The Canucks powerplay: not broken

In his post-game presser, John Tortorella answered a question about the powerplay the same way I might have:

“Yeah, it could have helped us tonight. It wasn’t good tonight. Other night’s it’s been good, but tonight it wasn’t good.”

It may have drawn a bit of surprise from some commentators. Canucks powerplay… good? But they’re clicking at just 9.3% on the season! They were 0-for-3 against Detroit and have scored just 4 powerplay goals on the season. That doesn’t seem particularly good. The problem is that “good” too often conflates with “results” in the narrow minds of broadcasters and fans that focus all too often on what happens in a small sample. Leading up to the game, Tortorella made some points about the shots and chances the Canucks have been getting with the man advantage. He says some interesting things and it’s worth a read.

The Canucks powerplay success rate was pretty brutal in the 48-game season, and it’s been pretty brutal this season, but I’d say that the two powerplays are by no means similar. I can’t speak directly to Xs and Os because that’s not exactly what I do, but the entries appear cleaner and the Canucks are generating a tonne more shots:

Here are the 5v4 shots per 60 minutes and goals rates, dating back to the beginning of 2008, from Hockey Analysis:

  5v4 Shots 5v4 Goals
2008 50.4 (8th) 5.85 (15th)
2009 50.0 (12th) 5.60 (24th)
2010 53.1 (9th) 7.41 (5th)
2011 55.9 (5th) 9.20 (1st)
2012 55.4 (3rd) 6.99 (4th)
2013 43.7 (22nd) 5.33 (20th)
2014 63.3 (4th) 3.53 (26th)

There are two takeaways from this:

  1. The first is that in every year the Canucks have had a declining Shots per 60 rate, they’ve also had a declining goals per 60. The only (completed) year that somewhat bucks the trend is if you look at the team rankings from 2011 to 2011, when they jumped from 5th to 3rd in shots, but declined from 1st to 4th in goals. Fairly noisy, but if you looked at only the rates and not the ranks, you wouldn’t be surprised by the drop from one year to the next.
  2. Second takeaway is that the team is usually pretty close in shots rate and goals rate, as far as ranking goes. The biggest gap over an 82-game season is 12 rungs in the latter, while they’re 22 so far in the early going. While it’s very possible that the shots rate slinks back closer to the the high 50s, I envision no scenario where the Canucks finish the season shooting just 5.6% on the powerplay.

Just to further the point about shot rates and goal rates, if you look at a three-year sample between 2010 and 2013, the difference in shooting percentage was about 6 percentage points between Edmonton (14.65%) and Columbus (8.71%). If you look at the 48-game sample, the difference is about twice times as large. Washington scored on 20.5% of shots last season, while Carolina was just 9.48%.

(This leads to the question… how the heck did Columbus have such terrible shooting numbers between 2010 and 2013?)

The lesson isn’t “shoot a lot, and you’ll have success”. There’s some happy balance between taking shots from anywhere, and actually having a high shooting rate. It’s easy to dismiss that Canuck shots haven’t been of quality and they need to be better there, but I don’t think that you can work with such a high powerplay shot rate without getting some very quality looks. If all you did was take shots from the point, there’s a better chance of those shots being blocked, or missing the net, or easily recovered by the defence or goalie.

The lesson is “if you’re getting a lot of chances, you’re probably taking a lot of shots”. Vancouver’s favourite Oilers blogger mc79hockey had a fun post about this and thinks there’s some indication that teams are increasing their focus on shot-generation. Tortorella’s quote makes me think that the Canucks aren’t going to try too many different things on the powerplay, since it really has been working. It has the skeletal structure of some of the better powerplays we’ve seen in the NHL, and there’s no need to change the approach.

I mentioned in the post-game that the Canucks took just two shots in 4:54 of 5v4 time, which works out to 24.5 shots per 60 minutes, well down from how the team was doing going into the night. It’s why I agreed with what he said “wasn’t good tonight, has been good other nights” was Tortorella’s angry, monosyllabic way of essentially saying “trust the process”, one of coach Alain Vigneault’s favourite tropes.

It was apparent that last season the Canucks had a stagnant powerplay, and I don’t think it’s fair to compare to the two, particularly since they’ve replaced the coaching staff. That, on its own, hasn’t led to a super successful powerplay, and neither has letting Jason Garrison fire bombs from the point all night like people were demanding last season, but it’s clear this powerplay is headed in the right direction, game against Detroit aside.

The game against Toronto Saturday should be a fun test. The Leafs are essentially the opposite in the early going. On the penalty kill, the team has given up a tonne of shots, but few goals relative to those shots. Interesting test from both clubs.

  • Peachy

    “While it’s very possible that the shots rate slinks back closer to the the high 50s”

    3 of the last 180 team seasons have had a SF60 above 63.3 at 5v4.

    8 of the last 180 team seasons have had a SF60 above 60.0 at 5v4

    Only San Jose (65.5) and Anaheim (57.1) have a SF60 above 52.8 at 5v4 in the last 3 seasons.

    It’s a lot more than “very possible” that the shot rate at 5v4 will begin to fall…

    • Peachy

      I suppose the question is how much it will regress and whether or not it’s coming at the expense of shot quality, ie will shooting percentage regress back to 14%? If the system is fundamentally different this year, the two stats may not regress as far as we think.

      The other interesting thing is that PP shot rates are up across the league.

      MC79 offers some insights.

      • Squibbles

        Considering how infrequently teams have maintained a SF60 above 60.0 at 5v4, I’ll take the under on the Canucks.

        While Dellows’ article is interesting, it’s really only something to keep an eye on at this point.

        It’s a lot easier for a baserunner to limit the stolen base attempts than for a power play unit to shoot more while maintaining shooting percentage, I’d argue.

        If it was as simple as “shoot more!” everyone would do it.

        And perhaps many teams are doing it and the new elite level baseline for SF60 at 5v4 will go up as a result…

        Of course, the relative value of the Canucks employing this strategy may be nullified.

        The Canucks have played 10 of 15 (66.7%) games against the Eastern conference.

        That’s the highest percentage amongst the big 5 of the Pacific division.

        By the end of the year it will be 30 of 82 (36.6%).

        I’m skeptical about a lot of what we’ve seen so far based on the competition level…

    • Squibbles

      Yeah, I read that as ironic understatement, seeing as how that would fit in with the numbers displayed immediately above the sentence.

      That said, I’m not quite certain what your point is, since (again, referencing the chart) the Canucks have themselves been above 52.8 in two of the past 3 years. As the one year that they did not represents only 58.5% of a regular season, I’d be skeptical that the 8 under for that season outweighs the 3-4 over in the previous seasons. Assuming roughly equal opportunities per 60 minutes, my rough guess would be the Canucks should average about 53-53.5 SF60 over the last 3 seasons.

      Did you mean that only SJS & ANA have been above 52.8 at 5v4 in each of the last 3 seasons? If so, that really doesn’t disagree with anything Cam said.

      • pheenster

        Considering what teams have been doing in the recent past, the shot rate is almost certainly unsustainable.

        It’s no different than some of the crazy high (and crazy low) save percentages we’ve seen in the early part of this season and every season since the beginning of time.

        And while this is more of a response to Tyler’s article than to Cam’s, the focus on shot-generation stuff is pretty simplistic.

        Because the flip side, assuming shooting percentage doesn’t suffer as a result, is that teams are going to all of a sudden have lower penalty killing percentages.

        Because I’m sure all it takes for all 30 teams to improve PP percentages is to shoot more…

        There’s nothing groundbreaking about any of this.

        Also, there is no such thing as the “Canucks Powerplay”.

        It’s a construct. There is player and coaching fluidity.

        The Sharks, for example, can’t just stick 5 of their replacement players on the ice and maintain their PP dominance…

  • Peachy

    The Columbus data point is interesting and perplexing. They’re so far away from anybody else that it’s incredible. And their data should probably be thrown out. Difference between 1 and 29 is ~4%.

    Question: were the Canucks to regress to their 3 year average shooting percentage (yikes, second in the league at 14%, go Sedins), where would their goals per 60 land?

  • Squibbles

    I wonder if not having a practice for 10 days in a row has contributed to poor powerplay performance? It seems like one of the things that practicing as a team together could really improve as opposed to 5-on-5 play that is more dynamic and based on on-the-fly decisions?

    • Squibbles

      I think so. It’s pretty obvious from the shots vs goals, or just watching, that we’re lacking the finish and it’s something that is obviously declining. We’re not putting pucks INTO the net. A lot of it is timing, like a getting it off the stick a split second earlier. Getting better wood on a one timer, hitting and picking spots. Things which practice sorts out. We didn’t have a lot practice time last year and I think some of the kinks in our game, like the PP, couldn’t get worked out because of it.

      This year, I just think we’ve been more focused on getting the system right. Now that we do somewhat I can see our PP getting more focus. No brainer really since Torts mentioned it sucked. We’ve seen results from most everything he’s done, so I’ll get ballsy and say our PP takes off in the next few games. There’s too much there for it not to.

  • Squibbles

    You are what your record is, and that PP rank is terrible. Last night the Canucks had about 12 shots through two periods? Torts better do something to get these guys scoring tough goals because bad luck becomes the norm very quickly. Too many guys are missing the net and if memory serves, morethan a few cheapies went in the net on the roadie because they just shot the puck! Shoot the puck and scrum!!

  • Squibbles

    Hey Cammy, still trying to delusde yourself into thinking the Cansucks aren’t a sucky team eh? Tell you what, your stupid charts and retarded graphs aint gonna do any damn good come this post season, where you Canucks will choke yet again, and it will be the 36th season of futility again. After you and 5minutesintheass and spleensteer wank yourselves off, you guys can jack off to your sedin sisters poster, one thinga for sure, wanker, I’ll be back this post season to remind you retards bout your Canucks failure, and there aint nohing you can do about it. Bottom line, Seins = 2 girls, no cup. Bahahahahahahahaahaahaha RETARDS! LMFAO

  • pheenster

    Is there a way to break down the shot rates and scoring rates per 60 minutes for each PP unit? For example, how is PP unit #1 (with Kes, Sedins, Garrison) doing compared to the 2nd unit. The eye test seems to show that the 2nd unit has been dreadful due to lack of talent – but looked immediately better with Burrows and Edler on the 2nd unit.

    The 1-3-1 formation that is all the rage in the league now (as seems to work very effectively) might work better if Garrison was the only dman at the top with Kesler, Burrows and the twins?