I'm going to write in defence of Capitals RW Alex Ovechkin, if only because there have been some recent attacks and, while I can certainly live with them being directed at Ovechkin, the analytical process ought to be better, particularly if it's going to be a hit piece on the league's leading goal-scorer.
Yes, I'm writing to defend a player that is on pace for a 59-goal season, when the second-best goal-scorer in the NHL this season, Toronto's Phil Kessel, is on pace for 43. Naturally, I didn't think that Ovechkin would be a player that needed much defending, but he's taken some hits lately.
On one side of the discussion, we have the Toronto Star, with their Department of Hockey Analytics, and while there are plenty of flaws in Ovechkin's game, they somehow determined that his goals-for/against percentage was the way to illustrate the problem. Never mind that goals for/against percentage is essentially measuring plus-minus.
I'll get to that in a moment but, first, also peruse the Hockey News piece by Ken Campbell who, after Ovechkin was minus-5 against Columbus, decided that Ovechkin has to decide what kind of player he is. After all, Ovechkin was minus-17 on the season after that game.
What's odd about using plus-minus to denigrate Ovechkin's contributions is that anyone doing serious analytical work in hockey has been against using plus-minus because it involves so many factors beyond an individual player's control (not least of which are the contributions of nine other skaters and two goaltenders when the game is 5-on-5) and, generally, involves small samples because goals are relatively rare events.
It's funny to find myself in this position, because I can be a bit of an apologist for plus-minus. You spend enough time around the game and that thinking can be pretty common, and when the sample is large enough, you can get a pretty decent list of players at both ends of the spectrum. (For example, here's the list of players with the best cumulative plus-minus since 2000, and here are the worst.)
But, I've at least learned that there are many other factors that go into whether a player is a plus or minus player, and they must be considered if you're going to attempt to pass judgment on a single season or, especially, a portion of a season.
So, let's take a look at some factors that are at play to make Alex Ovechkin a minus-17.
First off, the shooting percentage of others on the ice with Ovechkin at 5-on-5 is ridiculously low. His 6.3% is only ahead of fourth-liners Aaron Volpatti and Jay Beagle among Capitals forwards. The suggestion could be made -- and of course it has been -- that Ovechkin isn't making those around him better, but here are the 5-on-5 on-ice shooting percentages when Ovechkin has been on the ice for the past five seasons: 10.36%, 8.62%, 8.05%, 11.76%, 9.09%.
You're really going to have to dig for reasons, other than poor luck and ineffective shooters, to explain even-strength shooting effectiveness declining by 40% over last season, especially when Ovechkin himself is shooting 10.6% (18 goals, 170 shots) at 5-on-5.
Taking away Ovechkin's 18 goals on 170 shots, leaves the other Capitals to score eight goals on 242 shots (3.3%) with Ovechkin on the ice. Marcus Johansson, his most common left winger, has one goal on 51 shots. This undeniably effects plus-minus, right? Of course it does. Give Ovechkin an average on-ice shooting percentage (say, 8%) and that is a difference of about seven goals.
At the other end of the rink, Ovechkin is getting burned with a .909 save percentage at 5-on-5. Naturally, the argument will be that Ovechkin's defensive play is what leads to that low percentage. Keep in mind, that percentage is well below career norms for him (.922 over the past five seasons, including this one) and ranks near the bottom on the Capitals' roster. Use that typical save percentage, on 439 shots against, and that becomes a difference of 5-6 goals.
So, why not take a look at where the shots are coming from with Ovechkin on the ice?
According to Some Kind of Ninja's Shot Tracker, shots against the Capitals with Ovechkin on the ice at even strength, come from an average distance of 34.5 feet. In the previous five seasons, it has been 34.2, 36.6, 35.8, 34.6 and 35.9 feet on average. There can be an argument made, based on those average shot distances, that Ovechkin's most effective defensive performances were in 2010-2011 and 2011-2012 (the Dale Hunter season), but there really isn't a huge difference; goaltenders aren't suddenly flummoxed by 34-foot shots when 36-footers are easy pickings.
On top of that, best of luck trying to identify those particular seasons as anything close to Ovechkin's best. Why? Because he scored 70 goals and 150 points in 157 games over those two seasons, producing the two lowest point scoring rates of his career. Yes, even lower than this season, when virtually no one else puts the puck in the net with him on the ice.
Of course Ovechkin is not a defensive whiz, but that shouldn't stand as an indictment any more than it has for elite offensive players throughout the history of the game.
Seriously, take a look at the Top 10 goal-scorers in the league, none of whom put the puck in the net like Ovechkin, and identify the ones that are notably strong backcheckers. Some are: Joe Pavelski, Alexander Steen, Patrick Sharp, but there's no reason to believe that right wingers Kessel or Corey Perry or Patrick Kane are doing brilliant work in the defensive end. Here's the thing: they're all great players!
Part of the trouble for Ovechkin is that the Capitals haven't been able to win in the postseason, so he gets painted with the brush of failure for a whole host of team shortcomings. That's what comes with being a superstar.
This Capitals team is flawed. They rank in the bottom third of the league in Fenwick Close (measuring shot attempts, not including blocks, at even strength, with the score close), which is a good indication of team puck possession, yet Ovechkin has relatively solid possession numbers.
If you want to break down a player's overall contribution, and feel that you must use one statistic in order to do so (better yet, don't), then at least reduce the impact of others on the ice and look at the possession stats, because it won't matter that linemates aren't finishing or that, for whatever reason, goaltenders aren't stopping the puck.
Shooting and save percentages fluctuate and while they affect perception -- just ask Tyler Bozak -- they don't get to the bottom of a player's on-ice contribution, and so it is with Ovechkin this year, who is having a fine season, no matter what his plus-minus says.