Forty games in: Defensemen

The Washington Capitals are two and a half years removed from being one of the best defensive teams in the NHL. A lot has changed between then and now, including, ya know, a Stanley Cup Championship. To win their Cup Washington had to overcome a bunch of defensive problems, which now appear resurgent. If they wish to truly go back-to-back, the Caps will have to solve those problems all over again.

But whether defensive problems are actually the same thing as problems with defensemen is another matter altogether.

On Ice

• TOI = time on ice during 5-on-5 play
• SA% = shot-attempt percentage, the share of total attempts by either team that belong to WSH
• xGF% = expected goals-for percentage, the share of total expected goals (a function of shot volume and shot quality) by either team that belong to WSH
• GF% = goals for, the share fo total goals by either team that belong to WSH
• ZSR = zone start ratio, the share of non-neutral zone starts that begin in the offensive zone
• Sh% = team-level shooting percentage for WSH when player is on the ice
• Sv%  = team-level saving percentage for WSH when player is on the ice

Scoring, All Game Situations

As of January 5

Observations

• For obvious reasons, defensemen take the blame for a team’s defensive problems. The Caps, at least since the end of the 2016-17 season, have certainly had defensive problems. As a result, the blueliners get the first and most scrutinizing looks – but the picture isn’t tidy to me. For starters, there have been (or *had been*) some important progress in limiting opponent chances over the fifteen-to-twenty games.

• As the presumed number-one shutdown defenseman and a generally well-respected player, Matt Niskanen has nonetheless seen a huge drop-off since 2016-17. In late November, Kevin at Japers Rink lamented Niskanen’s struggles, which he (and I also) suspected were caused by injury. Since then — since pretty much exactly then — Niskanen seems to have turned his game around, driving play in Washington’s favor during his shifts. I have a hypothesis about how this has happened, but I’ll save it for another time.

• We can dismiss one possible cause for Niskanen’s and the team’s turnaround. Defensive pairings have not been significantly altered in the last twenty games (aside from third-pair fiddling). Our top four is pretty stable: Orlov-Niskanen and Carlson-Kempny. You can see their basic stats (shot attempts, expected goals, and actual goals) as pairings below, along with every other pairing that has been together for more than 15 minutes.

• I utterly hate those numbers for Dmitry Orlov and Niskanen. They’re playing a ton of minutes (512), but the Caps are doing poorly within them (under 45 percent in shot attempts and expected goals, well under 40 percent in goals). There are two comforting pieces of context I’ll offer. First, Orlov-Niskanen has improved significantly in the last month, and second, their usage has been noteworthy and may explain their struggles somewhat. I’ll address the latter point in a bit.
• Another way of viewing defensive pairings is below. Where each row and column intersect is the percentage of shot attempts that belong to the Caps when those two defensemen are on the ice together. Only pairings with more than twenty minutes together are shown.

• So we see that Orlov and Niskanen are having lots of trouble regardless of their partners. We also see that Madison Bowey gets his best results when with Brooks Orpik, and the best numbers of all pairings come from Jonas Siegenthaler and Christian Djoos.
• Don’t be alarmed; that’s not some secret synergy: Siegenthaler and Djoos played nearly all of their minutes together in dominant games against the Coyotes and Rangers, two weak teams. It’s a side effect of irregular schedule strength. Jonas Siegenthaler didn’t play in those very tough late October/early November games; he’s instead gotten most of his ice time from the much more forgiving December schedule. Let’s see what happens in March, a brutal stretch of games for the Caps.
• Aside from a nagging hand injury, one of the likeliest causes for Niskanen’s problems might be his usage. This chart shows where defensemen take their zone starts (left means more defensively, right means more offensively) and the quality of their competition (using their opponents’ average expected-goals percentage). The size of the circle means the percentage of 5-on-5 ice time the player gets (bigger means more). The very small blue dot is Brooks Orpik.

• It looks like Niskanen is getting the majority of the tough defensive-zone draws — perhaps to shelter younger defensemen like Siegenthaler or to optimize scoring defensemen like Carlson. Niskanen’s zone-start ratio of 41 percent is the 13th toughest among 191 defensemen with at least 300 minutes played. He’s got a bafflingly tough workload.
• On the flip side, we can see that John Carlson‘s deployments are favorable, and he’s absolutely reaping the benefits. Carlson is five points from the league lead in points among defensemen, and one could even argue he’s been a bit unlucky in lacking secondary assists.
• Carlson seems to have a good thing going with partner Michal Kempny. Kempny looked rough early in the season as he recovered from a concussion, but his play of late with Carlson has been solid. Penalties remain a bit of a problem for him though.
• I’ll admit that I don’t yet have a cogent position on Jonas Siegenthaler. I like what I’ve seen, but his games have been easier, his opponents less skilled, and deployments very favorable. Those are ingredients for an exciting NHL debut, but I doubt they reveal the quality of a player. I’ll keep an eye out, but I’d like to see Siegenthaler get tested a bit more — more ice time and tougher matchups. Let’s see what we’ve got here.
• Brooks Orpik continues his impressive age-38 season, outscoring opponents 11 to 9 with similarly positive underlying numbers. The usage chart makes it clear that Orpik is being sheltered (he’s the small blue dot in the bottom-right quadrant). Orpik played less than 15 minutes a night since the game on new years’ eve, in which a 91-second (!) penalty-kill shift ended in a backbreaking goal for the Predators.

• That’s not necessarily a fault of Orpik, and his PK numbers have been perfectly respectable — though he appears to be getting burned by Washington’s goaltending. But I wonder what Orpik’s lowered ice time portends. As Christian Djoos recovers from his lower-body injury, Madison Bowey is now the odd man out — so that Orpik can play low minutes on his way to performance bonuses. This is by no means a scandal, but it at least raises my eyebrow. Is there a plan here?
• We’ll wrap up now by dropping some lovely heatmaps of opponent shot locations against the various Caps defensemen. These images show the Caps’ side of the ice, with red meaning more opponent shots coming from that location compared to league average and blue meaning fewer.

• Pretty much no one is good at limiting shots from up close, which might suggest systemic problems in Washington’s defensive play. Then again…

• I began this article with the supposition that defenseman are accountable for a team’s defensive problems, but that’s not the whole story. All players are supposed to contribute to offense and defense all the time. And if just one big-minute player has big problems in play without the puck — even if that player is a forward — it can hurt defensive stats for everyone he encounters, turning a mediocre team defense into a shambolic one. I wonder.

Author’s Note: This article should have been delivered twenty games ago along with the rest of the Twenty Games In series. I ran into some health problems after Thanksgiving that kept from me from doing the deep dive warranted at the time. I’m sorry for the delay, and I thank you for your patience.

This article would not be possible without Corsica, Hockeyviz, and Natural Stat Trick. Please consider joining RMNB in supporting them via Patreon. All data as of January 4.

Headline photo: Elizabeth Kong