On Mason Raymond’s Playoff Production, or Lack Thereof

Photo by Gerry Thomas/NHLI via Getty Images

For a variety of reasons, most notably his slender build and the fact that he had a tremendous run of awful puck luck in 2010-11, and followed up on it with a disappoiting season in 2011-12 coming off of a serious back injury, Mason Raymond catches a lot of flack from Canucks fans. He’s a perimeter player who falls down a lot is the general critique – nevermind the solid underlying numbers – and his offense dries up in the playoffs.

But does it really when you break it down? In comparison with your average NHL player, does Mason Raymond have a unique habit of suddenly becoming an ineffective offensive player once the playoffs begin? I’ll look into it on the other side of the jump.

Advertisement - Continue Reading Below

To figure this out we’ll pivot off of Eric Tlusky’s fine work over at NHLNumbers.com last fall in a post titled "Do Some Players Elevate Their Games in the Postseason?" Using Eric’s methodology, which is brilliant in its simplicity, let’s take Mason Raymond’s career production per minute, and extrapolate an "expected total" for postseason points. We can then see how well he’s performed against reasonable expectations set by his own performance over the course of his career.

So far Mason Raymond has put up 175 points while playing 5570:22 minutes of regular season hockey in his career. That gives him a scoring rate of .031 points per minute. In the postseason, Mason Raymond has played 845:52 of hockey, so based on his career regular season scoring rate we’d expect him to have 26 career postseason points.

Now here’s a statistical quirk, when Eric added up all of the expected points for every player in the league since the lockout, he found that expected playoff point totals outpaced actual playoff point totals league-wide by about 12%. Eric posits that "the mix of better defenses and fewer penalties actually suppresses scoring slightly in the playoffs," and that strikes me as compelling. It also gives us a sort of playoff production "translation number" of 0.88.

When we apply that translation to Mason Raymond’s pace, we’d expect him to have put up 23 playoff points in his career so far, which is better than the number of points he’s actually produced. That number is 16 points, by the way. So Mason Raymond has under-performed his expected offensive output in the postseason by seven points.

Here’s the rub: Raymond isn’t underperforming his own regular season production in the playoffs by a sufficient amount for it to be all that meaningful statistically speaking. After all, we’re talking about a fifty-one game sample and, as we know, variance and random chance can greatly impact results over a relatively small sample of games(*). To account for the impact of random chance, Eric used a "standard error" which is basically a tool that allows us to ferrett out whether or not a deviation in our expectations is just variance, or something more sinister.

Advertisement - Continue Reading Below

(*) Yes, I consider fifty games a small sample when it comes to judging an individual’s true talent – if you don’t believe me, I own a perfect replica of Steve Mason’s Calder Memorial Trophy for you to spit shine.

So to calculate a "standard error" for Mason Raymond’s playoff production, we simply find the square root of 23: sqrt (23) = 4.79. We can then determine that Raymond’s actual playoff production is only one and a half standard deviations away from where we’d expect it to be. That’s a slim enough margin for us to reasonably presume, based on the current evidence, that it’s mostly "bad luck" in Raymond’s case.

The evidence supporting that conclusion becomes even more compelling when you consider that his personal shooting percentage is a third lower in the playoffs than his career regular season shooting percentage. Also that his lack of production mostly stems from an unsustainably low 4.58 on-ice shooting clip during Vancouver’s extended 2010-11 playoff run…

That Mason Raymond falls within a standard deviation and a half of his expected career playoff production also makes him, quite unremarkably, like 95% of other NHL players. So in comparison with every other player in the league, Raymond isn’t a meaningful outlier on the "he just doesn’t produce in the playoffs" spectrum at this point in his career.

Here’s a fun table of every current Canucks skater with more than 10 games of playoff experience, and how they’ve performed offensively in the postseason when set against expectation based on their record of regular season production since the 2005-06 season:

Advertisement - Continue Reading Below
Skaters Actual Playoff Points Expected Playoff Points based on regular season production
Mason Raymond 16 23
Henrik Sedin 71 62
Daniel Sedin 64 59
Alex Burrows 29 33
Ryan Kesler 36 34
Alex Edler 27 24
Dan Hamhuis 18 16
Kevin Bieksa 24 26
Jannik Hansen 14 18
Chris Higgins 19 23
Derek Roy 25 27
Maxim Lapierre 13 15

So here’s a list of other Canucks who have underperformed in the playoffs offensively when judged against their "expected" totals: Alex Burrows, Kevin Bieksa, Jannik Hansen, Chris Higgins, Derek Roy and Maxim Lapierre. Hansen and Higgins have both underperformed their expected totals by a full standard error.

So where are the articles and the tweets bemoaning the inability of those players to produce in the playoffs? If your answer is that "such statement would be silly because those players haven’t underperformed expectations by much," you’re right. Such statements would be silly.

By the way, here’s the information from the above table in graph form. You might notice that the plotted data has a rather linear look to it. If there are any outliers among current Canucks skaters, well that’s certainly not obvious:

Advertisement - Continue Reading Below

Look, I’m not saying that Mason Raymond will for sure be a reliable offensive threat in the postseason, anything can happen over a small sample of games. But I do think it’s significantly more likely that Raymond’s limited postseason production so far in his career is a result of variance and chance, rather than some vague, inherent inability Raymond is presumed to possess that prevents him from producing offense in the playoffs…

  • Cale

    Are the results still the same if you just look at even strength numbers? Raymond’s only had 76 minutes of power play time during the playoffs which seems like an awfully small sample size.

    In the years that the Canucks made the playoffs, 26% of Raymond’s points came on the powerplay during the regular season. During the playoffs, only 19% of his points have come on the powerplay.

    Maybe it’s just the powerplay that’s not particularly effective during the playoffs?

  • Cale

    I also wonder if players like Raymond and Hansen who are posting their best years (or close to them) have had their data skewed.

    Meaning, if Raymond/Hansen are putting up more points per minute this year than in past years their “playoff point/min expectation” will have been raised. However, they haven’t been to the playoffs this year during which they would have a chance to increase their actual playoff points/minute.

    In essence – aren’t you constructing one set (expected points/minute in the playoffs) by gauging this year’s contribution? Wouldn’t it be more fair to look at points/minute only in past seasons?

  • BrudnySeaby

    Interesting to see that the Sedins both produce slightly above their expected playoff output, since they also have a (ridiculous) reputation for “disappearing” in the playoffs.

  • BrudnySeaby

    But if I convince myself Raymond is weak and unmanly, despite being significantly more successful in his career than I am in mine, then I feel better about my lot in life.

  • Cale

    Fair enough Thomas.

    Really great analysis here, and I I appreciate your commitment to sample sizes. Something I get to stress in an hour when I have a quarterly meeting/report for work.

    Thanks for the lunch distraction!

  • I hope MG can get some of those 5% guys on the Canucks!!

    “That’s a slim enough margin for us to reasonably presume, based on the current evidence, that it’s mostly “bad luck” in Raymond’s case.”
    Based on what current evidence?

    I don’t think the fact that his shooting percentage in the playoffs is a thrid lower supports this….perhaps the harder hitting more interference/hooking/holding atmosphere of the playoffs suits his playing style much less than players like Higgins or Hansen.

  • BrudnySeaby

    That table was incredibly helpful for solidifying why I never bought the bunk that Raymond or the Sedins “regress” when playoff games start.

    You are a voice of reason and I thank you.

  • MayRay’s reputation is sound, in spite of the statistical evidence to the contrary. I will always remember his missed backcheck on Kopitar in Game 3 last year. He hustled back and braced for impact as Kopitar stopped and MayRay slid on by. Kopitar then pops it under the cross bar.

    It’s nice to read MayRay’s point production is greater than what’s perceived, however soft is soft. He has all the speed/finish in the world, but can’t up his physicality when it’s needed most; therefore meeting a statistical expectation is completely overlooked. Either score more than expected or at least stop the opposition with good defensive play; i.e. Kesler.

  • ChrisB

    I wonder what the expected playoff points for the Sedins would be if one were only to take into account their last 5-6 seasons since they broke out.

    Obviously, the expectations of when they were 40-50 point 2nd liners are different than since they’ve been putting up ppg.

    I get the idea is to try and use as large a sample as possible, but when you’re talking about expectations, those change over time. So shouldn’t you really be using the “data” that people are basing their expectations on?

    For the record, personally I think they’ve done well enough.

  • It blows me away how many people simply refuse to notice how scoring goes down for *everyone* in the playoffs.

    The NHL, being the world’s only league who arbitrarily decides to stop calling penalties in the part of the season that it has (equally arbitrarily) determined is the “real season”, pretty much sees to this.

    • ChrisB

      FTR – This is not the whole story?! the last 5 Cup winners have actually outscored their season average by a significant margin. (5 /7 in total)
      So it is not true that it is any tougher for successful teams to score in the playoffs.

      The fact is: Cup winners need to ‘Overperform’
      their reg. season stats considerably -beyond the luck/variance margin!!! –

      That’s the key point –Van. fans & media keep missing. It is not enough to just meet expectations.They need to reach a new level of performance. In addition, the GD has surprisingly stayed consistent ~I believe ~ave .70 for Cup winners. Even though teams are facing much harder comp.

      Its clear Cup winners must dominate.
      The Nuxs have not. Their clear wins (2 goal+ victories) and GD over 53 gms since coach took over proves this.

      Since AV took over Nux have underperformed in GF & in GA (more than 12%). He has not got it done!The team has had home ice adv. 9 out of 11 series. Team is only 6 and 5. And team has not beat one better team! All signs they can’t reach this higher level of performance!

      The mistake of fans and Media supporters of Av is simple. They mistakenly believe that it is ‘harder to score in playoffs’ and the comp. is tougher so can’t expect ‘better’ performance.

      When this is EXACTLY what is demanded.

      The team desperately needs a new coach to have an opportunity to get more from this talented group before window closes.To not do so is clearly one of the greatest mistakes of this sorry franchise.

  • ChrisB

    The methodology of calculating the SE is incorrect.

    It should be: sqrt# of reg season games * SD of expected value

    I would suggest a two tailed confidence interval would be a more robust way to determine outliers.