More than ten years ago, I was worried about Marcus Johansson’s play. Getting game-by-game statistics was much harder in those days, so I had to ask some favors to get good data to back up my hypothesis. Turned out he was struggling, so I wrote about it, ultimately suggesting that the Caps might want to move on from him. A couple weeks later, the Capitals revealed Johansson had been playing through a concussion. Johansson rested and recovered, and eventually he was his old self again, but I was changed by it. It wasn’t that my analysis was wrong– the numbers were “right” — but the presumption behind the bad numbers was wrong. The player wasn’t bad; he was hurt, secretly.
Since then, I’ve tried to split my analysis into different domains. The quantities are somewhat (but not entirely) objective and straightforward to discuss, but the reasons behind those quantities are a lot harder. Sometimes those reasons are what you could call “narrative” — compelling stories, but too often spurious. I have become more doubting about narrative over the years.
So if anyone tries to pin Washington’s recent improvement on Evgeny Kuznetsov’s entry into the player-assistance program, I’m calling bullshit.
Kuznetsov played his last game at the end of January. The Capitals played their best month of the season in February. I can understand the attraction to make a sweeping and maybe cynical conclusion, but I beg you to resist it. The chart below shows how much of the on-ice events Washington controlled during five-on-five play over the season. Above 50 percent means better than even.
The Capitals hit rock bottom almost two weeks before Kuznetsov left. They were on a dramatic upswing before the all-star break. Subtracting one troubled player from the lineup doesn’t explain Washington’s improvement. If anything, I’m tempted to think Oshie’s return from injury on January 11, Sgarbossa’s addition on February 6, and the increasing chemistry of the McMichael line were bigger drivers. But I don’t know, and that’s the important thing. The snapshot series starts from the numbers and then tries to go a level deeper — but never without keeping in mind the limits of our knowledge, which I try to harp on as much as I can. This is just an intellectual exercise, so ultimately it has to defer to the human element. I hope Evgeny Kuznetsov gets well soon, because I like him as a player and person, and I’d like to see him happy again. That’s what matters most.
Let’s do the snapshot. The tables below show the team’s performance when each player is on the ice during five-on-five play. I’ve highlighted stats of note, and we’ll discuss those below.
Forwards
| Player | TOI | SA% | xGF% | GF% | PDO |
|---|---|---|---|---|---|
| Sgarbossa | 128 | 55.0 | 56.3 | 50.6 | 0.99 |
| Mantha | 657 | 49.9 | 53.4 | 58.0 | 1.03 |
| Protas | 664 | 49.2 | 51.1 | 50.5 | 1.00 |
| Strome | 798 | 49.1 | 50.1 | 41.1 | 0.97 |
| Wilson | 759 | 48.8 | 45.6 | 31.1 | 0.95 |
| McMichael | 694 | 48.2 | 49.8 | 39.6 | 0.97 |
| Ovechkin | 764 | 48.1 | 44.4 | 34.0 | 0.95 |
| Pacioretty | 272 | 47.9 | 47.1 | 41.1 | 0.97 |
| Lapierre | 255 | 47.0 | 40.6 | 45.8 | 1.01 |
| Milano | 304 | 46.4 | 49.4 | 52.2 | 1.03 |
| Oshie | 478 | 46.3 | 45.6 | 35.2 | 0.97 |
| Aube-Kubel | 523 | 43.5 | 40.7 | 62.2 | 1.07 |
| Phillips | 249 | 42.8 | 51.3 | 44.6 | 1.00 |
| Kuznetsov | 563 | 42.3 | 38.8 | 34.3 | 0.98 |
| Dowd | 557 | 41.2 | 41.3 | 60.0 | 1.05 |
| Malenstyn | 639 | 40.0 | 39.2 | 54.6 | 1.05 |
Defenders
| Player | TOI | SA% | xGF% | GF% | PDO |
|---|---|---|---|---|---|
| Edmundson | 615 | 47.9 | 46.5 | 46.2 | 1.00 |
| Fehervary | 763 | 46.6 | 47.3 | 41.1 | 0.98 |
| Carlson | 1044 | 46.5 | 46.5 | 42.8 | 0.99 |
| Sandin | 944 | 46.3 | 46.2 | 45.3 | 1.00 |
| van Riemsdyk | 770 | 46.1 | 48.5 | 43.7 | 0.99 |
| Jensen | 895 | 46.0 | 44.9 | 50.1 | 1.02 |
| Bear | 286 | 45.2 | 38.1 | 35.1 | 0.95 |
| Alexeyev | 198 | 42.0 | 41.2 | 34.2 | 0.95 |
Glossary
- TOI – Time on ice in minutes. Only five-on-five play is included here.
- SA% – Shot-attempt percentage. The share of total shots attempted by Washington while the skater is on the ice during five-on-five play. 50 percent means even.
- xGF% – Expected goals percentage. The share of expected goals generated by Washington while the skate is on the ice during five-on-five play. Expected goals weighs how likely to become a goal each attempted shot is. 50 percent means even.
- GF% – Goals-for percentage. The share of total goals scored by Washington when the skater is on the ice during five-on-five play. 50 percent means even.
- PDO – The sum of Washington’s shooting percentage and saving percentage when the skater is on the ice during five-on-five play. One means league average. The acronym doesn’t stand for anything, and yes, I hate it.
Notes
- I love Mike Sgarbossa. After a few years of lean and unreliable forward depth, the Caps finally have a dependable veteran to call up. The team controls 55.0 percent of the shot attempts (SA%) and 56.3 percent of the expected goals (xGF%) when he’s on the ice, and I feel like the puck is safe every time he’s near it.
- Tom Wilson had a bad February, but it was almost all the fault of PDO, a measurement of finishing by summing the Caps’ shooting and saving percentages. Opponents outscored the Caps 10 to 3 during Wilson’s shifts last month, but that is way out of proportion with the team’s control of underlying events: 50.2 percent of attempts and 46.8 percent of expected goals. The results (goals) don’t follow the process (shots and shot quality), so the x-factor is finishing. Washington goalies saved a lowly .865 in that time, and Wilson’s on-ice partners converted only 4.9 percent of their shots-on-goal. So I’m not worried.
- Meanwhile, the line of Aliaksei Protas, Anthony Mantha, and Connor McMichael was superb last month. They outscored opponents 4 to 2, and they dominated the flow of play with 59.6% of the attempts and 68.8% of the expected goals. The McMichael line does well something that no other Caps line does: generate dangerous chances on the rush. That’s where Mantha is most dangerous, and that’s what he had been missing in his first few seasons here. McMichael doesn’t have the individual offense I originally thought he’d have, but he’s developed a talent for getting out the Caps zone and into the opponent zone cleanly. I can’t wonder how much more effective they’d be if they had a great puck-moving defender behind them.
- Because, yeah, the defensive roster remains the team’s biggest problem. Nick Jensen is somehow the only defender in the black in goals, and he’s just even. John Carlson’s got an on-ice goal differential of minus-11 (42.8 percent GF%; four of those opponent goals coming in the cursed Detroit game), and the depth options — Bear and Alexeyev — are barely even viable at the NHL level. I think Alexeyev’s usage is a bit too spare for me to have any strong opinions about, but Ethan Bear is getting favorable deployments (only Sandin has more offensive-zone starts) and doing not much with them. He’s fine in his own end, but his contribution to offense is really lacking. The visualization below, from HockeyViz, shows where the Caps are shooting when Bear’s on the ice. Blue blobs mean fewer shots from that spot; red means more. The minus-22 means the Caps’ offense is more than one-fifth off the league average. We know the team has a productivity problem in general, and reinforcements like Bear only hurt the cause.
- Meanwhile, with the Capitals racking up injuries on the blue line, John Carlson has been asked to do way too much. Playing 26:42 per night in February, he’s in the league’s top five in ice time, behind stars of various degrees like Rasmus Dahlin, Seth Jones, and Charlie McAvoy, and just above Carlson’s west-coast mirror-universe twin, Drew Doughty. Both 34 years old, Doughty does defense as well as Carlson does offense, and both defenders are kind of maddening in their tradeoffs.
- Alex Ovechkin finally, FREAKING FINALLY, broke out in February. He had 7 goals and 5 assists in 11 games. There’s two sides to this, but I’ll start with the fun part. I reject any notion that Ovechkin’s production is because he’s shooting more or better. His individual shot-attempt rate (the gray dashed line) barely ticked up since the all-star break, but his goal rate (the red line) exploded. I like to think that’s the universe paying him back for miserable shooting luck in December. In the snapshot, we still ask how many, but we also ask how, and sometimes the answer is Chaos.
- But Ovechkin’s individual scoring has coincided with bad stuff at the other end of the ice. The Caps controlled 43.6 percent of the expected goals during Ovi’s five-on-five shifts in February, better than only Carlson and Oshie, and the team got outscored 11 to 6. If we’re looking at Washington’s recent improvement in underlying stats, it has come apart from Ovechkin’s ice time. If we’re looking at Washington’s recent improvement in the standings, it has depended more on Ovechkin’s special-teams play.
- The fourth line of Beck Malenstyn, Nic Dowd, and Nicolas Aube-Kubel has started less than six percent of its five-on-five shifts in the offensive zone. That is by far the most defensive deployment in the league, and consequentially it colors the numbers for those players, which are bad. Malenstyn is worst, exactly at the tank line of 40 percent in on-ice shot-attempt percentage (SA%), but I’ve already written enough about him. Nicolas Aube-Kubel is the interesting guy right now. He’s strong in his own end (HockeyViz reckons he’s elite there) but he’s not a traditionally strong finisher, until he busts out a solo effort like the one below from the Tampa game. NAK pressures the puck loose, banks a pass to himself, settles it with his skate, evades a defender, then elevates a backhander to beat Vasilevskiy.
- For an unrestricted free agent on an expiring contract who had been on waivers in October, Aube-Kubel all of a sudden looks like a guy who would shore up the forward depth of any playoff contender. The Caps have a long and attractive stack of trade chips, and I think he should be near the top. Next to — sorry, I’m saying it: Charlie Lindgren.
This story would not be possible without Natural Stat Trick, All Three Zones, Evolving Hockey, and Hockey Viz. Please consider joining us in supporting them.
