Only two teams improved their shot-attempt differential from last season more than the Capitals: The Nashville Predators (boo) and the New York Islanders (hiss). With a 4.5 swing in score-adjusted possession, the trend has finally ended, and the Capitals are good again. Exactly how good– by the reckoning of the playoffs– remains to be seen, but looking back, I’m over the moon about this season.
But, in this week’s snapshot, the last of its kind, we ask, is the cake a lie?
Understanding 5v5 hockey using shot attempts starts with a number: 50 percent. That is even possession– one for the other team, one for your team, repeat. If your team is above 50 percent, you’ve either denied your opponent some attempts or you’ve managed to create a few extra of your own. Being “in the black,” with a possession number above 50 percent is a sign of an above-average team.
The Capitals are above 50 percent. They’re a 52-percent team– or just about. But I’m not so sure they’re truly or significantly above average. And it’s Buffalo’s fault, dammit.
If you ignore the Sabres (using some sloppy, back-of-napkin math), then the league’s average isn’t 50 percent– it’s more like 50.4 percent. And over on Puckon.net, the median teams have around 51.2-percent score-adjusted possession. It’s like grade inflation for hockey– making the Caps look just a bit better than they are because they, just like everyone else, got to beat up on the worst possession team of the modern era (and maybe longer; I’d love to know the 74-75 Caps’ shot-attempt differential.)
That has nothing to do with what’s in this week’s snapshot, but I thought it was curious.
What’s actually in this week’s snapshot: the Caps have really good young players, but are they fast enough for the Islanders?
For me, the snapshot has been about opening up the analytic process on a micro level– adding transparency and immediacy to my journey to understanding of how the Caps play every week. In the interest of furthering that transparency and because I can’t stand being insincere here, lemme say this: my enthusiasm is waning.
Part of that is me not having enough time to do the in-depth research and number-crunching, and part of it is acknowledging that the Capitals merely are what they are: A marginal playoff team with good special teams and one very special player.
However they looked in October and whatever hot streaks we’ve seen this season, the Capitals are not a championship team. (Or, if they are, we haven’t seen evidence of it lately.) They’re not bad like under Adam Oates (and for that I am grateful) but they’re not great. They’re just good. They’re a good team with a good coach.
Unless the bounces go bad or they draw a tough team, these Capitals should make it into the second round of the playoffs. No further.
In this week’s snapshot, which isn’t 36 hours late you’re just imagining it, it is what it is, but at least we’ve got Ovi.
There’s a stat called CHIP, as in salary Cap Hit of Injured Players. It measures the impact of injuries to a team based on how many games the players miss and how much they get paid. Up until recently, The Capitals had fared very, very well on the injury front this season. They had Dmitry Orlov and John Erskine missing from the blue line but were otherwise mostly unscathed. The Caps had one of the lowest CHIPs in the league.
Then March happened. Ovechkin missed a game, both Brookses are banged up, Peters got hurt, Latta is out, are and there’s a stomach bug going around. Just as the Capitals are mounting their final push for the playoffs, they’re all of a sudden a shambling mess.
Add to that the team’s performance since the all-star break (except for the week they roughed up the scrubs), and you’ve got some genuine worry about this team. They’re barely holding on to a playoff spot, and they’re one game away from squandering their longest home stand of the season.
It seems like everything’s broken for the Caps right now.
Okay, maybe you could say that a week of games against Buffalo, Toronto, and Columbus isn’t exactly indicative of the broader NHL’s competition level. But with only 15 games left in the regular season, every shot, goal, and win counts that much more– even if it just means the team is more confident as they face down some much tougher teams next week.
In this week’s snapshot, we look back with curiosity and forward with optimism.
I don’t know if what I’m doing with the snapshot is really “analytics.” Hearing all the buzz at the Sloan Conference and this mostly inane Deadspin piece, there’s a lot of stuff wrapped up in that term that don’t really apply here.
The snapshot isn’t about decision-making (we don’t make any decisions), and it’s definitely not a branding effort (it probably hurts the RMNB brand by being so stodgy). For me, these statistics are just new ways to understand the game.
My educational background is in literary criticism. In that field, people discuss writing using different frameworks (formalism, deconstruction, post-colonialism, queer, etc.). The goal isn’t to decide what writing is good or bad, but to appreciate the writing in new ways and learn more about it and ourselves by looking from different– and deeper– angles.
It’s not that much different for hockey. For some people, the only metric that matters is championships. It’s a simple binary: yes you won, or no you did not. Some go deeper: how far did you make it in the playoffs: zero rounds, one round, two rounds, or more (As a Caps fan, I suspect the “more” is a myth). More nuance, more understanding, but still nothing too deep. And then you can get down to wins. And then goals. And then shots. And then– and for some reason people resist this– shot attempts.
And there are layers of rich and complex data even further below, new angles from which to look. And when we acknowledge how that low-level information can bubble up to high-level results– like championships– we create an intellectual scaffold for richer understanding of the sport. It’s miles from the championship binary.
I see why some critics consider analytics to be a retreat from complexity: because it uses numbers, which are finite, instead of descriptions, which are not. It can seem reductive. But the spirit behind the analysis is quite the opposite: it’s a framework for looking deeper, and more closely– not to blithely draw conclusions.
In this week’s snapshot: No blithe conclusions. I’ll try.
Hockey legend Igor Larionov captured the hockey news cycle on Monday with a provocative article in the Players Tribune. Larionov’s thesis is that coaches are so stuck on hockey orthodoxy or so fearful of risk that they stifle and select against “creative” players.
The pullquote: “There’s a reason why Pavel Datsyuk went undrafted in 1996 and 1997.”
I’m skeptical about that example, but I think Larionov is spot-on about conservative coaching in general. We see it in all sports, but hockey seems to have a particularly pernicious strain of Goodoldboysclubitis, wherein exciting, finesse players are considered too “European” and risky, and safe and pedestrian players are wildly overvalued. I suspect that disease is behind symptoms like Top-line Beagle.
It’s a chronic infection that takes the form of valuing of a player’s characteristics over his actual effectiveness. Adding hard-working, “spark”-y Jay Beagle to the Capitals scoring line despite an embarrassingly convincing body of evidence that he actually hinders scoring– that’s an acute case.
If the point of hockey is to win games, why do so many people care more about a player’s description than a player’s production? The only answer is Goodoldboysclubitis.
We’ve seen itcreate necrotized flesh on the Philadelphia blue line after an injection of Andrew MacDonald, and in coming years we will see similar morbidity with the addition of Brooks Orpik in Washington.
“The effect is not going to be in goals and assists,” Brian MacLellan said in July. “It’s going to be in culture and winning and attitude, and that’s what Brooks Orpik does.”
That quote, likely uttered in the throes of a Goodoldboysclubitis fever, sums up the affliction perfectly. It’s like the brain is not able to separate the ways we talk about players (“gritty,” “hard-working,” “last name is also a breed of dog”) from the things those players do to actually help win games.
The teams who can rid themselves of the disease are the ones who do best in this league.
The only cure is information. Let’s do the snapshot.
There’s a great Mark Twain quote about weather in New England, but it sort of applies to how the Caps organize their forwards. If you don’t like the lines, just wait a few minutes.
Barry Trotz’s forward lines, which I’ve been tracking all season, get announced once or twice a day– at practice or morning skate and during warmups before games. And each time they’re followed by a chorus of criticism from professional and amateur hockey watchers, including me.
There’s always something to rail against: the guy on Alex Ovechkin’s opposite wing, who’s getting stuck on the fourth line, which pairings don’t work, who deserves a scratch but isn’t getting one, and who deserves a sweater but isn’t getting one. It’s instant fodder for content, fresh grist for the anguish mill, and an easy conversation starter.
But it’s also sort of cheap. Because there is no optimal line combination for Barry Trotz. If twenty of us were to make up our ideal lines, I doubt any two would match. There is no magic Rubik’s cube of forwards that make everyone love him and shut up. It just doesn’t exist. The lines are a loser every time, and Trotz, a coaching veteran with three decades of experience, knows it.
Every day he’s got to set back a rookie or piss off a forward. He’s got to give a sweater to a player he’d rather see traded, and maybe he’s thinking about how great some other player on some other team might be in that same spot. Even MacLellan has limited control over his roster considering the market forces and freak injuries that determine it.
That doesn’t mean we should hush up about the lines. I think we’ve got an exceptionally informed and passionate community here. We all know what it means for Evgeny Kuznetsov’s development when he takes shifts with Jason Chimera, and we should talk about it. (Though I bet Barry Trotz probably already knows as well.)
While it’s good for us to discuss and debate it, I’m going to always try to acknowledge that the lines will never be totally perfect and they’ll never be totally broken.
In this week’s snapshot, we explore the great space between those two extremes.
My educational background is in literature and my day job is in information architecture. In both pursuits I spend much of my time thinking about how people think. Understanding how they learn, what their different perspectives are, and what unconscious biases they may hold made me a better writer and a better software engineer.
Now that I’m doing the hockey thing, I see those same biases in play– especially concerning younger players. Most of the furor about the Erat-Forsberg trade was rooted in a (now verified) belief that Forsberg had the potential to be a star.
Were those angry people right or did they get lucky? When do our aspirations for a player hinder our ability to discuss him accurately? And what kind of new information is most likely to make us revisit an opinion of a player?
In this week’s snapshot, let’s talk about Evgeny Kuznetsov.
There’s this one peccadillo in hockey discourse that is bugging me more and more. I don’t know if anyone else experiences it this way; let me know.
If I say a player had a bad game, people might hear me say that the player is bad. If I say I’d rather someone not play on a top line, people might hear me say I don’t like that player. If I say a player’s contract is very bad, people might hear me say that player himself is very bad.
But I didn’t say that, and I definitely don’t think that.
I’m not sure how to do this more precisely in the future. How can we offer a criticism (or a joke) about Jason Chimera without committing character assassination? How can we separate discussions of Jay Beagle’s deployment and Brooks Orpik’s contract from descriptions of those player and those persons?
Because I like Jay Beagle and Jason Chimera and Brooks Orpik. A lot. They wear the right colors. They’re my guys.
In this week’s snapshot, we go a little bit deeper– because “this guy sucks” is never helpful to the discussion and almost never correct.