People like to point out that James van Riemsdyk is not a very good defensive winger.  This is true; he isn’t.  Lately, though, this take has mutated into the opinion that he is so bad defensively that it cancels out all of his many offensive gifts.

This is demonstrably not true.

JvR Is A Very Good Player

Let’s start with this screenshot:

Under Mike Babcock, van Riemsdyk has been a positive presence at 5v5 in any complete metric you care to name.  This is compounded by the fact that he demonstrably is closely involved in the offence whenever he’s on the ice, whereas people seem very keen to attribute a lot of defensive responsibility on a bad defensive team to a LW.  Since Babcock came, among players with min. 30 GP, JvR has the second P1/60 on the team (behind only Auston Matthews.)  That’s a major contribution.

This gets even odder when you factor in his power play scoring.  People are prone to dismiss powerplay scoring as something basically anyone can do, as if goals on the power play don’t count.  In powerplay P1/60 since the start of 2015-16, van Riemsdyk is first on the Leafs—ahead of even Nylander.  Actually, in that span, he’s fourth in the entire NHL, amongst players who have had at least 100 PP minutes.

Further to that: people have attempted to make robust all-in-one metrics, such as Goals Above Replacement, that weight the relative importance of different numbers.  JvR, as per these metrics, is indeed not good at even-strength defence. He’s not even average at it. But his offence is so good he was rated as the best player on the team last season anyway.

He is not a great defensive winger.  He is still one of the best offensive wingers in hockey, and his offence outweighs his defensive flaws considerably.

This is a point that we have to hit home. You cannot reasonably expect someone to come in and replicate this performance. And yet, regularly, on Twitter and elsewhere, we see people arguing that the Leafs can replace him with no trouble. This isn’t just a random set of fans saying this, either. This is the Athletic’s James Mirtle, in reference to a (protected) tweet of a roster that doesn’t include van Riemsdyk, commenting on the potential issues losing him brings.

With all due respect to Mirtle here, this is batshit.  It feels like a bunch of people are trying to be sophisticated hockey hipsters and claim that the high-scoring winger gives it all back on the other end. Except it’s demonstrably not true.

Look at that first screenshot again. Positive Corsi. Positive Corsi Rel. Positive xG. Positive xG Rel. Positive goal differential. Unless you want to include Carlyle years (where the entire team looks like dogshit, unless your name is Jake or Nazem), JvR is quite clearly a net benefit to a team. And even under Carlyle, he only had one year with a negative Corsi Rel (2014/2015).

Now, the immediate counter is, “but he’s sheltered!”. This is vastly overstated. It’s easy to point at a player’s zone starts and bleat when it’s greater than 50%. Smart people like Micah McCurdy have looked into this - turns out, they mostly don’t matter. Only ~5% of players have their zone-adjusted CF% change by more than 1%. As far as we can tell, JvR has never been one of them. His usage is simply not extreme enough to matter, certainly not in the same stratosphere as what players like Artemi Panarin have received.

What about competition? Much has been made of the ‘third’ line of Van Riemsdyk, Bozak, and Marner feasting on depth players. Again, this is overstated.

He does, indeed, play less top-end forwards than the average player. You know who else did that last year? Auston Matthews.

Matthews probably played harder competition, don’t get us wrong. But it’s not by a huge amount.

None of this is to say that the Leafs should pay the exorbitant contract it would likely take to keep him. That’s a separate question, and one that has multiple reasonable answers.  However, in evaluating that we should go in with an accurate conception of what he actually is.

As far as we’re concerned, there is only one reasonable answer to the question of whether van Riemsdyk is a net positive. That answer is yes, and literally all the evidence that we have in the last two years supports it. It’s crazy when people argue that he is so bad defensively that he cancels out his offensive gifts, and it’s even crazier to argue that Kasperi Kapanen or Josh Leivo can replace that production without a hitch.

Leivo and Kapanen

Let’s start by acknowledging that both Josh Leivo and Kasperi Kapanen are good and interesting young players.  Each of them could be playing regular minutes for an NHL team with less wing depth than Toronto, and both of them have shown encouraging signs of potential.  It’s also possible that the Leafs will have to accept the downgrade from van Riemsdyk to one of these players as the price of the upcoming cap crunch.

But again, people go overboard and insist that you can replace him with either of them and experience no overall loss, or even an improvement.  This is nuts.

Starting with Josh Leivo, the man who cannot escape the pressbox: Josh Leivo has shown signs of being a decent powerplay presence.  He’s not especially quick and his defensive reputation has never been great (albeit improving), but in what little NHL time he’s gotten over the past two years (14 games), he’s looked quite good and been highly productive on a rate basis.  If you take his numbers in those 14 games at face value, Josh Leivo is, in fact, an elite offensive player.

The thing is, we like Leivo and think he’s done enough to earn some ice time.  But you have to contextualize his numbers a bit.  He went just short of point per game in the AHL at 22, or a little better than Connor Brown’s partial season at 21.  Should we expect that player to score at James van Riemsdyk’s rate?

Christian Roatis’ NHLe calculator, as a starting point, projects Leivo would produce at about a 36-point rate in the NHL (funnily enough, Brown’s exact point total in his rookie year.)  Josh Leivo’s career points per game over all his NHL stints are 0.45, or a 37-point pace.

Yes, his P/60 over these last 14 games has been spectacular.  But is it realistic to expect a player with Josh Leivo’s track record, who is struggling to crack the Leaf lineup, to replace a player who’s been at a 55-point pace or better for his last four seasons, and who routinely produces 27-30 goals?

None of this is to rain on Josh Leivo, who is doing everything he can and who can likely be a decent NHL player.  A guy who can go 20G-20A-40P—which seems like the high end for Leivo—is generally useful.  But that doesn’t put him in JvR class, unless that player also has very remarkable defensive value.  Leivo, with all due respect, doesn’t.

Let’s move on to Kasperi Kapanen. He’s a good prospect. 20-year-olds who excel in the AHL to the degree that he did last year are not dime a dozen, as much as Craig Button may think they are. He is also not remotely likely to be as good as JvR offensively.

At 21, van Riemsdyk was in the midst of putting up a 20-goal, 40-point season in the NHL in just 14 minutes a night for Philadelphia. NHLe projects Kapanen for about 38 points in the NHL, if we use his numbers for last year. So in that sense, he can be reasonably expected to deliver somewhat similar offense as van Riemsdyk did at 21 (though it’s obviously better to actually put up those numbers than have a model say you could - of course, it’s not really Kapanen’s fault he isn’t getting those opportunities).

Unfortunately, we’re not replacing van Riemsdyk at 21; we’re replacing him at 28, who is significantly better. And it is not reasonable to expect Kapanen to do that. It’s easy to say that Kapanen’s defensive value will let him surpass JvR’s overall value, despite lesser offense. It is much harder to prove that. In his very limited NHL time, there isn’t much evidence that Kapanen is much of a play driver. He can play on the PK, but that doesn’t tell us a lot about his even strength defense. The desire to see him as some sort of defensive stopper seems more fanciful than grounded in reality. He will likely be better than JvR on defense. But there’s still almost surely a huge gulf offensively.

Summing It Up

It’s easy to take players that have been around a while for granted. You become inured to their strengths, and start to see their flaws. This is especially true when their flaws are so readily apparent, as they are with van Riemsdyk. For exactly that reason, many Leafs fans still think Jake Gardiner is a horrific defender, even though all the evidence points to Gardiner being a net positive in spite of his occasional brain farts.

The same is true for van Riemsdyk. It’s fair to say that he’s bad defensively. It’s unpleasant when you watch him miss a defensive coverage, or when he’s on for another goal against—and not coincidentally, he’s been burned badly for on-ice save percentage this season.  Even people who try to ignore +/- and the eye test can’t help getting frustrated when he gets scored on.  We all feel that way.

It’s when people say that his defence negates his truly elite offence that they leave the world of fact and enter a world of fiction. As we’ve learned more about hockey, it’s become clear that there are empty-calories scorers who give all their offense back through anemic defense - Thomas Vanek is the quintessential example.

From every angle we can look at, James van Riemsdyk isn’t one of them, and it’s bad analysis to say he is.

Analysis