Ekblad vs. Klingberg: Derailing the Hype Train

Last week I wrote a piece about NY Rangers’ defenseman Dan Boyle and why I think he isn’t as good as some fans make him out to be. This time, we’re looking at the Calder Trophy race in the NHL, which goes to the best rookie in the league. The three finalists were Aaron Ekblad, Johnny Gaudreau (yay New Jersey product!), and Mark Stone.

Ekblad is a bit of a prodigy; he was the #1 overall pick last year and started 81 games for the Florida Panthers as an 18 year old. In writing this, I don’t mean to take anything away from him. He was a phenomenal rookie and he has an extremely promising career ahead of him. I also firmly believe that John Klingberg (age 22) of the Dallas Stars had a better season than Ekblad, but didn’t end up a finalist for the Calder due in large part to Ekblad’s hype as an 18 year old phenom.

So what are the arguments in Ekblad’s favor over Klingberg, taking age out of the equation? First off, we have that Ekblad ranked 29th in the league among defensemen in CF%, while Klingberg was slightly lower at 33rd. Another is that Ekblad’s CF% Relative to that of his team (CFRel%) is higher than Klingberg’s, 3.76 vs 1.84, respectively.

There are issues with both of these statistics, however. For starters, those CF% rankings are misleading. While Ekblad is ranked above Klingberg, the actual differences are negligible — Ekblad’s 53.85% is virtually identical to Klingberg’s 53.55%. This means that both players effected possession while on the ice roughly equally.

The other statistic is CFRel%, which is the difference between the player’s CF% and that of his team. The higher the number, the better, and yes, it can be negative. Ekblad’s CFRel% is higher than Klingberg’s, which could suggest that he was the better player. However, since both of their percentages were roughly equal, what this actually tells us is that Florida was a better possession team than Dallas. Now, one can certainly make the argument that being an equal player on a worse team is impressive, and it is, but I don’t believe it is enough to put Ekblad over Klingberg.

So why is it that I believe Klingberg was the better player this season? The first stat I want to look at is a simple one, which is points. While a defenseman’s job isn’t necessarily to score points, it is often a large part of what makes the great defensemen great. A defenseman who is good at both ends, after all, is more favorable who is good on one end. Looking at 5 on 5 production (since this makes up the majority of hockey, and is often the more repeatable production), Klingberg put up 24 points in 65 games, while Ekblad managed 22 points in 81. More points in less games is great, but we can truly see this effect when we look at the points per 60 minutes of ice time, which puts them on more level ground. In this regard, Klingberg tallied 1.30 points per 60 minutes of 5 on 5 ice time, while Ekblad had an even 1.00.

Again though, these are defensemen, and points aren’t everything. Let’s look at their usages. I made a point last week that a defenseman who starts less frequently in the offensive zone/more frequently in the defensive zone, is often considered favorable than the inverse. A player like that typically faces tougher assignments and is less likely to have shots taken by his own team when on the ice. Klingberg’s offensive zone to defensive zone start ratio is 343:374, or an offensive zone start percentage of 47.8%. More than half the time he gets on the ice for a faceoff, it is either in the neutral zone or his own end. Ekblad’s ratio is 522:338, or 60.7%. Not only is that a much higher percentage than that of Klingberg, but it’s as high as 30th in the league among players with at least 500 minutes of ice time. Again, that is 30th among players, not just defensemen, but all players with at least 500 minutes, which counts 563 players. For comparison’s sake, Klingberg ranks 359th on this same list.

Let’s take a look now at their level of competition. According to War-On-Ice, where all these stats are taken from, Klingberg’s TOIC% (competition-weighted time on ice) is 17.6% to Ekblad’s 17.3%. Looking at the same stat, but for teammates rather than competition, I’m seeing 16.8% for Klingberg and 16.9% for Ekblad. Clearly, both players face a similar level of competition, with a similar quality of teammates on the ice with them. Pairing this statistic with the fact that, despite starting in his own defensive zone much more often than Ekblad does, Klingberg’s CF% is roughly identical tells me that Klingberg was a more reliable player.

Now, I don’t mean to make this look entirely one-sided. Ekblad was very good this year, especially for a rookie and even more special due to his age, since defensemen tend to take longer to develop than forwards. He also posted similar offensive numbers while on a much lower shooting percentage, 5.2% compared to Klingberg’s 11.9%. You’d think that this is in Klingberg’s favor, but an 11.9% shooting percentage is very high and is very likely to come back down to Earth next season. For those who believe that individual players have a significant effect on their goalie’s save percentage (I’m still not convinced), Ekblad’s on ice save percentage was 0.930 vs Klingberg’s 0.921. Ekblad’s 100.8 PDO tells us that his numbers should be about right, meaning he wasn’t a product of tremendous luck either good or bad. Klingberg, on the other hand, had a PDO of 101.3, which tells us that his production might have been somewhat lucky. For those who don’t know, PDO is the sum of a player’s on-ice shooting percentage and on-ice save percentage. It is thought that, over time with large sample sizes, those two will typically add up to 100, give or take a bit. The further a player is from that number, the luckier/unluckier they may be. Numbers far exceeding 100 are thought to be lucky, while numbers far below 100 are thought to be unlucky. Klingberg’s on-ice shooting percentage was pretty high and is likely to regress next season, so we’ll see if he can keep up his production.

So while it’s certainly not the case that every statistic favors Klingberg, I believe that enough of the important ones favor him to the point where I think he had the better season last year, and it should be Klingberg as a finalist for the Calder over Ekblad. I think Klingberg was the less “sexy” pick because Ekblad was the #1 overall pick next year and his performance at his age makes him a potential generational talent. They both have bright futures ahead of them, and I look forward to watching them develop over the years.