#ShadowOvechkin.  #ShadowOvechkin when he's on the Capitals' power play, I mean specifically. It's something I've tweeted many, many, many times. It even lends itself to a character-saving play on words. I've been in reasonably-civil arguments online about it with people who make decent points. But so far, I remain steadfast in my belief: When on a 4v5 PK against the Caps, the penalty-killing team should have a F glue himself to Ovechkin to limit his shots from the Ovi Spot. Last night, the NYR let him fire away from that spot, and it ended up leading to the first goal of the night (originally credited to Ovi, but later changed to Justin Williams). 

Prior to "Saving the Avalanche", Arik Parnass briefly looked at teams that tried to #ShadowOvechkin. His article was short, and not particularly conclusive (though it's worth a read if only for the Rangers joke). The good folks at JapersRink also touched on the role of Ovechkin on the Capitals' PP here and here though without focusing on #ShadowOvechkin as a tactic specifically. 

And let's get this clear up front: I fully understand that the Caps' PP is going to get looks, whether you shadow Ovi or not. Even if you managed to physically lock Ovi's arms behind his back (perhaps by putting his version of kyrptonite - an Olympic gold medal - on him),  (1) he will still manage to get shots off somehow, and (2) you are probably opening up other chances for Oshie, Backstrom, Shattenkirk, etc.

But, in a game like hockey, where every goal matters, and special teams can be the difference, I think you are better off taking your chances that one of those other guys beats you, than letting Ovi have free reign to fire away from the faceoff dot.

I wish I had time to really look at video and try to develop some strategies, but I don't. In the meantime, here are some stats regarding Ovechkin and the power play that I think lend support to my idea.

data from espn.com

Over the course of his career, Ovechkin has become much more of a "Power Play Specialist."  I think Capitals fans would point to the introduction of a 1-3-1 PP format under Adam Oates in the lockout-shortened 2012/13 season as the key change.  As this chart shows, for the last 5 seasons, Ovechkin's ratio of "Even Strength" to "Power Play" goals has been more of a 50/50 split than earlier in his career.  [Note: These are all ES/PP, not just 5v5/5v4]. Putting this another way, before the lockout, about 1 in 3 Ovechkin goals was a PPG.  Today, if Oveckhin scores a goal, it's almost as likely to be a PPG as an ESG.  This season, if it doesnt change in the last games, will be the first time in Ovi's career he's scored more PPG than ESG in a season.


data from Corsica.Hockey

Note the shift from "ES/PP" to "5v5/5v4" here, since I'm now looking at Corsica data. Also note that the x-axis starts at 2007/08, since that's as far back as Corsica data goes. I would guess that the earlier years would be similar to the early years shown, but I'm not sure it would matter much either way.

Ovechkin scores almost 50% of his goals on the PP, but he only gets about 20% of his total ice time on the PP. To be clear, I wouldn't expect it to be exactly the same - PP shots are likely going to be better chances to score on average than ES shots. But looking at the trend, where his ratio of ES/PP time is basically consistent throughout his career, and then looking at this ratio of ESG/PPG from the previous chart, you again see that his PPG production has become much more influential to his total goal production.


data from Corsica.Hockey

Looking at some key metrics (individual Corsi for, individual Shots for, and individual expected goals for), we again see that trend. The blue line at the top is the percentage of total 5v5+5v4 time that Ovechkin spends at 5v5: This season its 79% (this is a carryover from the previous chart). I've then included the percentage of total iCF, iSF, and ixGF that happened at 5v5. So in 2016/17, 66% of Ovechkin's total iSF came at 5v5, and 64% of both his iCF and ixGF came at 5v5. These numbers have remained relatively consistent since the 2012/13 lockout season.

I think it's interesting (in part because it supports my beliefs ;) to see his ixGF fluctuates even more - especially since it was almost 10% points lower until this year.  Seeing individual xGF for lower than iCF/iSF makes sense to me - players on the PP are probably going to get better "quality" than at ES.  I don't know whether this year will be the new trend or not but I'd bet its the exception to the more recent rule.  

But even for his direct shot metrics, they are about 5 percentage points lower than his pre-lockout trends, which again shows that Ovechkin is consistently doing more of his damage on the PP these days. 

Also: yeah yeah, I know I shortened the y-axis range to make this easier to read. 

data from Corsica.Hockey

This is that same information, but at 5v4 (so the previous chart flipped). In sum, Ovechkin spends 21% of this time on 5v4, but gets about 35% of his production at 5v4. Again, I'm not expecting it to be exactly the same - but instead to show that as his career has gone on, that ratio has widened. 

So what difference does it make?

[update: 4.6.17, 5:05 pm]

How lethal is Ovechkin on the PP? Here's a list of the top 25 PPG scorers since the 2012/13 lockout. Oveckin has a 37-goal lead over 2nd place. That's the same difference as 2nd place has over *50th* place. Whatever NHL teams are doing to defend Ovechkin on the PK? It's not working.

data from hockeyreference.com

#ShadowOveckin isn't just an easy snarky comment. Since the 2012/13 lockout, Ovechkin's goal scoring and individual shot production have become more reliant on the freedom that being on the PP (and more specifically 5v4) provides. Maybe you will never stop that. Maybe the Caps' PP unit is just that good at moving the PK around to get Ovi that open Ovi Spot shot. And maybe if you did #ShadowOvechkin, you'd just get killed by Oshie and the others. But if was a coach, and had only this information available to me, I'd be looking to find ways to minimize letting the greatest goal scorer of (maybe) all time do this:

source: hockeyviz.com



Again I'm right in my analysis.