• A Faceoff Age Curve

    by  • November 13, 2013 • Hockey • 1 Comment

    I’ve never seen anyone address this, so I thought that a quickie post on the topic might be helpful. There’s a general sense, I think, that young players tend to struggle in the faceoff circle but I’ve never seen anything specific about how they improve or what the curve looks like.

    I gathered the data for all players who took 100 ES faceoffs in consecutive seasons and coded it by age. This gave me groups of players who played at, say, 18 and 19. I used ES faceoffs only because it’s harder to win faceoffs SH and easier to win them on the PP, as a general rule. I then calculated the average faceoff winning percentage for each group of players, at 18 and 19, 19 and 20, etc. This gives me a an ES winning percentage for each age group. Here’s what we see.

    Screen Shot 2013-11-13 at 2.35.59 PM

    Gagner didn’t take a lot of draws in his rookie year but he was pretty bad. His next two seasons were actually ok, relative to what we generally see from young players – he took a big jump in year two. He took another big jump in year three, actually being above the average for twenty year old players at that point. That was Pat Quinn’s year.

    From there though, it’s a disaster. He collapsed in Renney’s first year, although he had a big improvement in Renney’s second season. He was terrible again last year and is off to a really poor start in 2013-14. As I’ve said, I’m not a big believer that faceoffs matter but there comes to a point where they’re so bad that they do start to hurt. I think that this was a big part of the Gagner line’s difficulties in Chicago.

    RNH is young enough that I wouldn’t be too worked up about him, particularly given that he had shoulder issues last year and is coming off surgery this year. That said, a 40% centre means a lot of time chasing the puck, which costs you points.

    The fact that Gagner’s experienced some success at the past, relatively speaking, makes me wonder what was different when he was succeeding versus now. It’s probably worth someone on the Oilers trying to figure that out. If he was heavily sheltered, taking faceoffs against nobodies in favourable spots, which certainly seems possible given that his two good seasons he had Shawn Horcoff in front of him both seasons and Eric Belanger in front of him in 2011-12. When the preferred faceoff centres missed big chunks of time, as they did in 2010-11 and 2012-13, Gagner’s numbers nosedived.

    It’s something that has to be figured out though. If Gagner can’t do better than 43% unless he’s being heavily sheltered by other centres taking on the other team’s best, then he’s not a top two centre. It’s as simple as that. He has other fine qualities as a hockey player and can still be a valuable winger, one hopes, but he’s not a top two centre in those circumstances. The challenge for the Oilers is to sort this out and do it relatively quickly because it’s hurting them, particularly given RNH’s (more defensible) struggles to win faceoffs.

    Email Tyler Dellow at tyler@mc79hockey.com


    One Response to A Faceoff Age Curve

    1. November 14, 2013 at

      Great work to get all this data together. Very logical & interesting in own right. Neat to see how faceoff percentage improves deep into 30′s. It’s a very specific skill where experience trumps all, apparently. I recall reading a Bill James Baseball Abstract where he discussed career curves, and how specific skill sets follow different curves. As I recall (from reading it 25-30 years ago), ability to draw bases on balls was a late, late curve with players basically improving throughout their careers — becoming more patient, knowing the umpires strike zones and other tendencies, having a “book” on the opposing pitcher, on and on. Seems a very reasonable analogy to me.

      While gross faceoff percentages are a good rough indicator, one refinement that is readily accessible at source is special teams faceoffs. IIRC the team with the mad advantage wins ~55% of draws (haven’t checked this for awhile, don’t imagine it’s changed). So a given player’s individual percentages might reflect changes in his role. For example, Gagner’s faceoff numbers dropped last season at least in small part due to his gig on the penalty kill. Not a lot, because they were still only a few percent of his overall draws, but he got mashed on the PK (14/41=34%) & those 41 draws would be enough to shave a little off his overall rate.

      Here are Sam’s # faceoffs by season: EV + PP + SH (FO%)

      2007-08: 223 + 75 + 1 (41.4%)
      2008-09: 560 + 126 + 14 (42.0%)
      2009-10: 589 + 119 + 1 (47.4%)
      2010-11: 771 + 140 + 24 (43.8%)
      2011-12: 613 + 84 + 4 (47.6%)
      2013-13: 587 + 113 + 41 (43.9%)
      2013-14: 91 + 8 + 4 (43.7%)

      If you prorate it, not a gigantic factor, but a factor. Likely things like individual match-ups (like Toews … ugh) would have a greater influence, but that’s way harder to tease out of the data I would imagine.

    Leave a Reply

    Your email address will not be published. Required fields are marked *