Yesterday, the Toronto Maple Leafs made a signing that delighted many and terrified just as many. There’s really no middle ground when it comes to Roman Polak, he’s a player that you respect the hustle of when you watch him, you cringe when you open up the spreadsheets, and depending on which of those two elements mean more to how you evaluate the team, you usually end up with a decisive conclusion.

From a “grab a beer and watch the game” perspective, Polak is legitimately one of my favourite Leafs of the last several years. I grew up on tough hockey, and while I often question the holier-than-thou attitude that it stems in the fan and casual culture in the game, there’s something to be said about a player who lays it all on the line whenever he plays. Polak is absolutely that; he’s not a great skater but he practically fights to the ice get where he need to go, he sticks up for his teammates, and he’ll sacrifice his body to make sure the puck stays out of the crease and out of the net.

The question, of course, is does it really work? Last year, it did and it didn’t. Toronto had a better goal differential with him on the ice at even strength (55.4%, or as traditionalists will point, out, +10), but nothing about the actual process seemed to align with that. The Leafs game up more shot attempts with him on the ice than any other defenceman. He was a bit better with unblocked attempts, suggesting he got in the way of a lot of what was coming his way but was still a net negative when factoring in his impact on offence. Expected Goals models all estimate him as having a negative impact in your typical season, but again, last year was atypical.

Polak’s biggest benefit was the obscene save percentage that Frederik Andersen and Curtis McElhinney were on the ice with him. With Polak on, they had a 5-on-5 save percentage of 0.943, which is the fifth-most fortunate run of goaltending that any regular defenceman in hockey had last year (Shea Weber, Andrei Markov, Brooks Orpik, and Karl Alzner, who all played in front of Carey Price or Braden Holtby, were above him). That “prowess” continued while shorthanded, as the 0.890 they had there placed Polak 8th of 26 penalty killers to play 200 minutes.

Alas, even a player who keeps players out of the “danger areas”, or take away “good shots”, won’t have a permanent impact on raw goals or save percentage. The goaltender’s decisions still matters the most on a save, so any impact that a defenceman will have on the quality of shots on goal will be margin-of-error tight, not the difference between making his goalie an ECHLer or a Vezina winner. We have this argument about a handful of “grittier” defencemen and forwards every year, and almost every time, the player is incapable of defeating random distribution and ends up with a pedestrian or inverse-outlier number. The aforementioned Alzner is a good example; he’s sitting at a cool 0.847 at 5v5 to start the year. Or even look further back on Polak career to date; in his last 10 seasons, he’s been above the team average on-ice save percentage three times, and below it seven times.

Of course, it’s incredibly difficult to turn around to a fanbase or even a coaching staff and say “hey, we know you like what you see of this guy, and it’s working on the scoresheet right now, but it won’t last”. Sometimes, a player runs hot and you don’t want to move on from them until you’ve got confirmation that the potion is, indeed, not magic, and that rings especially true when it’s someone you want to root for. Every single one of us is guilty of hyping up our favourites and calling out our goats during their peaks and valleys, and if Polak is still in the midst of a peak, it’s not shocking that he’s going to be the safe-bet, devil-you-know option when a problem you believe he can solve a problem of yours.

Right now, the Maple Leafs, who are mostly flying high with a 6-2-0 record and +9 goal differential, have two particular issues that seem somewhat more pressing than the others. Those are:

  • Frederik Andersen’s sudden inability to stop hockey pucks and
  • Finding a balance on the penalty kill

It’s easy to draw a line between Polak and each of these issues, and why they’d presume he could fix them. Andersen stopped a lot of pucks with Polak on the ice last year, so maybe he helps him this year too. The penalty kill featured a lot of Polak this year, so it presumably could again and do fine. A few things to keep in mind, though:

Not Your Save-iour

Stats from Corsica, Home Plate from War On Ice, Heat Map From HockeyViz. tl;dr on the viz: Red means more shots against from that area than the average player, blue means less.

There’s not a lot of evidence that what Polak brings to the table will help Andersen out very much. Polak’s reputation is based on his ability to clear traffic in front of the net, something that’s evidenced by the fact that he’s always hitting and shoving guys there. Perception doesn’t match reality, though; Toronto actually gives up more shots in front of the crease, in the highest danger areas, when he’s on the ice. Not that high danger shots have even been Andersen’s problem this year; his numbers there are actually equal in all situations and better at even strength, only lowering on the powerplay (almost assuredly a sample size thing; high danger powerplay shots against are fancy talk for ‘shorthanded breakaways’), and a bit on the penalty kill. Andersen’s biggest drop-off at the moment is on low-danger shots, which could be a screen issue, or perhaps even just luck (again, this is a team making a decision based on eight games, three quarters of which they won).

Whatever it is, it’s not something that lines up with what we know of yesterday’s signing. The Senators game is a good example; there’s not a single goal against there that I think his skill set would’ve prevented, either because they were goal types that would’ve have exploited weaknesses, or because the current defencemen made decisions similar to those he’d make. If you expect Polak to make Andersen better, you’re basically just gambling on raw odds; which is also possible by sitting around and assuming that Freddy hasn’t suddenly turned into a beer leaguer and will sort himself out soon.

Still In Unfamiliar Territory

Here’s a secret: Despite all the pressing concern about the penalty kill, Toronto’s unit has done… better than last year so far?

Against (Per Hour) 2016/17 2017/18
Shot Attempts 109.8 (27th) 94.9 (7th)
Unblocked Attempts 81.2 (27th) 67.9 (4th)
Shots on Goal 56.0 (23rd) 49.6 (4th)
Goals Allowed 5.9 (9th) 5.4 (6th)
Expected Goals 7.7 (27th) 5.6 (6th)
Save Percentage 0.895 (3rd) 0.891 (12th)
Penalty Kill  % 82.5% (7th) 84.4% (7th)

Unlike last season, where Andersen carried the load and made up for a unit that worked hard but bled shots, Toronto’s penalty killers have had a lot of success this year, forcing opponents to refrain from shooting and keeping it out of the net once they have. The issue, though, is who has been sent over the bench. Not so much that they’re choosing the wrong players, but on defence, they’re essentially only running one pair:

Ron Hainsey would be Toronto’s third-most-used penalty killer, a minute shy of second if you only used the time he spent on the unit in the season opener against Winnipeg. He and Zaitsev haven’t just been Toronto’s top shorthanded pair, they’ve played 78% of their shorthanded minutes on defence. In the last four games, they’ve had a minimum of 85% of the allotment, including every single second against Washington.

That won’t last, and even if it was going to, it’ll gas these two out as the season progresses, which seems pretty suicidal. Toronto no doubt needs a second pair there to even everything out, and Polak presumably gets them a little closer. But this still begs the question of what exactly it is that’s preventing Toronto from using two of their other four defencemen more often on this unit, and how one player will be the solution if that just brings you to three trustworthy guys out of four?

My gut feeling here is that they’ll pair him with Andreas Borgman, who many hoped was going to be the player that replaced Polak in the sandpaper department, rather than becoming his second layer. That way, they can have a “heavy” pair that pushes around their opponent, which may or may not actually work. Borgman would presumably learn a thing or two as well, which is one of the reasons I’ve been let down to see him average only a minute a night on the unit so far; if this is his long-term role, it seems to me to get him out there while everyone is in their own early-season adjustment period.

It’s not a terrible combo, but I wonder if it would have made more sense to continue running with him and Morgan Rielly as the PK 3+4’s. The counter to that is that Rielly is playing more on the powerplay this year in Nikita Zaitsev’s place, but the double-counter would be to give those minutes to Connor Carrick, or go full modern era and run one of the powerplays as a five-forward unit. That way you have three of your top four defencemen playing PK minutes, without gassing anyone else out.

Better Kills Cost You More Kills

Let’s say you still firmly believe, though, that Polak is the right player to have out there with a man down. Hey; all power to you if you do; at this point, I can’t really change your mind if I haven’t already. But I’ll warn of a caveat: a lot of the time, he won’t be there to help you kill the penalty, whether you’d like him to be or not?

Why? Well, there’s a good chance he took it. Even as Toronto’s sixth-most played defenceman last year, and even without playing the first 10% of this year, no defenceman on the Leafs roster has taken more penalties since the start of 2016/17 season. If you’ll remember, I wrote a last week on penalty tendencies within the team. Using the same counting system, Polak has taken 24 penalties and drawn 8.

Taken Drawn
Interference 7 Fighting 2
Holding 3 Hooking 2
Slashing 3 Instigator 1
Boarding 2 Interference 1
Fighting 2 Tripping 1
Tripping 2 Roughing 1
Hooking 2
Puck Over Glass 1
Roughing 1
High Sticking 1

You’re effectively stuck adding 3-4 extra powerplay goals against over the course of the season just from him being in the box. That doesn’t seem like a lot, but if you only give up 40 while down a man over the course of the season, having ~10% of your unit’s “failure” coming from their stalwarts getting into his own trouble feels like a bit of a conflict. It’s also rather telling what types of penalties he gets involved in: half of what he draws are by-products of scrums, and most of what he takes are chasing / step-behind penalties (holding, tripping, hooking, boarding), net-front-commotion calls (slashing), or a combination of both (interference).

Polak’s aggression is a double-edged sword, to the point where you wonder if a less effective penalty killer that’s more disciplined would lead to fewer goals against at the end of the year. Toronto, like 26 other teams, are on pace to take more penalties (77 more, no less) than last year thanks to various league-wide crackdowns; adding a player that could make them one of the most penalized teams in the league, through his own decisions and people having to make up for the more deficient parts of his game, could its own problem down the stretch.

Depth Drama

One also has to consider the butterfly effect at play when talking about a move like this. “Who cares, he’s just a #6 defenceman” sounds great, but besides the fact that there’s no such thing as “just” anything in the top league in the world, where “being relied upon” as a defenceman means playing 40% of the game and “unimportant” still means playing over 25%, adding one more player to the bubble trickles a lot of situations down.

Now you wonder if Connor Carrick gets regular game action, despite the fact he’s been largely fine and plays a puck carrying/moving style that compliments the core group. You wonder how much ice Calle Rosen sees, and whether he becomes another Frank Corrado, Josh Leivo, Alexey Marchenko type situation where the player sits in purgatory last-push of development age. If they instead send Rosen to the Marlies, that’s another mess; that gives them nine defencemen who are either NHL-ready, capable quick call-ups, or within a year of being either of those things for most teams in the league. There are probably close to a dozen players who this “no big deal” signing immediately impacts on the depth chart, changing the course of the development to some extent.

The signing also puts the Leafs at 49 contracts, which means they’ll either be shedding someone soon or standing pat until something substantial comes around. In some ways, 49 is just as dangerous of a number as 50, in the sense that it makes you hold back on any move that doesn’t look like a grand slam.

That means something. You can argue how much, but it’s something.

So What Are The Leafs Gaining?

The Toronto Maple Leafs are gaining theoretical answers to issues that they’ve created for themselves, if they even exist at all there’s really no other way to say it.

After all, maybe the answer to “How do we fix Frederik Andersen’s save percentage” is “Didn’t we just wait it out last time and that worked?” or “Isn’t Carey Price an 0.881 right now?” Maybe the answer to “Who can we get to compliment Nikita Zaitsev and Ron Hainsey on the penalty kill” is “How does a team with Jake Gardiner and Morgan Rielly give up on using them after four games?” Maybe the answer to “How do we unleash Andreas Borgman’s edge?” is to play him more regularly and give him the rope to learn that you’d give a veteran to survive.

But ultimately, Polak is a safety net, he’s familiar to them, and it’s been a while since his results have caught up with his process. Ultimately, if you’re searching for a shortcut and it looks like it could come in a guy that you’re rooting for, a real soldier that’s worked hard for you in the past and worked harder to get healthy again, you’ll take the chance on it if the risk doesn’t handcuff you too hard.

That much I’ll give; he’s easy to root for, and it’d be awesome if he proved everyone wrong. It’s also low-enough cost contract to throw aside if things go awry. There really isn’t much risk, and it’s far from Armageddon.

But all the same, adding him doesn’t seem like an optimal solution to anything, his on-ice performance is likelier to make the group worse than it is to make them better, and when this team is leaping forward at light speed, it’s still a bit of a letdown to see them rely on old crutches. It’s one of those things where one shouldn’t be mad in a vacuum about the move itself, but skeptical of the reward and perhaps left disappointed in what it represents.

Thanks for reading! Hopefully you enjoyed this post. If you did, don't hesitate to share it on Twitter or Facebook; having more readers will help the site grow. As well, consider a subscription if you're interested in reading additional work that isn't available to guests.