Drance Numbers: Kesler and Booth have chemistry

David Booth’s knock-knock joke falls on deaf ears. #NoChemistry

Advanced statistics and quantitative analysis have consistently proven useful in hockey, but any honest hockey math nerd will admit that there are factors the numbers can’t quite measure. Some things operate on the U.S. Supreme Court’s “I know it when I see it” principle. Supposedly, one such unmeasurable factor is chemistry, which has been a major talking point among Canuck nation of late.

During the regular season, David Booth played roughly 35 total even-strength minutes with the Sedin twins. He played more with former Florida Panthers teammate Tomas Kopecky. Yet, with the team facing elimination in Game 4 of the Canucks’ first round series with the Los Angeles Kings, Alain Vigneault modified his lines, bumping Booth up to the top line to skate with the twins. That alteration to the team’s forward lines separated Booth from Ryan Kesler, his linemate all season.

Despite being somewhat bemused by Booth’s move to the Sedins’ right wing, many cheered the split from Kesler, as the two apparently have “no chemistry.” Oh, but they do.

Was Vigneault’s adjustment motivated by “chemistry”? Certainly Burrows (who replaced Booth on the second line) and Kesler are thought to “have it.” They’ve come up through the organization together, they’ve often been used together on the penalty kill, and they memorably teamed up to dominate the Chicago Blackhawks in Game 7 of last year’s Western Conference quarterfinal. But it’s unlike Vigneault to pursue something as fleeting as “chemistry” when he decides how to employ his lines. Think about it — if “chemistry” was a priority for him, he wouldn’t turn so ruthlessly or regularly to his Magic Bullet line blender.

When the playoffs began, Alain Vigneault was criticized by Canucks fans and media for not rolling the team’s best two-way trio, the so-called “Amex line” of Ryan Kesler, David Booth and Chris Higgins. When Vigneault adjusted in Game 2, and matched the “Amex” line against Anze Kopitar’s group (who had plastered the Pahlsson-centered checking line in Game 1), Vancouver’s embattled head coach was criticized for continuing to deploy Booth with Kesler, despite their apparent lack of chemistry. There are two lessons here: first, you can’t win in the Vancouver market no matter what you do if you’re Alain Vigneault, and second, chemistry matters, apparently.

Jeff Paterson pointed out that Ryan Kesler and David Booth didn’t pick up assists on one another’s goals with much regularity this season:

 

Considering the two skaters spent nearly 530 minutes together at even-strength, that’s a surprisingly dismal level of crossover. The usefulness of slicing up the data in this arbitrary way, however, falls apart under further scrutiny.

You can count the number of powerplay shifts that Ryan Kesler and David Booth played together this season on two hands; Kesler spent more time on the power-play with Mikael Samuelsson than he did with David Booth in 2011-12. I think it’s fair, then, to throw all of both players’ powerplay goals out the window (that’s 8 for Kesler, and 3 for Booth).

David Booth doesn’t play on the penalty kill either, so let’s defenestrate Kesler’s 1 shorthanded goal as well.

Furthermore, Booth joined the team in late October, and didn’t play his first game with the Canucks until October 25th. He was injured a few months later and missed six weeks of action. During that period of time, Kesler tallied 3 even-strength goals.

So Kesler scored 10 even-strength goals with Booth in the lineup, and Booth assisted on half of them. Meanwhile, Kesler recorded only 10 even-strength assists with David Booth in the lineup, and 40% of those were on goals scored by Booth.

If we look at it in terms of scoring rates, Kesler and Booth, when playing together, helped the Canucks manufacture 2.15 goals for every 60 minutes of even-strength ice-time they played. When Kesler played without Booth on his wing this season, the Canucks scored 2.25 goals for every 60 minutes of even-strength ice-time. The difference between those two rates is negligible. We can say with a good degree of confidence that Booth’s presence on his wing does nothing to neuter Kesler’s offensive efficiency.

Anyway, as you know, I don’t put much stock into production anyway; I care significantly more about controlling possession and generating shots and chances, and here’s where things get interesting. Let’s take a quick gander at Booth and Kesler’s WOWY (with or without you) numbers, courtesy David Johnston’s hockey analysis site:


Game State Corsi%
Kesler with Booth 60.4%
Kesler without Booth 51.3%
Booth Without Kesler 50.6%

I’d define “chemistry” between two players as, “when they play together, their line becomes more than the sum of their parts.” With Booth and Kesler, that has been the case this season, as they’ve controlled 60.4% of Corsi events. When they’ve skated together they’ve totally dominated play.

Let’s put their “chemistry” into context.

Perhaps no two players in the league have better “chemistry” than Daniel and Henrik Sedin. They’ve shared a womb, they’ve been teammates their entire lives, they were drafted back-to-back, and they’ve spent their entire NHL careers as linemates. They operate like homing beacons, always knowing where the other one is on the ice, always anticipating one another’s next move. It’s uncanny and magical to watch. I think most would agree that the chemistry between the two twins is a big part of what makes them such special players, and one of the most feared offensive duos in hockey.

Yet, when they’ve played together this season, the twins have controlled 59% of Corsi events. That’s a dominant number, but it’s also slightly and notably lower than the percentage of events that Booth and Kesler have managed to control when they’ve skated together this season. How about that?

There’s a reason you wear gloves, goggles and occasionally nose plugs when performing experiments in your high-school chemistry class. Reactions can be unpredictable, and if the experiment goes wrong you may have to deal with an explosion, or your compounds might emit an unpleasant odor. Odd reactions and noxious smells – that sums up the “Booth and Kesler have no chemistry” argument pretty well.

Tags: , , , ,

44 comments

  1. bbchez
    April 20, 2012

    Not really sure what you’re trying to say here.

    VA:F [1.9.16_1159]
    Rating: -14 (from 18 votes)
    • Brandon
      April 20, 2012

      He’s trying to say that Booth and Kesler DO have chemistry, lol did you even read the title?

      VA:F [1.9.16_1159]
      Rating: +15 (from 15 votes)
  2. akidd
    April 20, 2012

    interesting article. but i think that corsi event stat is misleading in booth’s case. booth takes a lot of low-percentage shots. he drives the puck forward but rarely does much scoring result.

    to the eye it seems like booth and kesler could be playing on seperate teams with the amount that they use each other. not a match made in heaven imho.

    VA:F [1.9.16_1159]
    Rating: 0 (from 2 votes)
    • akidd
      April 20, 2012

      kesler and burrows otoh are always looking for each other out there, most notably(this year) on the pk.

      VA:F [1.9.16_1159]
      Rating: +1 (from 1 vote)
      • ArtemChubarov
        April 20, 2012

        In terms of driving play – Kes and Booth have been awesome. I know from your past comments AKidd that you’re skeptical about Booth’s finishing ability etc. but there’s nothing in his career to indicate that his presence suppresses on-ice sh% for his line-mates, or anything like that.

        VA:F [1.9.16_1159]
        Rating: +4 (from 4 votes)
        • akidd
          April 20, 2012

          and i’ve also really grown to appreciate the energy booth brings every night. he works hard. but once you give him the puck it’s not very likely you’re going to get it back in a high percentage shooting zone. he often drives forward until he hits a roadblock then then either shoots for a low-percentage spot or dishes desperately to a teammate who is covered.

          i don’t think he’s selfish on purpose, just that he lacks peripheral vision. and maybe av’s system is not a perfect fit for him. on sutter’s heavy forechecking kings for example his style of play might be really effective.

          VA:F [1.9.16_1159]
          Rating: +3 (from 5 votes)
        • David Johnson
          April 20, 2012

          Actually, Booth’s on-ice shooting percentage over the past several years is pretty low. Back when he played a lot with Nathan Horton (who is excellent at driving on-ice shooting percentage) he was putting up good numbers, but the last 3 seasons he has been below average. He ranks 191st of 221 players with 2000 minutes of 5v5 zone start adjusted ice time over the past 3 seasons. His 3 year fenwick percentage is 52.3% but his 3 year goal percentage is 42.4%.

          VA:F [1.9.16_1159]
          Rating: +1 (from 1 vote)
          • ArtemChubarov
            April 21, 2012

            Hadn’t realized that David, I quickly checked his last five or so seasons using TOI.com and saw a 9.1 and another above 8.5 and just figured the last two seasons were luck.

            VA:F [1.9.16_1159]
            Rating: 0 (from 0 votes)
          • akidd
            April 21, 2012

            well that explains a lot. nice one david. stats and obsevation can match up…once you find the right stats.

            VA:F [1.9.16_1159]
            Rating: 0 (from 0 votes)
      • akidd
        April 20, 2012

        gotta love burrows. he is invaluable. one of my favs for sure. when’s the last time we saw the camera focus on burrows after an opposition goal(meaning he is the goat of the play?) has it happened this year?

        VA:F [1.9.16_1159]
        Rating: +3 (from 3 votes)
    • TC
      April 21, 2012

      If it’s true that Booth takes a lot of low-percentage shots, then his Corsi numbers both with or without Kesler would suffer. Unless you’re arguing that he happens to take even more low-percentage shots with Kesler on the ice, and is more selective when he’s away from Kesler, and I can’t fathom any reason that would be the case.

      VA:F [1.9.16_1159]
      Rating: 0 (from 0 votes)
  3. Cableguy
    April 20, 2012

    I like this article because it basically shreds Jeff’s argument. And that guy is an arrogant idiot with little of value to say. Ever. Well done.

    VA:F [1.9.16_1159]
    Rating: +2 (from 6 votes)
    • ArtemChubarov
      April 20, 2012

      I’ve worked with JP and discussed the Canucks at length with him – I think he’s a smart commentator and I have a lot of respect for his work. This wasn’t meant as a shot at him, was just using some data to make an argument about the team, and debunk what I believe is a misconception.

      VA:F [1.9.16_1159]
      Rating: +6 (from 6 votes)
    • akidd
      April 20, 2012

      i think that patterson’s intonation gets in the way. i don’t mind what he says but it sure is hard listening to him. compared to sekeres though, he’s a genius. how did sekeres get the job he has? there must be more to punditting than meeting deadlines and being able to talk for four hours straight.

      VA:F [1.9.16_1159]
      Rating: +3 (from 3 votes)
  4. RG
    April 20, 2012

    And this is why I read PITB instead of the Sun.

    VA:F [1.9.16_1159]
    Rating: +17 (from 17 votes)
  5. sarah
    April 20, 2012

    Finally! A US Supreme Court reference on PITB! Nicely done, Mr. Drance.

    Great article too. Personality wise, I imagine Booth gets along with most everyone and Kes gets along with almost no one. They’re basically a sit-com ready odd couple. Booth and Kesler in the morning!

    VA:F [1.9.16_1159]
    Rating: +11 (from 11 votes)
    • Harrison Mooney
      April 20, 2012

      “Finally! A US Supreme Court reference on PITB!”

      I’m putting this on my list of things I never expected anyone to say ever.

      VN:F [1.9.16_1159]
      Rating: +9 (from 9 votes)
  6. Brent
    April 20, 2012

    So you are saying put booth back with Kessler? Who plays up front with the Sedin’s? Back to Burrows, or do we keep him with Booth and Kessler?

    And what about Ebbitt? Not sure what his numbers are like but he has scored some clutch goals. He has played up there before. Surprized he sits while Kassian plays. Not sure what his numbers are but he hasn’t made a huge impression but he hasn’t got a lot of minutes.

    So many questions, not enough numbers.

    VA:F [1.9.16_1159]
    Rating: +2 (from 2 votes)
    • Phileo99
      April 21, 2012

      Now there’s an interesting combo: booth, kess and burr!
      I’d like to see AV try out that combo for this next game . Either Lappy or mayray can play w/ the sedins.
      4th line still our Achilles heel, as evidenced by their lack of ice time vs. Lak 4th liners. Hopefully AV comes up with a fix for that in game5
      And i didn’t see what was wrong with the 3rd line in game4. Can anyone fill me in on that?

      VA:F [1.9.16_1159]
      Rating: 0 (from 0 votes)
  7. jim
    April 20, 2012

    Am I the only person who wants a Sedins-Higgins and Booth-Kesler-Burrows line?

    VA:F [1.9.16_1159]
    Rating: +7 (from 7 votes)
    • akidd
      April 20, 2012

      it would be worth a whirl. but if burrows had a twin we wouldn’t be having this conversation:)

      VA:F [1.9.16_1159]
      Rating: +1 (from 1 vote)
    • Michael
      April 21, 2012

      I think my just asploded.

      VA:F [1.9.16_1159]
      Rating: 0 (from 0 votes)
  8. Chris the Curmudgeon
    April 20, 2012

    To echo RG, it is pretty amazing how much more you can learn about hockey and the Canucks from this blog as compared to people who’ve had entire careers to watch and write about the team. Very nice work as always Thomas. Incidentally, do you have any idea what Kesler and Burrows’ corsi numbers are together? Is that supposed duo actually as good as their reputation suggests?

    VA:F [1.9.16_1159]
    Rating: +8 (from 8 votes)
  9. hold_the_stats
    April 20, 2012

    You never fail to impress. I love IWTG but DN is by far my favorite read here. Keep up the great work!

    VA:F [1.9.16_1159]
    Rating: +3 (from 3 votes)
  10. RJ
    April 20, 2012

    Personally, I’ve seen this “no chemistry” argument being used as a possible explanation when trying to explain Ryan Kesler’s massive drop-off in production from last season (which can be explained partly by his off-season injury that caused him to miss the early part of the regular season, and partly by simple regression to the mean. He’s not a consistent 40-goal guy, unfortunately).

    However, while driving play is great and all, at the end of the day, the team with the most goals wins, not the team that had “the run of play.” And when Capgeek shows some of your cap hit comparables are players like Callahan, Ladd, Eriksson, Fisher, Hartnell, Michalek, Lucic, Horton, J. Staal, etc., it’s pretty clear what Booth is being paid to do: work hard and produce offensive numbers. At a level of production that would put him at around 23 goals and 19 assists over the course of a full season, a cap hit of $4.25 M/yr is simply too expensive for him to retain this mediocre level of production. The Canucks would be better served to have last season’s Mason Raymond (who was also excellent at driving play in the Canucks’ favour) at $2.5 M than David Booth going into 2012-2013, especially with an uncertain salary cap ceiling looming on the horizon.

    VA:F [1.9.16_1159]
    Rating: 0 (from 2 votes)
    • ArtemChubarov
      April 20, 2012

      Booth is probably over paid by about a million dollars, no doubt about that. But he’s a useful two-way player, plays a solid possession game and his PDO was low this season. I think he’s a 50 point guy with if he played a season with normalized percentages + 1st unit PP time (neither of which he got this season) which, is about right for 4 mill.

      VA:F [1.9.16_1159]
      Rating: +4 (from 4 votes)
      • Mack
        April 20, 2012

        I remember Booth saying when he first came here that he thinks he could hit 30 goals again playing with Kesler. Maybe I’m just being optimistic, but I think that he could hit that number again after having a full summer to train. Just look at Chris Higgins. He had an uneven performance after he was acquired last season, but this year he looked energized and his level of play picked up immensely. And having a healthy Ryan Kesler can only help production.

        VA:F [1.9.16_1159]
        Rating: +4 (from 4 votes)
  11. ktownfan
    April 20, 2012

    While I don’t want it to seem like I’m totally disregard Corsi, or Fenwick stats, at the end of the game, say game 3 for instance, the announcers didn’t say and The Canucks win 70% corsi to the Kings 30%. No it was a Kings win no matter what the advanced stats say.

    “So Kesler scored 10 even-strength goals with Booth in the lineup, and Booth assisted on half of them. Meanwhile, Kesler recorded only 10 even-strength assists with David Booth in the lineup, and 40% of those were on goals scored by Booth.”

    60% corsi together or not, those are not the numbers that should be acceptable from the 2nd line. Chemistry, lack of chemistry. Fact is that combination is ice cold production wise and has been for most of the year.

    They are both the same player. Neither use each other well and you see it every game. Both would rather drive the net or take a shot, no matter if it’s a good scoring chance or not than dish the puck to each other to create a quality scoring chance.

    VA:F [1.9.16_1159]
    Rating: +1 (from 5 votes)
    • ktownfan
      April 20, 2012

      further, I don;t think you can base the “chemistry” argument on just this year’s Corsi numbers and make a case this line has Chemistry. Maybe if you want to base the argument on corsi, togther they were better than any other options THIS YEAR.

      You would have to compare to last year, when Booth wasn’t on the team and how did Kesler do with the line mates he had during his career year.

      VA:F [1.9.16_1159]
      Rating: +1 (from 3 votes)
      • ArtemChubarov
        April 20, 2012

        I wouldn’t make the argument based on Corsi alone, but, looking at the gap between both players Corsi without the other, and when they’re together is striking. It’s preposterous to think that two players, who are demonstrably more effective when teamed up together, don’t have chemistry. They’re a solid duo, they proved it all season.

        VA:F [1.9.16_1159]
        Rating: +3 (from 3 votes)
        • ktownfan
          April 20, 2012

          I beg to differ at least production wise. They didn’t prove it all year. The production numbers for both were well below the output expected from a 2nd line, especially at a combined 9,5mil. Their corsi numbers taking in conjunction with production numbers look more like a solid 3rd line.

          I know the blogoshpere loves the corsi and fenwick to try and prove a sport via math like every blogger is the Billy Bean of hockey.

          Booth has a career high in assists at 29. Kesler had 1 season at 50 but avgs less than 30 assists a year. They are both shoot first pass second players and together had horrible shoot% numbers, which whil it gives a high corsi, indicates to me they didn’t work well to put themselves into good scoring chance positions. Whole lot of solo outside attempts.

          Their games may mesh in a pure corsi sense, but corsi is not the be all end all of anything really. It’s a stat and looking at it in isolation shows nothing. Come to me with scoring chances from homeplate numbers and maybe I will believe it. Corsi alone. Nope, especially when the production numbers show an entirely different tale.

          VA:F [1.9.16_1159]
          Rating: +2 (from 4 votes)
  12. bitterguy
    April 20, 2012

    god i hate corsi every time i read one of your articles. you probably think that 5 fishing poles with no bait make you more likely to catch a fish than 1 with. who cares how many shots kesler and booth direct toward goal? how are they being productive by doing that a lot? they don’t score because each of them plays in a bubble and shoot every chance they get, regardless where on the ice they are and regardless of whether another player is open in a better position. both shoot mostly long, weak wristers with no screen and both miss the net a lot, yet that gives them a lot of corsi events. bfd. i’d take half as many events from the twins and still get the same production.

    VA:F [1.9.16_1159]
    Rating: -5 (from 5 votes)
    • Daniel Wagner
      April 20, 2012

      Would scoring chances rather than Corsi make you feel better? Those are shots that are, by definition, not from long distance. Booth and Kesler are +47 and +46 on the season, respectively, good for 4th and 5th on the team behind the Sedins and Higgins.

      http://canucksarmy.com/2012/4/10/canucks-even-strength-scoring-chance-totals-2011-12

      VN:F [1.9.16_1159]
      Rating: +4 (from 4 votes)
      • ArtemChubarov
        April 20, 2012

        Even though I worked hard (along with Mr. Charron) to compile the data in the link Daniel shared above, I don’t believe in shot-quality. One of the main reasons – how closely our “scoring chances” shot quality data corresponds with players individual Corsi numbers.

        Shot quality matters a little bit, over a small sample of time – but analytically, over a full 82 game season it evens out. So I like to use the metric with the largest sample (i.e. Corsi) it eliminates noise, is the best proxy for possession we’ve got, and provides the most value, analytically speaking.

        VA:F [1.9.16_1159]
        Rating: +1 (from 3 votes)
        • Daniel Wagner
          April 20, 2012

          I agree with you, but I think some people might be more receptive to scoring chance numbers than they would be to Corsi. Coaches and players talk about scoring chances; very few talk about Corsi.

          VN:F [1.9.16_1159]
          Rating: +2 (from 2 votes)
          • TC
            April 21, 2012

            It’s a well intended thought, Daniel, but I hope Thomas never changes the content of his posts to appease “some people”. There are hundreds, if not thousands, of hockey blogs dedicated to those people. On the other side, there are maybe a dozen good hockey bloggers who even attempt any sort of quantitative analysis.

            VA:F [1.9.16_1159]
            Rating: +2 (from 2 votes)
        • David Johnson
          April 20, 2012

          I am curious what your thoughts are on my most recent article: http://hockeyanalysis.com/2012/04/19/on-ice-shooting-percentage-is-sustainable

          VA:F [1.9.16_1159]
          Rating: 0 (from 0 votes)
      • bitterguy
        April 20, 2012

        to further my point, with regard to chemistry i don’t even think that scoring chances are a good proxy for chemistry. most people define chemistry as how the players play OFF of one another, (eg. the sedins) versus these statistical analyses of how they play WITH one another. whichever one of kesler or booth that has the puck does what they do without regard for the other players on their line to a much greater extent than any other players on the canucks. neither pass very much when on the rush, they don’t cycle very well, they just get the puck and shoot it as soon as they can see daylight. sure, they’re both good skaters, hard workers, good battlers so naturally they control the puck a lot when they’re out there because they cover a lot of ground and don’t give the puck up easily but i don’t see them playing a team concept style of game with the puck so to me it doesn’t really matter whether they play together or not, they’re the same players regardless.

        if you define chemistry as being what makes the sedins so good together and better than the sum of their parts as a line i don’t think corsi or scoring chances is how you measure that.

        VA:F [1.9.16_1159]
        Rating: +1 (from 3 votes)
        • Daniel Wagner
          April 20, 2012

          This issue is that there are different types of chemistry. If you define chemistry as what the Sedins do, then no one else in the NHL has chemistry. Different lines play in different ways. Maybe the best linemate for Kesler is a player similar to him, who goes hard to the net, shoots first, and does his best work off the rush. Kesler paired with a player who wants to cycle the puck down low might not have much success, simply because that’s not Kesler’s game.

          Where you see two players not passing well or using each other well on the ice, maybe the team sees two players who complement each other’s style by being both so aggressive on the rush and such good skaters.

          All I’m saying is that if you want every line to have the same chemistry as the Sedins, you’re dreaming. I think Drance was being facetious when he suggested that Kesler and Booth have better chemistry than the Sedins, but you can’t deny that they dominate possession and create scoring chances when they are together. Yes, they also take a lot of shots from the outside, but that leads to better, higher quality opportunities.

          VN:F [1.9.16_1159]
          Rating: +3 (from 3 votes)
    • akidd
      April 20, 2012

      agreed, bitter guy. nice post.

      VA:F [1.9.16_1159]
      Rating: 0 (from 0 votes)
  13. tom selleck's moustache
    April 20, 2012

    Thanks for the article, Drance. It’s hard to know what to really make of the number of goals Booth and Kesler are assisting on for each other without some context though. Certainly, it’s not as bad as Paterson makes it out to be when you dig deeper, as you have done. But I think there needs to be something to reference as well (say, comparing them to other duos).

    For example, I wouldn’t be surprised if the Sedins assisted on 80-90% of each others goals. If we assume them as being on one extreme of “chemically compatible linemates”, then the 40-50% assist rate of Booth and Kesler doesn’t look as great. Maybe comparing them to the Sedins isn’t fair; but even compared to other duos would be an interesting exercise.

    VA:F [1.9.16_1159]
    Rating: +2 (from 2 votes)
  14. taco
    April 20, 2012

    Ya, I think the use of corsi is a bit weak here. My thinking is this, you could have two completely different pairs of players.

    Say pair one is the sedins. The majority of their play is done as a combined attack, they always look to each other and rarely go it alone. On the other hand pair number two is Ryan and David. They mostly go it alone and rarely use each other.

    Both pairs of players could have good with-or-without you corsi numbers. The Sedins for the obvious reason that they have chemistry together. The AMEX pair because they take turns, one taking the puck, driving to the net or taking a shot, then the other taking his turn to do the same. My point is, would we really call this chemistry?

    Another way of saying this would be to picture a line made up of Crosby and Malkin. I bet their WOWY corsi numbers would be awesome. Good chemistry or just two good players.

    I bet my corsi numbers would be great with Hendrik. Does that mean we have good chemistry, or is Hendrik just maybe a bit better then most of my other usual linemates?

    VA:F [1.9.16_1159]
    Rating: -1 (from 1 vote)
  15. taco
    April 20, 2012

    Hmm… So then if I wanted to create the best lineup for a team, we could randomly mix all the players up and let everyone play with everyone else for as much of the season as possible. Then calculate all the WOWY numbers to compile a ranking which we could use to make the optimum lines.

    Sound familiar?

    Then we could toss all that out the window and stick Booth with the Sedins in the most important game of the year. WTF?

    VA:F [1.9.16_1159]
    Rating: +2 (from 2 votes)
  16. Reuben
    April 21, 2012

    Thanks for the article, once again you guys are an island of reason in a sea of reactionary nonsense.

    VA:F [1.9.16_1159]
    Rating: +2 (from 2 votes)