I was slightly disappointed by the Capitals’ loss to the Leafs on Saturday night. I hoped for a bounce-back performance after that humiliation at the hands of the Devils, but instead the Caps got mildly outplayed and earned themselves a 2-2-1 record after two weeks.

After the game, superfan Will asked me:

We worried yet?

— Regal Beagle (@will_po) October 14, 2018

The short answer is no. The longer answer is, wellllllllll…… ¯\_(ツ)_/¯

No Caps player has clocked even 100 minutes of 5-on-5 play yet (Orlov has 93). Before I start making conclusions about this team, I’d want the sample sizes to be more than twice that large. But that doesn’t stop me from noticing where the team stands overall right now.

With 44.3 percent, the Caps rank 26th in the league in shot-attempt differential, which means their opponents possess about 56 percent of the shot attempts in games. If you adjust those percentages based on score state (i.e. weighing blowouts lower), the Caps improve a bit – up to 45.1 percent, or 23rd in the league.

Twenty-third is also where the Caps rank in 5-on-5 scoring (12 Caps goals, 15 opponent goals), though those numbers appear to be driven by early-season weirdness: Washington has the 5th highest shooting percentage and the 5th worst saving percentage in the league, resulting in an unstable-but-even PDO (the sum of shooting and saving).

Special teams have been a mixed bag, with the Caps converting 36.8 percent of power plays (4th best) and being middle-of-the-road in the penalty kill with a 77.8-percent kill rate.

All that adds up to a familiar picture: a Washington team that doesn’t dominate play but squeaks out wins when they can based on brilliant individual performance.

Here’s a picture of Evgeny Kuznetsov (four goals, five assists):

While there may be system-wide issues limiting how much the Caps control play, it’s too soon to say. For now, the loss of Tom Wilson before opening night and the resulting changes have me curious about how the lines are performing. Using the centers as proxies for each line (which makes sense since they’ve been the most stable choices), here are the early returns.

A few things pop out to me here, enough to justify bringing out the old bullet points.

• The top two lines have been wildly high-event, with the Caps and their opponents adding up to 125 shot attempts per hour. These are probably the early-season jitters, but it’s also something the team should actively address with more deliberate breakouts and focused team defense, which already looked better to me against Toronto.
• Despite Alex Ovechkin and Evgeny Kuznetsov sporting individual offense rates similar to last season, their line’s total shot-attempt rates have dropped more than 10 percent. For whatever reasons, Brett Connolly and Chandler Stephenson have not effectively complemented Ovi and Kuzy. Wingers with higher individual rates like Vrana or Burakovsky might be better choices.
• The Nicklas Backstrom line (2nd) is eating up a ton of defensive-zone draws with 26, double the number they’ve had in the offensive zone. Reirden is depending on these guys to do hard work, and they’re doing alright so far (and I expect the high-danger numbers to draw closer to the attempts as the sample matures).
• Lars Eller‘s line (3rd) has not worked so far. They’ve been outscored 8 to 1 (with 4.6 percent shooting and 77.1 saving), but the underlying numbers are rotten too. One thing that jumped out at me is the high-danger chances; Eller’s line has three in total, two of them from Lars himself. Andre Burakovsky‘s offense has been nearly non-existent so far (his various individual rates down by roughly half), especially south of the faceoff dots. Perhaps with Brett Connolly back on the line and some easier matchups, we’ll see an improvement here. Or perhaps Andre would be better served with a high-profile, sink-or-swim opportunity on the top line.
• I don’t have specific expectations for the Nic Dowd line (4th) yet, but here’s one thing they’re doing well: unlike the top six, game pace slows to a crawl for Dowd, and opponents have seen their lowest rates of scoring chances and high-danger chances – forty to sixty percent lower than what they get against Kuznetsov’s line. That’s very encouraging and a trait coaches generally like to see in their fourth lines.

Here’s the same exercise but with D-pairings, using Orlov, Carlson, and Orpik as proxies.

• The difference in deployments for Dmitry Orlov (1st pair) and John Carlson (2nd pair) jumps out at me. Orlov is being asked to start a lot of shifts in the defensive zone (33) so that Carlson’s pair can participate more in offense (39 offensive-zone starts). Goal percentages for each pair are bearing out that difference so far. This might lead to Orlov again having brutal numbers but nonetheless serving an important role with positive externalities.
• Michal Kempny has returned from concussion, but his numbers have not. Playing with Carlson on what I consider the second pairing for the last three games, Kempny’s been caved in with a shot-attempt percentage under 40 percent.
• For the Brooks Orpik pairing (3rd), which has included both Djoos and Bowey, the most curious factor so far is its steadiness. Orpik, 38, has started all five games, which have included two zero-rest games. Does this suggest that Todd Reirden has no intention of healthy-scratching Orpik, whose contract includes bonuses for playing more games? If so, the consequences of such a decision might be seen both in Orpik’s on-ice numbers, which haven’t terrible so far, but also in Madison Bowey‘s development and career prospects. We still haven’t seen well how Bowey can play apart from Orpik, but we know how they play together: horrifically.

Finally, there’s reason to believe that everything will look better in a couple weeks. Washington’s early schedule has been daunting: five opponents who all made the playoffs last season, averaging 104.6 standings points last season and projected to be roughly as good this season according to Manny Perry’s and Dom Luszczyszyn’s models. The next five games are the exact opposite: five opponents who missed the playoffs last season, averaging 81.6 standings points last season and projected to be borderline at best this season.

And there will be no back-to-back games.

While I’m not yet worried, at least now I know what I’m looking out for and what adjustments I’d like to see. I hope the team does too.

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

Headline photo: Patrick McDermott