The Toronto Maple Leafs' woes on the penalty kill over the past few years has been a frequent topic of discussion here at Pension Plan Puppets and virtually everywhere else in Leaf fandom since Ron Wilson became head coach of the team, and probably before then too. While many ideas have been floated about what the reason(s) may be, I decided that I was going to try to collect a bunch of hard data and see if I could put together some kind of well-supported explanation as to what's been going wrong. The results took the form of a comparison between two teams that seemed like they should have had similar PK results, but didn't - the Leafs and the Minnesota Wild.
I decided to divide this into two parts to keep it readable. In Part 1, which I wrote about a week and a half ago, I took a look at the goalies to see if there were any noteworthy differences between the two teams in net. The conclusion that I came to was that the Leafs goalies seemed to be somewhat to blame for the team's poor results. In Part 2, I'm going to be looking at what the other 4 skaters on the PK are doing and see if my initial conclusion about the goalies holds up. If you haven't yet read Part 1, I'd definitely recommend it, since it will provide important context to Part 2 such as "What would drive a person to spend their free time watching video of old Minnesota Wild penalty kills?" Follow me past the jump, where I finally reach a solid conclusion about The Problem With The Leafs PK.
In Part 2 of What's The Problem With The Leafs PK I decided to watch video of every single PK goal that either the Leafs or the Wild allowed this season and see if I could discern anything about the strategies that they use while killing penalties that might cause either team to be more or less likely to allow powerplay goals against. To try to organise this info in something resembling an objective manner, I recorded whether each goal was caused by a specific kind of defensive lapse or breakdown. Here's a breakdown of those categories
Forward(s) in Behind the Defence - I created this category mostly to keep track of goals allowed on account of fronting. If you're not familiar with fronting, it's a technique where defencemen stand in front of the forwards in front of the net in an attempt to block shots. The downside to this technique is that it leaves forwards wide open for tip-ins and rebounds. I also included in this category any goal that was scored by a forward getting in behind the defence in some other manner; for example, if the defence skated too slowly and a guy skated unobstructed to the net on a rush.
Goalie Screened - any goal where either the defence or an opposing forward blocked the goalie's view of the shot that went in, but didn't actually redirect the puck.
Rebound - any goal scored on a rebound when the defence were not fronting.
Odd-man rush - in practice this turned out to be a category to record breakaway goals. I had intended to include 2-on-1s and such, but none actually showed up in the video.
Uncovered Defenceman - a goal in which a defenceman scored directly on a shot where there was no screen and no forward got out to the point to attempt to get in the shooting lane.
Other - any other goal. A couple of these were goalie miscues, some were just really good shots, etc. Basically anything I couldn't pin on some sort of clear defensive error.
Here are the results in table format:
|Goalie Screened||F behind D||Uncovered D||Rebound||Odd-Man Rush||Other|
A useful way to summarise the data is this - Toronto has allowed 21 goals that I was able to pin on a defensive lapse of some sort and 11 that I couldn't, a ratio of 65.6%. Minnesota has allowed 11 goals that I was able to pin on a defensive lapse and 12 that I couldn't, a ratio of 47.8%, significantly lower.
However, I think this overstates the degree to which rebound goals are actually the defence's fault; they're often the result of scrambles where the defence is outnumbered, and there's a great deal of luck involved when six or seven players are all wildly hacking at the puck. If we take out the rebound goals, Toronto's ratio of "bad"-to-"good" goals stays basically the same, while the Wild's falls to just 33.3%, a huge difference.
I think it's clear that there are some pretty big PK systems issues at play here. Toronto's goalie is more frequently screened, has to deal with more unobstructed traffic, and is left out to defend more breakaways. I recognise that there is a bit of selection bias here in that I was watching plays where goals were scored, but I tried to address issues such as shot quantity and location in Part 1. If anyone wants to spend time necessary to chart the level of goalie-screenedness and so forth for each individual shot, I'd be perfectly happy to re-examine my conclusions (I estimate that would take maybe 15 hours or so). Based on the video I watched (and I did watch some entire PKs to get an idea of each team's technique), I think that the results here pretty closely match how each team plays the PK.
What are the differences between each team's actual style, beyond the results? Toronto's defencemen front more or less all the time. There is virtually never a Toronto defender tying up a forward in front of the net on the PK, and the strong side forward will often skate well out from the net to try to block shots. Minnesota does front on occasion, but their defenders generally play back closer to the net and they're far more engaged with the other team's forwards out front. Toronto's defenders also seem to be much more likely to chase the puck into the corners and behind the net. I don't know if that's a systems issue or just some bad decision making. In addition, Toronto's defenders are much more aggressive than Minnesota's at the blue line, which likely explains why the Leafs have allowed more breakaway goals while killing penalties. Minnesota's forwards are also far more aggressive in preventing shots from the point. Toronto's forwards collapse deep into the zone on the PK, frequently leaving both pointmen unguarded. Minnesota's forwards rarely collapse as deeply into the zone, and are generally very active in preventing defenders from having the time or space necessary to get good shots off.
Having watched all of these goals, I'm prepared to revise my tentative conclusion from Part 1. I don't think goaltending is a very good explanation for the Leafs' penalty kill problems this year. While examining the goals the Leafs have allowed on the PK this year, I found it hard to fault the goalies on more than 1 or 2 goals. I would say that Minnesota fared worse in that regard, though I didn't track it closely. I think the ultimate explanation for the Leafs' horrid PK results really is coaching. Specifically, I think the problem is fronting. It has been a factor on most of the goals the Leafs have allowed on the PK this season.
It will not make Leafs fans any happier to hear that the Leafs are still employing the same penalty killing strategies that have created so many problems for them this year. I think the success of the PK in January is a mirage, a combination of the fact that the Leafs had to kill so few penalties while mostly playing teams with poor powerplays, with a little bit of luck thrown in. I expect to see the Leafs continue giving up PP goals at a high rate once they begin playing better teams, especially if they start taking penalties more frequently.
* * *
If all you wanted to know was What's Wrong With The Leafs PK?, then you can probably stop reading here. However, having looked at all of the information I've collected, I've got one additional conclusion to draw that will likely be of interest to those of you who are into the Fancy Stats. I think what I've gathered here is proof of shot quality.
We know from Part 1 that both teams give up shots on the penalty kill at roughly the same rate (Minnesota gives up slightly more than Toronto). We also know from Part 1 that both teams face shots from the same locations at roughly the same rate (with Toronto once again looking slightly better). However, the Leafs clearly allow goals on the penalty kill at a far more frequent rate than Minnesota does. And after watching the tape to see why that is, it seems pretty clear to me that there are significant structural differences in the way that each team kills penalties that results in Toronto giving up more goals. Perhaps an easier way to say that is this - a shot against the Leafs from a given area on the ice is far more likely to be a good scoring opportunity than a shot against the Wild. While both teams give up a similar rate of shots from similar areas, a shot faced by a Leafs' goaltender on the PK is frequently much more difficult to stop than a shot against the Wild from the same spot. And if I'm not mistaken, that's shot quality.