Drance Numbers: Bolland is hardly Sedin kryptonite

Drance Numbers is the silly research wing of PITB. While Messrs. Wagner and Mooney blog nationally and solve mysteries, Drance Numbers will look into the minutiae of quantifiable NHL data and bore you with it every Friday.

***

When Chicago Blackhawks’ defensive ace Dave Bolland referred to the Sedins with the hackneyed “Sisters” moniker this week, he set off a new wave of feeble trash-talk between the Vancouver and the original six club, who are something of a perpetual thorn in the Canucks side.

It was the latest chapter in an increasingly heated rivalry, not just between the two teams, but on an individual level between the Sedin brothers and Dave Bolland as well.

Every protagonist must have a foil, and the Sedins’ foil is undoubtedly Bolland. He’s the Rommel to their Patton, the Prince Joffrey to their Rob Stark. If you listen to Blackhawks fans (which, for the record, I don’t recommend doing), they’ll tell you a tale of how Dave Bolland “has the Sedins’ number” and how the twins “just can’t figure him out.”

Many in the media will probably tell you the same thing.

Now I suppose there is some superficial merit to this take. After all, the Sedins were crushing the Blackhawks with nine combined points in the first three games of their first round series last April. When Bolland came back, however, the series turned pretty dramatically. Bolland had a statement game in game 4, putting up four points and absolutely dominating the twins.

And, after Bolland returned to the Blackhawks lineup, Henrik Sedin managed only one point and was -7 through the final four games of the series…

There was also the iconic “Daniel freaking out at Bolland” moment in game three of the series in 2010. It was a moment of frustration and annoyance and, according to many observers and yarn spinners, it was a window into the team’s utter lack of discipline.

The Sedin “freakout” taught us a subjective lesson about why this Canucks core could never succeed. In some ways it was like the Marchand punch, before the Marchand punch was the Marchand punch.

I’m always skeptical when it comes to conventional wisdom, and this is another one of those fairy tales that always struck me as less than wholly true. It seemed to me in April that the Blackhawks and the Canucks were more evenly matched than their one and eight seeds would indicate. In fact, by the numbers that matter, the Blackhawks were among the best teams in the Western Conference last season. So did Bolland personally turn the series around, or did he have a big game at an opportune moment as the series turned for other reasons?

I figured I’d look into the history of the Bolland v. Sedin matchup to examine whether or not Bolland dominates the Sedins to the extent we’ve been told.

Now, when it comes to this matchup’s “history,” we’re not talking about a particularly large sample of games. Interestingly, Bolland didn’t start the 2008-09 series or the 2009-10 series matched up against the twins. In 2008-09, Joel Quenneville originally trusted Sami Pahlsson with the bulk of minutes against the Sedins at even-strength, and to start the 2009-10, series Coach Q deployed the Toews line against the twins in game one. In both cases, he adjusted mid-series and the adjustment was effective.

Dave Bolland only became an NHL regular in 2008-09, and I’ve only looked at games in which Bolland (or his line) were the Sedins’ primary matchup.

We’re talking about 13 playoff games (4 games in 2008-09, 5 games in 2009-10, and 4 games in 2010-11), and 10 regular season games (including the 2 games in November this season). This means we’re dealing with under 120 minutes of even-strength playoff ice-time in which the Sedins have been matched up with Bolland and roughly sixty minutes of even-strength ice-time in the regular season.

To get an accurate idea of their head-to-head ice-time I used timeonice.com’s head-to-head ice-time script, which is indispensable.

To start with, I looked at results. I calculated the Sedins’ production rate (points per 60 minutes) during the regular season (since 2008), and then I calculated their rate of production (Pts/60) when they’re on the ice at even-strength against Dave Bolland.

Looks like the Sedins don’t have much of an issue scoring on a team with Dave Bolland in the regular season:

Regular Season: EV P/60 Overall EV P/60 H2H v. Bolland
Henrik Sedin: 3.06 4.77
Daniel Sedin: 3.18 6.36
Combined: 6.24 11.13

 

Those numbers are pretty eye-opening. Far from being “a Sedin stopper,” in their ten regular season meetings, the Sedins have managed to burn Bolland at almost double their usual even-strength scoring rate.

But Bolland’s reputation wasn’t built in the regular season, it was forged in the playoffs. So how did the Sedins fare against Bolland in the last three postseason series against Chicago?

Postseason: EV P/60 Overall EV P/60 H2H v. Bolland
Henrik Sedin 1.86 2.1
Daniel Sedin 1.73 1.05
Combined 3.59 3.15

 

Note: Because the playoffs give us a smaller sample, I used the last five years worth of data to come up with the Sedins EV P/60 Overall Number.

Against Dave Bolland, the Sedins have combined to score at a marginally lower-rate than they have otherwise in the playoffs. Of course, the whole story is that Henrik scores slightly more while Daniel scores significantly less.

The net result is flattering to Bolland. Seeing as the Sedins are two of the best offensive players in the world, the fact that he can slow them down is impressive. It also makes life a lot easier for his teammates, and we know how valuable an “enabling” third line center can be.

But make no mistake, based on the results, that’s what Bolland is doing: slowing the Sedins down, a far cry from the “total domination” some would have you believe he’s brought down on the twins the past three years.

Now, results are one thing, but what’s more interesting to me is whether or not Dave Bolland can control events when he’s on the ice against the Sedins. I went through about thirty play-by-play sheets at NHL.com to compile the data I used in the next section, and got a hand along the way from Cam Charron. Here’s the distribution of shots on goal (including goals) when the Sedins are matched-up against Bolland in the regular season:

Regular Season SOG% 2008-09 2009-10 2010-11 2011-12 Overall
Dave Bolland 57.1% 33.3% 40% 16.7% 42.1%
The Sedin Line 42.9% 66.6% 60% 83.3% 57.9%

 

It’s funny that Dave Bolland chose this season to mock the twins, seeing how in about eleven minutes of head-to-head even-strength ice-time this season, Bolland and his line-mates have managed two shots on goal, while the Sedins have put up a shot-on-goal per minute (and have put two pucks past Crawford) in that time.

But is the postseason a different story? Does Bolland lift his game in the playoffs, while the Sedins wilt under pressure? Nope:

Postseason SOG% 2009 2010 2011 Overall
Dave Bolland 42.9% 37.5% 47.2% 43%
The Sedin Line 57.1% 62.5% 52.8% 57%

 

Pretty similar, eh? I’d point out that the 2010 series featured a number of Chicago blowouts, and I suspect the Sedin’s 62.5% number is unduly influenced by score effects as a result.

While he may not control events, the Sedins never seem to score any dagger goals with Bolland on the ice. I know that Bolland has a sky-high postseason PDO, so I wanted to check out whether or not Bolland has been mostly lucky, or if he is legitimately able to somehow suppress the Sedin’s shot totals on a game-to-game basis? Well…

Sedin SF/60 (since 08-09) Daniel Henrik
Regular Season 34.1 33.8
Regular Season V. Bolland 46.64 41.97
Postseason 33.2 31.3
Postseason V. Bolland 36.37 36.37

 

Ouch. These numbers are probably over-dramatic because of the small sample size we’re looking at, but they still reflect a basic reality: namely that Bolland’s supposed “domination” over the Sedins is as over-rated as the Aerosmith classic, “Dude (Looks Like a Lady).”

While Bolland remains an elite two-way player, his “Sedin nemesis” title seems to me to be based on perception that has been mostly shaped by a couple of isolated incidents, rather than facts. In the larger picture, the Sedins have had a good deal of success against Dave Bolland, and will likely continue to should their respective team’s meet again this postseason.

Tags: , , , ,

13 comments

  1. Jeff C.Ho
    December 16, 2011

    Any reason why you used shots instead of Corsi or Fenwick?

    VA:F [1.9.16_1159]
    Rating: 0 (from 0 votes)
    • ArtemChubarov
      December 16, 2011

      I wanted a more “results” based metric, and a metric more accessible to folks who may not necessarily be well versed in hockey analytics.

      VA:F [1.9.16_1159]
      Rating: 0 (from 0 votes)
  2. Anonymous
    December 16, 2011

    Great points made here! Things sure are different when you use your brain and analyze rather than listen to what the media spouts. ;)

    Interestingly enough, the pre-game show before the Carolina game did a little highlight package of all the things Bolland does to ‘shut down’ the Sedins while they were discussing his comments. I got trigger flashbacks from our last few Blackhawk playoff series (oh how I wish I could forget! Remembering the absolute loathing I had for that team that was allowed to whatever they wanted to us, ugh. Byfuglien – ok no let’s stop while we’re ahead.) Seeing it all put together like that, it was ridiculous how much cheap crap he got away it. Hacking, slashing, slewfooting, and nothing being called. The refs turn a blind eye and everyone is up in arms about how the Sedins can’t play in the playoffs! No one’s asking why the refs are letting everything go on them, people are allowed to do whatever they want when if this was a regular season game they’d have penalties. I don’t understand that way of thinking, but hey maybe that’s the just the way hockey works and we shouldn’t question it. But if the Hawks contingent praises him for being dirty and getting away with it, I suppose we have to give him kudos as well.

    VA:F [1.9.16_1159]
    Rating: +15 (from 15 votes)
    • tom selleck's moustache
      December 16, 2011

      But that’s the thing, no one has really and truly questioned the thinking behind allowing the refs to call the game differently in the playoffs vs regular season. The conventional wisdom of “not wanting to influence the game” and “allowing the players to decide the game” had just been accepted as gospel without actually thinking about the merit of the argument itself.

      But when you stop and really think about it; that wisdom just doesn’t make add up and, even further, is really comes off as kind of stupid when you break it down. Essentially, it’s saying that it’s ok to cheat in the post season. What other sport in the world deems that as acceptable? How is that supposed to be a good thing?

      VA:F [1.9.16_1159]
      Rating: +12 (from 12 votes)
  3. BeCanucks
    December 16, 2011

    But but but… you mean that narrative and figures don’t add in the NHL? What a novelty!
    Without kiddin’, good job at digging the data to expose Bolland for what he is: an honest checker.

    VA:F [1.9.16_1159]
    Rating: +2 (from 2 votes)
  4. noulp
    December 16, 2011

    This is some of the greatest writing in hockey history. Very refreshing to see.

    VA:F [1.9.16_1159]
    Rating: +1 (from 3 votes)
  5. Origamirock
    December 16, 2011

    Great article, I hope we do end up seeing a Hawks-Canucks series this year if nothing else just to get a larger sample size (and who cares about compelling hockey?). Having said that, you spelled Robb Stark’s name with one b. FOR SHAME!

    /nerdrant

    VA:F [1.9.16_1159]
    Rating: -1 (from 1 vote)
  6. Logic Queen
    December 17, 2011

    I’m sorry, but there are some major holes to be considered here. You simply cannot add together point production when considering the “effectiveness” of the Sedins against Bolland. It would be far more accurate to use goals in this instance because more often than not, when one Sedin scores it’s because the other one assisted. This situation would result in “double points” in your analysis and give an extremely skewed result, probably what you were looking for. I’d be far more willing to listen, if you took the time to figure that out. Also, shots on goal?!? Shots percentage would be far more telling, if you want to talk about effectiveness. Stats are very easy to manipulate if you don’t tell the whole story.

    VA:F [1.9.16_1159]
    Rating: 0 (from 6 votes)
    • Daniel Wagner
      December 17, 2011

      Drance didn’t “add together point production.” He took the point production of each Sedin separately and showed that in the regular season each one individually scores more than their season average against David Bolland. There’s no double up effect there.

      Shooting percentage wouldn’t actually tell us anything about defensive effectiveness. There has been an open challenge amongst advanced stats enthusiasts to find a way to quantify shot quality and haven’t been able to do it. What has been seen, however, is that shooting percentage varies widely largely due to luck. Good defensive players have been shown to cut down on shots against, not shooting percentage.

      There’s no manipulation here and your logic is flawed.

      VN:F [1.9.16_1159]
      Rating: +2 (from 6 votes)
    • Benny
      December 18, 2011

      The only problem here is that Drance is comparing the sedin’s performance with Bolland on the ice vs when they are not playing against Bolland, thus whether they get “double points” for assisting on one another’s goals is irrelevant.

      As to Shot’s on goal not being an effective stat I ask you this:
      Really? That’s where you’re gonna go?

      Thank you and goodnight.

      VA:F [1.9.16_1159]
      Rating: -2 (from 2 votes)
  7. Logic Queen
    December 18, 2011

    Here is my point, and for the sake of brevity I’m focusing on the playoffs for now, I can do the regular season research if necessary.

    In 19 playoff games against the Blackhawks (09, 10, & 11) Henrik Sedin has 5 goals and 12 assists fir 17 points (10EV, 7PP). Daniel Sedin (in the same games) has 8 goals and 8 assists for 16 points (8EV, 8PP). These 33 points came on only 23 events. Of those 23 events, only 6 of them occurred while Dave Bolland was on the ice (26% of the time). During the same time period Dave Bolland played in 16 playoff games against Vancouver. In those games he had 8 goals and 9 assists for 17 points (10EV, 3PP, 2SH, 2EN). Of those 17 events, 11 of them came while either or both of the Sedins were on the ice (64% of the time).

    The main point I wanted to focus on with these statistics is that while the Sedins had 33 points in these games they occured in just 23 events. This is exactly what I was talking about with “double points”. It is completely relevant to consider this fact. I also looked at all on-ice situations, not just even strength, the Sedins score a lot of power play points and Bolland had 2 huge shorthanded goals in the history of this series. There a lot of things to consider when looking at this match-up and pulling out 2 random statistics is not going to give you a thorough understanding of how effective either the Sedins or Bolland are.

    VA:F [1.9.16_1159]
    Rating: +1 (from 1 vote)
    • ArtemChubarov
      December 19, 2011

      Hey QL, despite your haughty handle, your hostile assumption of bad faith on my part, and the shallowness of your relationship with statistics and hockey stats in general, I appreciate you chiming in here.

      I looked at even-strength numbers because that’s the more relevant game state for a head-2-head analysis. On the PK all you can do is keep position, but a Sedin PP goal likely has very little to do with Bolland’s play. Also, any SHG Bolland scores is probably the fault of the PP point man when you consider how the Sedins generally set up low in the zone.

      “Double points” are worth looking at, but their presence doesn’t have any impact on this research. Your stats made no effort to show that my results change AT ALL using your metric of “every Sedin goal must count only as one.” If you want to go through 3 seasons worth of Sedin points and show that they score at a faster rate against other players than Bolland – then do that. Merely pointing out that line-mates often assist on the goals of the other line-mates doesn’t throw any of my work into doubt.

      Using points per sixty in both the regular season and the playoffs as my “rate” doesn’t have any “double points” bias, seeing as “double points” would surely show up in both rates, right?

      And anyway, scoring rates were a small part of this piece and merely a smell test. The rub is in the shots against numbers, and I’d urge you to read up on the relationship between shot suppression, shooting percentage and “defensive skill.”

      All the best my dear Queen of Logic.

      VA:F [1.9.16_1159]
      Rating: 0 (from 0 votes)
  8. shoes
    December 18, 2011

    I did some further in depth research which really opened my eyes. Bolland got away with hacking, whacking and highsticking at a much higher rate than any other player in history ever has or it appeared so anyway. Sure enough….research showed that he had a picture of several refs in a compromising act with a goat…….so I checked around all the costume rental places within a mile of the Bell Center and sure enough….a anonymous guy who paid cash and signed the bill…”Dave B.” had rented a goat suit after carefully trying it on and seeing if it had all the features he wanted, such as lipstick and ‘secure’ zipper.

    VA:F [1.9.16_1159]
    Rating: 0 (from 2 votes)