43
Photo Credit: Vancouver Canucks / Twitter

Do Faceoffs Really Make A Difference?

The faceoff circle has been the center of major discussion throughout the Vancouver Canucks’ 2018/19 season. Whether it’s Bo Horvat leading the league in draws taken by a ridiculous margin, the contributions of Jay Beagle, or Elias Pettersson’s continued struggles on the dot, faceoffs have been the focal point of multiple Vancouver storylines. That probably makes this as good a time as any to discuss whether they’re even worth talking about in the first place.

Despite the wide swath of advanced statistics now available, faceoff win percentages continue to be looked at as a quintessential hockey number—but their impact on the game may be highly overrated. As a case study, we’ll take a look at the Canucks’ faceoff-related stats this season to see what, if any, correlation can be found between faceoff success and success in general.

The Canucks’ Faceoff Trend

Season 2014/15 2015/16 2016/17 2017/18 2018/19
Vancouver Faceoff % 46.7% 45.4% 49.5% 51.1% 47.9%

Before we embark, it’s important to look to the past for a bit of context. The Vancouver Canucks’ overall performance in the faceoff circle is down by more than 3% this season from where it finished last year. Overall, the Canucks’ faceoff percentage has been all over the place in recent seasons, and while it had seemed to improve under the coaching reign of Manny Malhotra before this year, it’s hard to say there’s been any correlation between it and the team’s win percentage—at least on the surface.

Does Winning Faceoffs Correlate To Winning Games? 

Vancouver Record When Winning Or Tying The Faceoff Battle Vancouver Record When Losing The Faceoff Battle
6-10-0 14-11-4

There’s no obvious connection between the Canucks’ long-term success and their faceoff percentage, so next, we’ll look at its correlation to their success on a game-by-game basis. As of this writing, the Canucks have played 45 games, which makes for a relatively decent sample size.

As the numbers reveal, if there’s any correlation between the Canucks’ winning percentage on the dot and their winning percentage on the scoreboard, it’s the opposite of what one would expect. So far, Vancouver actually does significantly better overall when they lose the faceoff battle to their opponents than when they win or tie it—with records of .375 and  .552 respectively.

Of course, this doesn’t necessarily suggest that the Canucks are better off losing draws and should start doing it on purpose. Instead, it’s probably evidence that there isn’t a strong relation between winning faceoffs and winning games—at least as far as the 2018/19 Canucks are concerned. If there is any correlation, it must be on a situation-by-situation basis, which would be much more difficult to track.

Does That Jive With Results Across The NHL?

The numbers do not support a correlation between the Canucks’ faceoff success and their overall success, and that absolutely jives with most of the leaguewide inquiries that have been done on the subject. A handful of in-depth studies—including this one from Sports Illustrated in 2017—were unable to find any tangible connection between faceoffs and achievement for teams or individuals on either a short- or long-term basis.

 

Faceoffs Across Different Situations 

Even Strength Faceoff % Powerplay Faceoff % Shorthanded Faceoff %
48.1% 48.6% 44.9%

One small inkling of faceoffs being consequential can be found, however, in the Canucks’ faceoff record across different situations. The team’s win percentage dips dramatically when they’re shorthanded, and that certainly seems to correspond with their woeful performance on the penalty kill. Vancouver’s struggles with shorthanded faceoffs aren’t the sole reason for their PK troubles, but they’re probably a factor.

 

Faceoffs Across Different Zones 

Offensive Zone Faceoff % Neutral Zone Faceoff % Defensive Zone Faceoff %
48.6% 44.8% 49.7%

Since we looked at faceoffs across different situations, we may as well look at them across different zones, too—though there’s not much to report. The Canucks have been significantly worse at draws in the neutral zone, but it’s hard to imagine why that would be—and they are, fortunately, the least impactful kind of faceoff.

 

Jay Beagle’s Impact On The Lineup 

Vancouver Record With Jay Beagle Vancouver Record Without Jay Beagle
12-8-1 8-13-3

Jay Beagle was signed as a UFA this offseason to fill a number of utility roles, including that of faceoff specialist. Thus far, his winning percentage on draws has actually trailed Bo Horvat’s, but Beagle has still come mostly as advertised when it comes to faceoff prowess. But has that made a difference?

Vancouver does have a significantly better record with Beagle in the lineup than they do without him, and his absence from the team did coincide with the darkest days of the season thus far. However, Beagle brings a lot more to the table than just faceoff abilities, so it’s impossible to draw direct conclusions. It’s probably fair to say that his presence on the dot has made a difference this year, but it’s difficult to say exactly how much of a difference.

 

 

The Elias Pettersson Factor 

Vancouver Record When Elias Pettersson Wins At Least 50% Of His Faceoffs Vancouver Record When Elias Pettersson Does Not Win  At Least 50% Of His Faceoffs Elias Pettersson’s Production When He Wins At Least 50% Of His Faceoffs
6-4-0 10-13-4 17 points in 10 games

Elias Pettersson’s success, or lack thereof, in the faceoff circle has been the topic of much discussion this season—it as, after all, arguably the only area of the game in which he has struggled.

While the Canucks do perform better when Pettersson wins at least 50% of his draws, it’s a marginal difference and doesn’t point to much in the way of correlation. More intriguing, however, is the fact that Pettersson does much better as an individual when he dominates the dot—with 17 of his 42 points coming in the ten games in which he won at least half of his faceoffs.

 

Faceoffs Versus Corsi

Player Faceoff % Corsi For/SAT %
Bo Horvat 52.4% 46.2%
Elias Pettersson 41.9% 50.0%
Brandon Sutter 48.5% 38.8%
Jay Beagle 50.9% 39.6%
Markus Granlund 41.4% 47.0%
Adam Gaudette 44.2% 47.8%

As this author’s last piece noted, the Canucks’ Corsi leaderboard is a mess in 2018/19, and that certainly holds true when it comes to comparing it to the faceoff charts. While it would make sense that faceoff success might correlate to possession—as a faceoff win usually means gaining possession of the puck, at least temporarily—there’s not much to work with here. The team’s three best faceoff men—Horvat, Beagle, and Brandon Sutter—have the worst SAT% among centers, but that can probably be attributed to the heavier defensive minutes they are asked to shoulder.

Conclusion 

In the end, it appears that success in the faceoff circle doesn’t really correlate to success on the scoreboard—not for the 2018/19 Canucks, and not for any other NHL team, either. While winning a big faceoff at a key moment can certainly change the game, such an event is relatively rare, and the vast majority of faceoffs just don’t matter all that much in the grand scheme of things. Faceoffs can be important on a situation-by-situation basis, but their overall impact on the game appears to be highly overrated.



  • North Van Halen

    Sigh, here’s the thing about face offs that gets ‘lost’ in stats: They don’t matter until they do. In other words you could win 20 straight face offs, lose one on the penalty kill/in the last minute/straight back to the point, suddenly the pucks in the net and the outcome of the game changes.
    This is why you can show me 1000 statistics that show me face offs don’t matter and I’ll show you one replay of the faceoff won with 10 seconds left that leads to the tying or winning goal that renders every one of those stats completely irrelevant to me anyway.

    • Dirk22

      You could say this about a lost puck battle in the corner though. Or a broken stick. Or a mis-read by a defencemen. All of those can lead to goals too. The point is not that faceoff losses can’t hurt (as you can obviously get scored on after a faceoff loss). It’s that it’s not a hugely relevant statistic when it comes to the success of a team. I thought the author made that pretty clear:

      ” While winning a big faceoff at a key moment can certainly change the game, such an event is relatively rare, and the vast majority of faceoffs just don’t matter all that much in the grand scheme of thing”

      • Bud Poile

        This is so much twaddle. The vast majority of faceoffs are taken in the neutral zone and of the least significance.
        Beagle’s career is in the defensive zone and why the Canucks win games with him in the lineup.
        Sutter and Beagle allow Bo and Petey to offensively dominate in the circle. That has been more than demonstrated this season when Sutter and Beagle were in and out of the lineup.
        Bo’s offensive production falls off dramatically when he has to cover for Sutter or Beagle.
        Faceoffs are and always have been one of the most critical components of the game.

    • Erik Lonnrot

      Did you read the last paragraph? “While winning a big faceoff at a key moment can certainly change the game, such an event is relatively rare, and the vast majority of faceoffs just don’t matter all that much in the grand scheme of things. Faceoffs can be important on a situation-by-situation basis, but their overall impact on the game appears to be highly overrated.”

      For every replay of a faceoff won in the last minute that leads to game winning goal I bet I could find 10 where it doesn’t. The whole point of the article is that while at an individual level in the big picture they don’t make much difference.

      • North Van Halen

        Until it does….Literally face offs don’t matter util they do. You can show me 1000’s of face offs that resulted in nothing but I’ve seen and been in enough situations where they do. Just because the events of the previous 100 meant nothing doesn’t mean then next one means nothing. Goals/high events situations happen at no set statistical time. You can’t know when the face off is important to win until it happens. 95% of face offs have absolutely no effect on a game doesn’t mean winning a key face off at a key time isn’t extraordinarily important.
        You could argue to me winning faceoffs at a higher rate on average doesn’t correlate to wins, thats very true. Telling me having a guy that wins key draws at key times doesn’t help you win, I say bull puckey.

        • Dirk22

          “You could argue to me winning faceoffs at a higher rate on average doesn’t correlate to wins, thats very true.” – NVH

          That’s exactly what the article is about.

          • North Van Halen

            On average not never and those are 2 completely different concepts. Being a 40 % faceoff won’t matter for 98% of draws. But on 2 it will.
            Talent will always be most important to winning, Saying that a faceoff percentage has a correlation on winning is ridiculous, over time talent prevails. But if 2 teams are dead even then it’s the little things that will make a difference. Having a good faceoff guy when it matters is crucial and could be the difference between winning and losing. Faceoffs don’t matter until they do, and when they do, it doesn’t matter that the last 100 were irrelevant.

          • North Van Halen

            sorry that should say faceoff % correlate to effect on winning % not just winning, of course it helps win games but obvioulsy the biggest factor in winning is talent..

        • Erik Lonnrot

          I think we may actually agree here. Individual faceoffs can totally matter but higher level face off stats don’t (to any significant degree at least). Stephen said that in his conclusion.

          The difference between a good face off team and a bad one is a few possessions per game. Sometimes one of those possessions is going to be a game winning goal one way or the other, most times it won’t.

    • Erik Lonnrot

      For reference, the top faceoff team at the moment (by a lot) is Philadelphia who are 30th in the league and the worst is Washington (7th). Tampa Bay are the best team in the league and have a decidedly average 50.3% win rate, only 0.4% better than last place Ottawa. Vegas is half a point worse than Ottawa at 49.4%.

  • DJ_44

    Similar simplistic stats have been trotted out for years. Just because this it the only data you know how to extract from Corsica.hockey does not make it valid analysis.

    What is a players SAT% when they win a draw? Powerplay percentage when they win a draw? PK% after a successful zone clear immediately following a draw. GF/GA down by one/up by one when the win or lose a draw.

    Everything has context; you are missing the relevance of faceoff by looking at the wrong metrics to assess their worth.

  • Killer Marmot

    It might be highly situational. Winning a faceoff in your own zone when you’re shorthanded is a considerable advantage as you have a good chance of icing the puck, running down the clock by as much as 20 s (with a little forechecking). If you’re at even strength, however, then you’re still trapped in your own zone and must try to engineer a controlled exit. Not such a big advantage.

  • Defenceman Factory

    This was a silly debate each time it has been brought up and still is. Looking for correlations between 2% differences in FO winning percentages and wins over a season will never show up. Are there any winning teams that average under 40% in FO wins? Any losing teams who win more than 60% of face offs?

    It is always better to start with the puck. Always. It might not be enough to help a weak team many games but every time it provides an opportunity to have an impact on the game.

    This discussion invariably degenerates into a debate about how much should strong face off guys get paid and whether being good at face offs is important enough to offset other glaring weaknesses. The answer to that question will never change. How bad is the player in other areas of the game and how good are they at face offs? Obviously winning 52% of face offs isn’t enough to offset the inability to skate, shoot, pass or defend but if a player wins 60% of draws they may not need to be quite as good in other areas to warrant a roster spot.

    • Erik Lonnrot

      The reason there are not winning teams below 40% or losing teams above 60% is that there are never any teams in those ranges. In the last 10 years only three teams have been below 45% and none of them were below 44%. Over the same period only three teams have been above 55% and only one of those was above 56%.

      The worst FO team in the league at this point is Washington at 46.5%. Only one team (Philly) is above 53%. Obviously if we were looking at teams going 70% or 30% it would make a difference just as a driver of possession, but most years there is less than a 10% spread between the best and worst teams. Being a good faceoff team or a bad faceoff team is only going to mean a couple of possessions more or less on average.

      • Holly Wood

        I think it really matters which faceoffs you win or lose. Winning draws in the neutral zone looks good on paper. Lose one on the P.K. late in a period, different story

  • TD

    I agree with NVH’s post about faceoffs, specifically referring to Boeser’s goal a couple weeks ago where he dropped back to the blue line and took the shot off the faceoff win. That goal doesn’t happen without the win. Many won draws in the defensive zone may similarly prevent goals.

    The reason the statistics may not reflect the value is that the difference on teams is minimal. The Canucks are at 47.9% this year, meaning the faceoffs are basically split in half with each team winning roughly half. The significance becomes more pronounced as the split becomes more dramatic. For example, what would the results be for a team with a 90% win rate? Every team believes/knows there is a value so they all practice the skill and the difference between teams in minimal. It is also only one factor in a 60 minute game, where faceoffs are just one of a huge number of factors that lead to an outcome. With how even teams are on the draw and how many other factors can affect the results in a game, I can see where winning draws does not have a statistical significance.

    As far as pk faceoff %, I would expect most teams to have a lower pk faceoff % as many faceoff wins are the result of winger wins and teams have one less player to assist with the win when short handed.

    • Erik Lonnrot

      I posted something very similar above so please forgive the repetition. It is uncommon for any team to go too far from 50% in the face off circle over a full season. This season there are only two teams that fall outside the 47-53% range. I wouldn’t call that “statistics not reflecting the value of faceoffs”. I’d say they do a pretty good job at reflecting the value of faceoffs within the range that actually exists. That doesn’t mean they’re worthless, every extra possession is a good thing and coaches should absolutely send out their best faceoff man to take important draws, but in the big picture the extra couple of possessions you get per game out of being a good faceoff team don’t make much of a difference towards how your team does over a whole season.

  • Goon

    While I have some sympathy for the conclusion, this isn’t exactly rigorous analysis. Looking at the Canucks’ record when Elias Pettersson is winning/losing faceoffs, or when Jay Beagle is in or out of the lineup is cute, but it doesn’t actually tell you anything useful about anything. I have a 0% occurrence of being mauled by a tiger when I keep this rock in my pocket. You’re just throwing out a bunch of random stuff that loosely correlates and drawing conclusions from it.

  • LACANUCK

    This is the main difference between fancy stars and eye test people. Fancy stats look at the macro while eye test people notice situational ie the micro.

    Sure overall face off wins don’t matter until your team is in a tough situation and needs control of the puck. Just like possessing the puck gives a team a better chance to win, getting control on PP, PK or up or down 1 at the end of the game gives you a chance to win. Facts are facts, stats can be manipulated

    • Erik Lonnrot

      Are you agreeing with Stephen’s conclusion that sometimes a specific faceoff win matters but overall they don’t make much difference? It’s hard to tell.

      I’d be curious to see faceoff stats for high importance faceoffs, say non-neutral zone when the game is tied or within one goal in the third. I wonder if there are players/teams who are better at those than their totals suggest. I could see a stat like that tracking better to wins than faceoffs as a whole. It would also probably look a lot more like the stats of the teams best faceoff men who get sent out in those situations.

  • Holly Wood

    I ask this question. How many draws can you lose in your own end before you become a winger? The thing is draws in the neutral zone are irrelevant, draws in the offensive zone have some impact. Draws in your own end, especially late in a period Carry so much weight,they should be a stat themselves

    • Erik Lonnrot

      Do you think there’s a big difference in most centres’ FO% in their own zone vs their total? It doesn’t seem like a skill that would vary that much based on zone aside from maybe a few real chokers/clutch players.

      • Charlie y

        I think that would be an interesting question to ask someone who plays center. At first glance, you would think a faceoff is a faceoff and the player would do more or less the same thing to win regardless of OZ or DZ. It kind of makes me think of football though, and how much different plays are in the red zone. A DZ faceoff has a much more compact formation behind the center than OZ. I don’t know how many faceoff techniques each center has in his pocket, but I imagine there would be situational variations that each player may be more likely use, and their proficiency at each one would vary. Also, the side they would want to win towards might change in each zone – toward or away from the net. Winning toward your off side must have much worse %.

  • Kanuckhotep

    Even prominent Indian mathematic theorist Srinivasa Ramanujan would have a hard time figuring out the deluge of statistical data that envelopes modern hockey. Intellectuals more often than not are dorks IMO arguing one side against another without any clear or concise conclusion(s) as to what they are discussing. Hockey should be very simple. The more face offs you win the more likely you’ll win the game. Dry numbers are a boring lot and how about some human interest articles where you don’t have to be Ramanujan to figure out numerical data which is, frankly , quite boring and uninteresting. Just sayin’.

  • Kootenaydude

    Does a team that has pulled the goalie have a better chance of scoring a goal when they win the faceoff in the offensive zone? Does a team that is on the power play. Have a better chance of scoring a goal when they win the faceoff in the offensive zone? I would have to say yes.

  • SCF2022

    Winning a faceoff is valuable for the same reason that controlled zone exits and zone entries are important – because you maintain PUCK POSSESSION!!!

    I am a big believer in analytics in sports and in hockey but I don’t understand why the writers on CanucksArmy continually try to dismiss the value of a faceoff win while simultaneously pushing the value of stats such as controlled zone exits which also can’t be directly correlated to “wins” in the manner that the author is using for his argument against faceoff wins. We intuitively understand that controlled zone exits are valuable because we maintain puck possession – why can’t the same intuitive conclusion be drawn for winning faceoffs?

    • kablebike

      Agreed. A team has either has possession or is chasing/waiting for it. Hard to score without the puck. Presumably chasing it causes more fatigue since it moves faster than a player.

      • truthseeker

        Depends on the energy exerted. If you are playing positional hockey and forcing the offensive team into the zones you want, you’re not going full throttle on most shifts. The canucks have won many games this year playing counter attack hockey. The penguins and caps are recent examples of teams with, frankly, terrible possession stats that have had good regular seasons and won cups. Playing a counter attack strategy is every bit as valid and successful as winning by dominating the possession battle.

        I will admit I much rather prefer a team that controls play than one that responds. It’s much nicer to watch. But in the end it’s the results that matter.

  • truthseeker

    Seems like the article is arguing a bit of a strawman. I don’t think any of the “face-offs matter” crowd have ever argued it from the perspective of every single face off and how that correlates to winning. (as a few others above have alluded to)

    And for a website so obsessed with corsi (although that seems to be dying off here as I don’t see it used as much by the writers) I have to wonder why the argument isn’t made more based on the correlation of face-off wins/loses in relation to shots on goal when only accounting for offensive or defensive zone face-offs 5v5, PP, PK. Because I’m pretty sure that every single “face-offs matter” person is only considering them in the end zones in terms of whether or not the might have an effect on the game.

    So…how often does a face-off win in the offensive zone lead to a goal or scoring chance? How often does a loss in the D zone lead to a scoring chance against? Measure it by goals, or shots or possession that leads to a shot…whatever. Seems to me that would be the information worth knowing. And that also seems to me to be an area where shot metrics might actually mean something.

  • Qualicum Wayne

    The problem with the stats used by the author is that they are not statistically valid. You can’t compare different teams from different years and expect to draw any meaningful conclusions about the impact of faceoff wins. There are far too many variables when comparing different teams. Unless one can look at individual lines (same 5 players) and then measure advance stats for the impact on fancy stats for the 30 seconds following the faceoff, then one isn’t looking at meaningful data. I realize why this hasn’t been done, it would be a tremendous amount of work. However, since the stats are not valid, neither are the conclusions drawn.

  • Goon

    The general consensus from both Stephan and the readers seems to be that faceoffs don’t matter most of the time, but occasionally there is an important faceoff that can really matter, and those occasional faceoffs tend to get lost in the noise of all the insignificant faceoff wins and losses that happen throughout the game.

    I can get down with that.

    What I think the more important and interesting question is, though, is “does faceoff skill matter when evaluating a player’s skill set”. If you’ve got one player who goes 54% in the faceoff circle, which would put him well above average, and a player who goes 47%, marginally below average, is that difference going to win you or lose you any games?

    In a situation where the faceoff pro is a mediocre player in other respects, and the subpar faceoff guy is above average in other respects, who do you put out for an important defensive zone faceoff? I’d argue you put the better player out there – nineteen times out of twenty you’re going to get the same outcome from those two players on the draw, and having a better player overall on the ice is, IMO, going to be much more important than coming up with the puck out of the faceoff one extra time out of twenty draws. If you lose the faceoff, and both those guys are going to lose the faceoff close to half the time, you want the guy out there who’s going to be more likely to recover the puck. If you win the faceoff, and both guys are going to win it close to half the time, you want the guy out there who’s going to be more of an offensive threat with the puck in his possession. Having a “face-off specialist” on the team who isn’t particularly good in other respects is not going to be very helpful for the team in the long-run.

    Of course, if you find a player like Manny Malhotra who is defensively dynamite, passable offensively until his eye injury, and wins faceoffs at close to 60% in his prime, that’s a guy who makes sense as a “specialist”. But there are only a couple guys like that in the league at any given time.

  • DogBreath

    This debate is bizarre. Someone who wins 9/20 FO’s is often considered a poor centreman; someone winning 10/20 FO’s is considered an average; and winning 11/20 FO’s is considered elite at FO’s and therefore highly regarded. Seems close to 50:50 to me with some winning slightly more and others slightly less. And yes, I do get the value of winning a FO in your own end when killing a penalty, protecting the lead etc. That said, the whole topic seems very over-analyzed.