How do the Canucks’ forward lines compare to the rest of the NHL?

Thanks to a litany of injuries (and, apparently, dehydration), the Canucks’ forward lines have been a complete mess recently. During one game, last season’s second line of Ryan Kesler, David Booth, and Chris Higgins was out, and Keith Ballard has now played three games as a forward on the third line. After passing up Jussi Jokinen on waivers, the Canucks are still relying on Andrew Ebbett to centre their third line until Kesler returns.

Despite all this, the Canucks are on a five-game winning streak, largely thanks to some stellar goaltending by Cory Schneider, as well as secondary scoring from the second and third lines. While some fans have complained about the Canucks’ depth at forward, it strikes me that their forward depth is actually pretty strong for the team to be missing so many players and still ice a lineup capable of winning games.

So how do the Canucks’ lines compare with the rest of the NHL? Despite missing Ryan Kesler, very favourably.

On Tuesday, Tyler Dellow published a post on placing the Corsi statistic in its proper context. He did this by sorting all of the forwards from last season into the equivalent of first, second, third, and fourth lines by ice time, then comparing their Corsi rating.

He did this because Corsi can frequently be abused. Simply saying that one player is good because he has a high Corsi number and another is bad because he has a low Corsi number is far too simplistic.

The starting point for me, when I’m looking at Corsi data, is always remembering that there are different expectations depending on where you are in the lineup. No hockey team is made up entirely of Datsyuks, puck possession wizards who crush the opposition in terms of shots when they’re on the ice. Generally speaking, as you go further down the lineup, the players get weaker in terms of their ability to gain and keep possession of the puck.

As a reminder, Corsi is basically an extension of the plus/minus statistic that uses a larger sample size by counting all shots — on goal, missed, or blocked — for or against when a player is on the ice. It has been found to correlate strongly with scoring chances. Basically, a good team that creates scoring chances and scores goals is going to simultaneously create a more shots than their opposition. Teams that are trying to get high-quality shots will also get a high-quantity of shots.

Here is the result of Dellow’s breakdown, which you can click to embiggen:

Dellow used Corsi as measured by percentage: above 50% indicates that when that player was on the ice, his team took more shots than the opposition. These results shouldn’t come as a surprise: first line players are the best at puck possession, then second line players, and so on. In general, this means that NHL coaches have a pretty decent idea of who their best players are and give them the appropriate amount of ice time.

Dellow took this a step further by splitting each of those groups of players into 5 groups to show the range of Corsi ratings and how that turns into goal differential. Once again, click to embiggen:

I find the goal differential column on the far right particularly interesting. Dellow set the lowest rated group from each line to a baseline of zero goals, which shows the difference in goals you can expect for each group. Essentially, if your first line is composed of players from the top-rated group of first line forwards, that works out to about 20-21 more goals for your team over an 82-game season than if it was composted of players from the bottom-rated group.

So, when you’re constructing an ideal team that is still realistic, you want each of your lines to be composed of players from the top-rated group for each line.

How do the Canucks fit into this scheme? Let’s take a look:

ES Ice Time is the determinant factor in figuring out which line each player is on, so these lines don’t precisely line up with what we might consider to be the Canucks’ actual lines. The players are listed by ice time. I also limited it to at least 10 games played, which eliminates Ryan Kesler, Manny Malhotra, Steve Pinizzotto, and Andrew Gordon, but accommodates Andrew Ebbett and David Booth.

It’s also important to keep in mind that this doesn’t take usage into account, such as zone starts and quality of competition. Both Chris Higgins and Jannik Hansen spent some time on a checking line, starting shifts mainly in the defensive zone and the same is true for Dale Weise and Maxim Lapierre. Lapierre has started just 22.5% of his shifts in the offensive zone as compared to the defensive zone, so his low Corsi rating shouldn’t be a surprise.

Still, this gives us some insight into how the Canucks’ lineup stacks up across the league.

The Canucks’ first line, unsurprisingly, is right at the top of the charts and is actually well above the average for the top-20 first-line players in the NHL. Their execution hasn’t been perfect recently, with Daniel in particular seeming to second-guess himself when he gets scoring chances, but their puck possession game is still fantastic and the Sedins frequently spend long shifts in the offensive zone.

On the second line, things get a little dicey. Higgins has struggled without Kesler and Booth. Placing him with the Sedins has seemed to spark his game a bit, but he seems to be showing that he’s more of a complementary player instead of someone who can drive puck possession on his own.

Hansen is sitting right at league average for a second-line player, but his usage has been so different from the start of the season to where he is now that it’s hard to judge if that’s accurate. Now that he’s playing on a scoring line with Raymond, he’s been very good from a puck possession standpoint, whereas earlier in the season he got soundly thumped while playing a more defensive role.

Raymond and Booth, on the other hand are both above average, with Booth posting closer to first-line Corsi numbers.

It’s easy to see why Jordan Schroeder earned his way on to the second line, as he is well above average for a third-line player, as is Zack Kassian. Both of them are right at the top-ranking for third-line players. Andrew Ebbett, meanwhile, is right at the average, while Lapierre is well below, sitting between the fourth and fifth rankings among third-liners, though his extreme zone starts should be taken into account.

Surprisingly, Tom Sestito is right at league-average when it comes to fourth liners. He’s been solid so far for the Canucks and far better than I expected. Weise, however, is below-average, though he was used in a similar way to Lapierre earlier in the season and had a positive Corsi at that time.

With Schroeder on the second line, the Canucks’ top two lines are above the league average, with the third line about average and the fourth line a little below average. That’s nothing a new third-line centre (or the return of a very good second-line center) won’t fix.

Tags:

16 comments

  1. peanutflower
    March 27, 2013

    I don’t know what to make of all this, let alone what to make of the word “embiggen”

    VA:F [1.9.16_1159]
    Rating: +9 (from 11 votes)
    • JDM
      March 27, 2013

      “A noble spirit embiggens the smallest man”. Someone doesn’t watch enough Simpsons. Or the rest of us watch too much. One or the other.

      As for the charts, I have to nitpick the Kassian one as it seems to me that he would have been boosted by playing 1st and 2nd line role for a while earlier in the season. Burrows has been on and off that top line and therefore has had his numbers depressed as a result.

      I continue to be amazed at how ridiculous Booth is at the possession game. Damn the hockey gods for stealing our AMEX away.

      VA:F [1.9.16_1159]
      Rating: +6 (from 6 votes)
    • ZeroTenancity
      March 27, 2013

      Embiggen is a perfectly cromulent word.

      VA:F [1.9.16_1159]
      Rating: +17 (from 17 votes)
    • Chris the Curmudgeon
      March 27, 2013

      Embiggens is a perfectly cromulent word.

      VA:F [1.9.16_1159]
      Rating: +4 (from 4 votes)
      • Chris the Curmudgeon
        March 27, 2013

        Ugh, beaten to the punch.

        VA:F [1.9.16_1159]
        Rating: +2 (from 2 votes)
  2. Chris the Curmudgeon
    March 27, 2013

    The problem with all of this analysis is that nowhere is there a statistic that adjusts for goals scored. A scoring chance is only as good as the player getting it’s ability to capitalize. Certainly, it has the positive aspect that a scoring chance at one end probably means the puck is at that end, however, goals are the statistic that ultimately dictate wins/losses. It seems like just about every IWTG spends time praising the team’s ability to generate high possession stats, however I feel like I read that here regardless of whether we’ve won or lost. The team may generate plenty of scoring chances, but they are severely lacking in finishing ability, all evening out to them being an average team offensively.

    Also, I find passing up the chance to grab a talented player like Jussi Jokinen to be just another sign of gross mismanagement, as the team is absolutely itching for a guy like him who has demonstrated scoring ability, is good in the circle (58.2% on the year) and would’ve cost the team nothing by way of draft picks or players in return.

    VA:F [1.9.16_1159]
    Rating: +2 (from 8 votes)
    • Origamirock
      March 27, 2013

      Except for the fact that the Canucks currently have 51 contracts and would have to lose one (or two) of Raymond, Lapierre, Higgins or Schroeder to pay Jokinen’s salary next year. So he wouldn’t have exactly cost “nothing.”

      VA:F [1.9.16_1159]
      Rating: +3 (from 3 votes)
      • Chris the Curmudgeon
        March 27, 2013

        If he doesn’t fit no need to keep him. I’m sure you could get something for him in the off season, or could just send him down as Carolina is now doing.

        Figuring that Ballard and possibly Booth will get bought out in the summer, and Luongo is probably going to be traded, I think the team should be able to make it work.

        VA:F [1.9.16_1159]
        Rating: 0 (from 0 votes)
    • Daniel Wagner
      March 27, 2013

      There’s no statistic there that adjusts for goals scored because it’s been shown statistically, time and again, that puck possession statistics like Corsi correlate with scoring chances and, well, goals scored. It may take time for it all to even out, but it does even out eventually. But let’s say that they were finishing all their chances right now, like they were during the 2010-11 regular season when they led the league in goals/game. They went from 3.15 goals/game during the regular season to 2.32 goals/game in the playoffs and only scored 8 goals in 7 games in the Finals. What happened to all those finishers in the playoffs or were they just suddenly not on the team anymore?

      Hockey fans need to understand that a large chunk of this game comes down to good and bad bounces. The best a team can do is consistently outplay their opposition, get as many scoring chances as possible, and do their best to finish their chances. Look at Daniel Sedin this season. His career shooting percentage is 12.4%. This season, it’s 9.7%. Has he become a worse finisher or is he just not getting the bounces? Mason Raymond’s career shooting percentage is 9.5%. This season, it’s 13.8%. Has he magically become a better shooter? Same with Chris Higgins, whose career shooting percentage is 10.5% but is at 14.6% this season. There’s your guy with elite finishing ability right there! Or, maybe, he’s gotten some good bounces this season.

      Heck, Tom Sestito’s shooting percentage is 25% this season. He may only have 8 shots, but he makes them count. Get him with the Sedins so they finally have someone to finish off all those great scoring chances they’re creating.

      What is it you’re looking for when it comes to finishing ability? You want the Canucks to score goals so they can win future games, right? Well, with sample sizes this small, the best predictor of future goals and wins is a team’s Corsi when the score is tied. It’s only over larger sample sizes that goals scored becomes a useful predictor of future goals. The Canucks, being one of the best puck possession teams in the NHL with the score tied, will probably score a bunch more goals in the future.

      I complete agree with you about Jokinen, however. I really wish Gillis had picked him up and I’m hoping that the only reason he didn’t is because he has something else in the works.

      VN:F [1.9.16_1159]
      Rating: +5 (from 9 votes)
      • Chris the Curmudgeon
        March 27, 2013

        I think the point was most relevant to the case of David Booth. While the guy generates a ton of scoring chances, his shooting percentage for his career is chronically low. I’d argue that either means he takes shots from everywhere (partially true) or that he lacks a sniper’s touch. Certainly some players have been unlucky while others have been lucky when you look at the bounces and the small sample size, but that’s the reason why a team needs to build scoring depth. Chicago had Kane and Toews and Sharp, but they didn’t just say “oh ok, well if our star players go on a slump, they’ll regress back to the mean at some point and meanwhile we have plenty of grinders that can pitch in when necessary”. They paid top dollar to add the best offensive player available at the time, Marian Hossa, knowing that the best bulwark against the ups and downs is to stack your lineup with guys with scoring talent, and when guys aren’t knocking him out with cheapshots, he’s paid off for them. When you look at the lineup Gillis has built, almost every player that he’s added (rather than inherited) is a two-way or gritty guy, but basically none are elite scorers, and some have come at the expense of guys with offensive ability. As a result, when the bounces don’t go the Sedins/Burrows way, or when Kesler is out with one of his many injuries, the team ices a lineup that pushes the tempo just fine, but doesn’t put the puck in the net. That might suffice against mediocre teams, but there are some NHL teams that are solid possession-wise and have a lot of players that can score. Detroit comes to mind as a good example, and they’ve pummeled us this year.

        I’d say that the power play problems stem from the lack of a power play QB type defenceman who can play on the right side. Also Gillis’s fault.

        I’d hope as well that he has something under his sleeve, but I’m also sort of dreading it considering his recent performance on the trade market.

        VA:F [1.9.16_1159]
        Rating: +2 (from 4 votes)
        • nanodummy
          March 28, 2013

          Would you sign Richards for 6.7? Semin for 7? Oooh, ooh, Kovalchuk for 6.6!

          Who is this magic free agent we should have signed?

          I seem to recall that Kassian and Booth (who at the time didn’t have his PDO problem, check out Drance’s article at nuck’s army) are both goal scoring bangers, you know, guys who will probably be 20 goal scorers with Burrows, Raymond, Kesler and Hansen for the next 4 or 5 years… Not that they’ll all be canucks, but…

          Oh right, we also signed one of the league leading goal scorering defensemen of the NHL in Jason Garrison, who also has evidence of being a pretty good defender at times.

          But shoot, we should have signed Semin for 7.5 million instead and had Alberts and Barker as our bottom pairing with Ballard and Edler playing soft parade. That would have worked out.

          Man, where is your sense of perspective? Elite scorers don’t grow on trees and we sadly don’t have the prospects to buy one on the trade market. Richards and Carter might have been the only deals we didn’t capitalize on in the last 2 years, and Carter clearly was gunning to go play with his buddy Richards. As for the Richards trade, we don’t have a Schenn to trade with Mason Raymond, and we already had a Kesler and a Malholtra at the time so why are we making that trade? We made a run at Doan, we made a run at Weber, Parise and Suter knew where they were going months before the FA season. Seriously, who is it we should have signed?

          This year, Iginla was never coming here, a divisional rival, and what other elite scoring talent is getting moved do you think? Vanek? St. Louis? Lecalvalier? Hemsky? That’s about it for bottom dwellers who have assets who aren’t prospects. Maybe Stevie Y. takes a bite at Luongo (he probably should entertain that idea, but he fired his coach so who knows) and we can rent St. Louis, but that’s probably where the gravy train ends, sadly…

          You can’t criticize GMMG for trades and signings that don’t exist.

          VA:F [1.9.16_1159]
          Rating: +6 (from 8 votes)
      • ikillchicken
        March 27, 2013

        That seems most plausible to me. There are some knocks against Jokinen but he’s still just what the Canucks need: A faceoff winning, penalty killing, defensive center with some scoring upside for the third line. And his contract is hardly an issue. With one year left at 3M we could easily move him, waive him, or even buy him out if he really doesn’t work out. There’s too little risk to not be worth the gamble on a useful piece…unless Gillis wants to save a roster spot/cap space for something else he has in the works. Somebody better…like Steve Ott.

        VA:F [1.9.16_1159]
        Rating: +2 (from 2 votes)
    • Daniel Wagner
      March 27, 2013

      Here’s another way to look at it. So far this season, the Canucks have 65 goals at even-strength, which is tied for 7th in the NHL. The only reason they haven’t been scoring as many goals as previous years is that their powerplay has been non-existent. Is that because they don’t have players who can finish on the powerplay or is it a combination of not taking enough shots and not getting the bounces?

      VN:F [1.9.16_1159]
      Rating: +1 (from 5 votes)
      • Lenny
        March 29, 2013

        Who knows? Corsi only tracks even strength numbers.

        VA:F [1.9.16_1159]
        Rating: 0 (from 0 votes)
        • Daniel Wagner
          March 29, 2013

          Well, you can get Corsi numbers for special teams as well, but they’re a bit more difficult to parse. I’ve been thinking a lot about how to approach advanced statistics with special teams and I’ve been leaning away from Corsi, as it includes blocked shots, which I consider to be a positive on the penalty kill. Too many blocked shots at even strength means you’re not controlling the puck well, but when you’re on the penalty kill, you assume that the other team will have possession of the puck, so restricting shot quality by blocking shots and forcing missed shots seems to be the right way to go. So, Corsi isn’t really the right statistic to use for special teams.

          But my point above is that the Canucks are actually one of the better offensive teams in the league at even-strength. They are finishing their chances and scoring goals. So the perception that the Canucks can’t finish their chances and need to trade for a sniper is misguided. If the Canucks were converting on the powerplay at even a league-average pace, their goals/game would actually be near the top of the league.

          VN:F [1.9.16_1159]
          Rating: +1 (from 1 vote)
          • Lenny
            March 29, 2013

            Yes. I like the team this year for its even strength numbers and added toughness. Our team +/- is actually very good but our goals differential is a dismal +6. Even the Leafs is at +8. When in full strength, I think we are a more playoff ready team than the one 2 years ago.

            Let’s just hope go deep enough in the playoffs for Booth to come back.

            VA:F [1.9.16_1159]
            Rating: +1 (from 1 vote)