Weird time to start sucking, but okay: Snapshot 4

We’re about to do the snapshot, but we’re gonna have to go a bit deeper than usual.

The typical data in this column includes the whole season to date. In that sample the Caps look great: top ten in controlling shot attempts, top-half in expected goals, a mortal lock for the playoffs. But since the all-star break, the picture is not so rosy: just 15th in controlling shot attempts, 23rd in expected goals, and the 25th best record in the league.

In today’s snapshot, we look around the room, make panicked eye contact with each other, and mouth the word yikes. This is not where the Caps want to be right now.

Forwards

Defenders

Glossary

• GP – Games played.
• TOI – Time on ice in minutes
• SA% – Shot-attempt percentage. The share of total shots attempted by Washington while the skater is on the ice. 50% means even.
• SA% Rel – Relative shot-attempt percentage. The difference in SA% when the player is on the bench versus on the ice. 0% means even.
• GF% – Goals-for percentage. The share of total goals scored by Washington when the skater is on the ice. 50% means even.
• PDO – The sum of Washington’s shooting percentage and saving percentage when the skater is on the ice. 1 means even. The acronym doesn’t stand for anything, and yes, that it is annoying.

Notes

• Like I said, the Caps look fine if you zoom out to the full-season level. I took the last snapshot just before the all-star break, and my headline was “maybe we’re all just spoiled.” A month later, and it’s clear I was wrong. The Capitals are trending in the wrong direction. The line graphs below are five-on-five shot share and expected-goal share over time. Fifty percent is league average.

• Like I mouthed at you earlier: Yikes. The Capitals have been dominated at even strength lately, controlling just 42 percent of the expected goals. That’s very bad. That’s miss-the-playoffs bad if the team hadn’t banked so many standings points earlier in the season.
• The hardest part of analyzing hockey is isolation. There’s so many players on the ice and involved in every play; it’s difficult to determine who’s actually the agent of a particular action. I can’t control for that here, but we can do a partial job by splitting the Caps into two groups: top-six forwards and bottom-six forwards. (I’m leaving out defenders for now since they play with both groups.) Here are line graphs for the expected-goals difference per hour for each, using Natural Stat Trick’s data. This is messy, but it’ll help us out when we get to the meat of this activity.

• The overall pattern is similar to the team-level line above. Right around game 50 and the all-star break, everything goes wrong. The first line (Ovechkin’s, in red) plummets. The second line (Vrana’s, in blue) does too. The third line (Eller’s, in yellow) has a dip but recovers quickly, and the fourth line (purple) does a surprise nosedive.
• Without jumping to a conclusion, we can at least presume that there’s no single forward we can identify as the cause of the cause of the drag. Every forward line is feeling this to some extent.
• Though let me real quick shout out Lars Eller, who — along with TJ Oshie — have been the team’s best forwards during this lull.
• Just to complete the set, here’s the same — much messier —  data plot, but for defenders. Color-coding is based on recent pairs. Check out Nick Jensen (light red) with the try-hard effort lately. Otherwise, I don’t think this graph is particularly enlightening.

• I’m already like 700 words into this, and we haven’t really gotten to the real snapshot yet. Still, I think it’s more important to set the stakes — the Caps have been genuinely bad for two weeks — and establish some kind of scale: this is systemic. The snapshot is all about looking for patterns for individual players and then diving deep. If I were to do that without having this foundation up front, we’d have a can’t-see-the-forest-for-the-trees problem.
• The trade deadline is coming up. Ian and I have chatted about how we expect the Caps to be quiet, and the recent troubles have not changed my mind. Swapping a player out might even do more harm than good — especially if it’s a depth forward. The team’s bottom-six centers, Nic Dowd and Lars Eller, have improved the team’s shot share by 5.0 and 5.1 percentage points respectively (SA% rel). I’m worried about the fourth line’s swoon, but in general I think the Caps’ depth forward roster is very strong.
• Up the lineup a bit, Evgeny Kuznetsov has missed a couple games to injury, and his wingers have done just fine without him (Oshie’s had two goals). The difference between the Caps’ shot-attempt percentage when Kuznetsov is on the bench versus on the ice is minus-8.7 percentage points. The same stat in expected goals is minus-8.1 percentage points. Here’s a line graph that shows the difference in Washington’s expected goals minus their opponents’ expected goals, depending on whether Kuznetsov is on the ice (red) or on the bench (blue).

• As genius as Kuznetsov is deep in the offensive zone, he’s a liability overall. And yet, seeing how the Caps were struggling when he was off-ice before his injury and continue to struggle now that he’s out of the lineup, that is further evidence that we can’t blame this trend on a single person player.
• Dmitry Orlov, Washington’s best defender, has the highest on-ice shot-attempt percentage (SA%) and relative shot-attempt percentage (SA% rel) among all D. He’s the only full-time defender with positive relative numbers, which means it’s really good to be his partner. Still, I’d rather see him paired less with Jensen in a shutdown role. Whether you need a goal or if you just need to counter an elite opponent forward, Orlov and John Carlson are up to the job (60.7 percent of shot attempts together, 58.7 percent of expected goals, low rates for opponent offense by any measure). I know that seems like a top-heavy defense, but at least we know it works. That’s something to build off for the other pairs.
• I try to keep goaltending and special teams out of this, but here’s something short to close on that I find so anguishing about this losing streak: it’s wasting Braden Holtby‘s turnaround. Holtby put up a .931 (all situations) in the Arizona loss, a .941 in the Colorado squeaker, a 30-minute shutout in relief in the Islanders loss, and a .935 in the Los Angeles win that was a disaster until Ovi saved it. (He also saved .720 against Philadelphia, but I’m conveniently not even mentioning it here.) For most of the season, Holtby had been the team’s biggest flaw. Now he’s finally putting it together — as all faithful Holtby stans knew he would —  and everything else has fallen apart. Maybe this is a perverse kind of encouragement. There’s still a lot of meat on this bone. If you throw back-to-form Holtby in a pot with some broth, some potato, and some competent even-strength play, baby, you’ve got a stew going.

This story would not be possible without Natural Stat Trick. Please consider joining RMNB in supporting NST on Patreon.