes one shot (or scoring chance) more difficult for a goaltender to stop than another. It does hint, though, in a way that’s visible in numbers, that there is something to look at in regards to what is happening with Bishop.

Ben Bishop is simply seeing a different mix of shots than the average goaltender in the NHL this season. He’s seeing more High Difficulty shots and fewer Low Difficulty shots. The league average mix of High/Medium/Low shots this season is 27.82%/26.77%/45.41%. Bishop’s seeing 31.80/26.83/41.37. That’s the second highest increase in High Difficulty Shots among goalies with 500+ minutes. [Data from http://war-on-ice.com/goalietable.html. ]

It would take a different project altogether to point out how high difficulty shots affect goalies or how all scoring chances are not created equal. Or to show those breakdowns happening in specific games and the specific consequences. Those are all projects worth doing, if only to show a different way of looking at how plays develop and break down. But for now, I hope to show that the numbers back up the idea that there’s a team effect here that makes Bishop’s experience different from the rest of the league.

This isn’t an indication of exactly what practices are driving that difference. It is an indication that there is a difference. And like I noted on twitter, I don’t believe the crux of the matter is the system the Lightning want to play as much as it is the errors in execution. The system is less to blame than the ways that players deviate from the system. Broken passes. Turnovers, especially close in, leading to undefended attacks and pucks coming in from unexpected angles. Screens. Bad pinches. All of these and more.

Furthermore, this isn’t an indication that goaltending is not a factor in goals against. A large part of the difference between Ben Bishop’s .932 ESSV last season and his .913 ESSV this season (or his .940 ADJ SV last season and his .926 this season) is that he was stopping pucks regardless of defensive errors last year. This year, he isn’t. Part of that—albeit a part of unknown size—is differences in technical execution on Bishop’s part.

But drilling down into the numbers can show that possession dominance on its own doesn’t give enough of the story to eliminate team defense as a factor in team goals against.

Advertisements