• Oilers Sign Andrew Ference: Part I

    by Tyler Dellow • July 6, 2013 • Hockey • 10 Comments

    Elliotte Friedman’s 30 Thoughts column had a really interesting bit about players playing on their off-wing a few months back that I seem to have referenced an awful lot. Let’s do it again.

    15. Had a terrific conversation with Oates last week about his dislike for players lining up on their off-wing. He is 100 per cent against it and explains why with an MLB analogy. “You look back through the history of baseball and every shortstop throws right. Why?” “Because it’s too hard to make the throw left-handed,” is the reply. “Right,” he says. “How many plays won’t be made because a left-handed shortstop isn’t able to turn, get set and make the throw with strength or speed?” He believes the same theory applies to a winger on the wrong side.

    16. One play he uses to illustrate this theory is Alex Burrows’ 11-second overtime winner in Game 2 of the 2011 Stanley Cup Final.

    Watch Andrew Ference off the draw. “He takes two steps, is closed off and has no play,” Oates said. He then pointed out something very interesting. Look at Boston’s roster now: four left-shooting wings. Four right-shooting wings. Two right-shooting centres. Two left-shooting centres. Three right-shot defencemen. Three left-shot defencemen. (It’s true, you can look it up.) That’s going to be the blueprint in Washington.

    If you don’t want to watch the video, let me tell you this: the Bruins defence pairing is Chara on the left side and Ference on the right. Ference, of course, shoots left – this was a lefty/lefty pairing. Bergeron wins the draw to Ference, who quickly finds that he has no angle – as a lefty shooter, he carries the puck on the left side of his body, closer to the centre of the ice, which enables the Canuck forward to close him down more quickly. Ference makes a soft play with the puck, because he hasn’t been able to get close to the red line, there’s a neutral zone turnover and, before you can blink, Vancouver scores.

    That’s an anecdote, albeit an interesting one. What if we looked at how Chara’s done with all of his defence partners since 2007-08 in terms of Corsi%? We’ll throw in how those partners did away from Chara.

    That’s awfully interesting. With a left handed shot on the opposite side of the ice, the Bruins aren’t bad with Chara. With a right handed shot out there with him, the Bruins crucify teams, which is all the more impressive because Chara tends to play the other team’s best.

    This kind of ties into a point that Twitter’s @Woodguy55 and @67sound made about who Ference plays against: he hasn’t been matched up against the other team’s best players. The thing with that is that the options the Bruins have include a monster who is seven feet tall on a pair of skates, shoots left and one of the greatest defensive defencemen in the history of the NHL. I don’t know that losing time to that guy tells us a lot about Ference.

    Now slide your eyes over to the column on the far right side. That’s the difference between how the player did with Chara on the ice and without Chara. Of the eleven defencemen who’ve played at least 200 minutes with Chara over the past six years, only three of them did worse with Chara than they did when he wasn’t on the ice. All three of them were left handed shots.

    Why do I bring this up? Well, Derek Zona doesn’t like the Oilers signing of Andrew Ference. He wrote, amongst other things, this:

    Last season, he was a 4/5 defenseman, playing even up zonestarts and had the worst relative corsi on the defensive corps. Two seasons ago, he was a 4/5 defenseman, playing even up zonestarts and had the worst relative corsi on the defensive corps. Three seasons ago, he was a 5/6 defenseman, playing even up zonestarts and was 4/6 in relative corsi on the defensive corps. Stop me if you’ve heard this before: four seasons ago he was a 5/6 defenseman, playing even up zonestarts and was 5/7 in relative corsi on the defensive corps.

    Ference is an okay bottom-pairing defensemen in the Eastern Conference, and when he plays higher than that, he’s not good.

    Here’s the thing about hockey data: it’s not an end in and of itself. It’s a reasonably objective record of what took place. I think that it’s better than relying on observation and recollection alone because people are not particularly well evolved to consume vast quantities of data by watching and then make good decisions. That being said, there is no number you can just look at and say “This. This says it all.” You’ve got to question the context.

    With the Bruins, there seems to have been a bit of a unique context. You have a defenceman who’s one of the greatest defencemen ever who happens to be a left handed shot. He performs much better with a right handed shot defenceman on the ice. If you’re a right handed defenceman, you’ll get the Chara Bump to your numbers when you play with him. If you’re a lefty, a) you won’t get to play with him much and b) you might actually do better not playing with him.

    Therefore, I’m not convinced that RelCorsi, which I generally like to get a quick and dirty feel of things, is a good metric to use with the Bruins and Ference. I suspect it tends to overrate their right handed shots and underrate their non-Chara lefties. There’s a hockey reason to suspect that that’s true – see the Oates comments above – and there seems to be some data in support of it. If I was charged with evaluating this decision for a hockey team, there are other things I’d do to look into the hockey truism but there’s a pretty solid prima facie case that Ference had circumstances acting against him.

    As far as what sort of a role Ference played, I think that Derek may, as he occasionally does, have let his rhetoric get away with the facts. 2009-10 is kind of arguable – it sure looks to me like Ference played in the top four in the regular season and then lost his spot to Matt Hunwick in the playoffs. From then on though, I kind of think he was second pairing fixture. He played there most of the season in 2010-11, lost his spot to Sexy Deadline Acquisition Tomas Kaberle for a little bit and then was a clear second pairing guy in the playoffs, playing 4:30 a night more than Kaberle.

    In 2011-12, the Bruins were kind of a five defenceman team. In the regular season, they used Chara (25:00), Seidenberg (24:02), Boychuk (20:36), Ference (18:53), Corvo (18:47) and then Adam McQuaid (14:57). I kind of see three different tiers in there: Chara/Seidenberg, Boychuk/Ference/Corvo and then the rest. Come the playoffs, Ference was clearly a second pairing guy.

    This year was more of the same. There’s a clear separation between Ference, fourth in TOI/G amongst Bruins defenders at 19:29, and Dougie Hamilton, at 17:07. The playoffs had an even cleaner split between the Bruins’ second pairing and the third pairing guys. Ference was third in TOI, with Johnny Boychuk fourth at 23:55. Fifth was Matt Bartkowski, who played 19:46 a night in only seven games; Torey Krug played most of the playoffs and was below 16 minutes a night.

    All of which is to say that it’s probably most reasonable to characterize Ference as being a second pairing defender. When the games mattered most, he was the guy Claude Julien kept putting out there. He overcame guys who the Bruins acquired to be top four defenders (Kaberle) or who they see as having big futures in Boston (Hamilton). I’ve indicated some of the difficulties that I have with RelCorsi as a metric in his circumstances; there’s a more generalized problem when dealing with players from good teams in that a guy can be a contributor and have a poor RelCorsi because it’s a good team.

    The money on this deal isn’t bad – $3.25MM a year. If the money’s structured as Hockey Symposium’s speeds suggested, it’s even better in terms of being easier to move in the future. The only real issue I have with the deal is the length of it, which is something that I’ll touch on in my next post.

    Email Tyler Dellow at mc79hockey@gmail.com

    About Tyler Dellow

    10 Responses to Oilers Sign Andrew Ference: Part I

    1. July 6, 2013 at

      I’m not sure you can justify shrugging off the gap between Seidenberg and all the other L’s. The L is already a pretty small fraction of the time; you could be drawing conclusions based on the equivalent of “performance vs the Atlantic Division”, no?

    2. Tyler Dellow
      July 6, 2013 at

      If I was going to be making decisions based on this, there’d be a few other things I’d do.

      First, I’d want to know if this was a league wide trend, whether guys tend to perform better when they’re paired with an opposite hand defender – is it like a platoon advantage in baseball, which we know exists across all players?

      Second, I’d want to control for the situation that those guys play in – maybe Seidenberg/Chara is a d pairing the B’s like for the offensive zone.

      At present, I think there’s enough there (and some old timey hockey wisdom) that it warrants further examination. This IS an issue that’s testable with data; you’d just need a team that was committed to directing some resources to it.

      So, when I said:

      If you’re a right handed defenceman, you’ll get the Chara Bump to your numbers when you play with him. If you’re a lefty, a) you won’t get to play with him much and b) you might actually do better not playing with him.

      I qualified point b) for a reason. Point a is pretty undeniable though and Johnny Boychuk, good Edmonton boy though he might be, isn’t a right handed shooting Zdeno Chara.

    3. Woodguy
      July 6, 2013 at

      I posted this at LT’s today.

      I find it interesting that Ference’s play didn’t drop off w/ lefty Siedenberg.

      I wonder if that’s due to QC or perhaps with Siedenberg he played the left side? I don’t know which side he played with Sidenberg, but he certainly played right side w/ Chara

      “You can laugh about Andrew Ference all you want, and yesterday’s thread was rife with putdowns and piling on, but Andrew Ference is an actual NHL player.”

      This is true.

      I initially hated the deal given his possession stats vs. where he played in the line up.
      They aren’t great, but judging D on possession numbers can be tricky business. Corsi works well for helping define forwards, but less so for D as its the forwards that are the prime drivers of possession.

      So if a D plays behind the 4th line, their shot attempt differential takes a beating. If they play behind a checking line it can be the same. If they play behind a offensive line that gets fed good zone starts and soft opp, their numbers rock…etc.

      One way of trying to suss out possession ability of a D is to look at their WOWY w/ D partners only.

      This still isn’t great because often when they change partners, they change assignments.

      In Ference’s case, when he played with McQuaid, he was playing 3rd pairing and got the 4th line often. When he played with Boychuck or Seidenberg, its was 2nds and of course with Chara its firsts.

      Another thing that can impact the play, is whether or not the Dman is playing his “correct” side.
      Ference played the right side in the playoffs often, and that’s pretty damn tough on the first pass ability under pressure, as if you are near the boards, its usually a backhand

      Ference shoots left, as do Chara and Siedenberg.

      Since the Bruin’s Dcorps have been fairly stable for the last 3 years, we can look at 3 years of WOWY’s and see how Ference does with his D partners.

      Here’s his WOWY with D partners from 2010-2013:

      Partner – 5v5 TOI together- CF% together – Ference apart

      MCQUAID, ADAM 1101min 50.2 49.7
      BOYCHUK, JOHNNY 538min 47.7 50.4
      CHARA, ZDENO 452min 46.5 50.5
      SEIDENBERG, DENNIS 339min 49.4 49.9

      http://stats.hockeyanalysis.com/showplayer.php?pid=568&withagainst=true&season=2010-13&sit=5v5

      Its interesting that he’s a little better away from Boychuck and Chara.

      Why? I would think that his assignments get easier when away from those players.

      Where I was really wrong, was in calling him a 3rd pairing Dman.

      If we assume (correctly I think) that when he was with McQuaid it was 3rds, SIedenberg and Boychuck 2nds and Chara 1sts, we can break down his assignments over the last 3 years like this:

      3rds – 45%
      2nds -36%
      1sts – 19%

      The end result is that I think we can say that Ference was a competent 4/5 Dman for BOS.
      I don’t hate the deal as much as I did when I wasn’t looking at all the info I should have.

      I think he’s a Nick Schultz replacment.

      MacT mentioned him helping take some of Smid’s load against tough comp. I’m not sure he’s up to that task though, of if he is, how many years he could do it from 34-38.

      I think that the Oilers still need a 1LD, but I’m not sure they aren’t done given the amount of D on the roster.

      LHD – Smid, Ference, N.Schultz-Belov-Klef
      RHD- Petry, J.Shultz, Larsen, Potter

      I’m glad he’s building up the D depth, but the 1LD still isn’t there imo.

      I also think Nick Schultz is on his way out of town.

    4. speeds
      July 6, 2013 at

      Having had more time to mull the deal over, I’m still not a big fan of it, but I think I was underselling the expected quality of Ference’s play in first year or two of the deal. I might have looked at other options with the roster spot, and other ways to use the money, but that doesn’t mean that Ference for the first couple years would not help the Oilers.

      A couple thoughts:

      (1) Arranging D by left shot/right shot doesn’t account for the general quality of D. If Chara’s RH D partners happen to have been better defencemen than his LH D partners, that might explain the differential even ignoring whether the players shoot right or left. Is that the case here? I don’t really know. Also, as Matt notes, there’s a pretty significant difference between Ference’s numbers and Seidenberg’s Is that based on usage, the relative quality of Ference vs. Seidenberg? I don’t know either.

      (2) I agree that it’s not reasonable to hold it against Ference that he hasn’t been playing hard comp – that’s more than understandable given the situation. However, BOS is a team with deep with quality forwards. If Chara is taking the majority of the tough matchups, Ference is still playing with pretty good forwards against somewhat lesser opposition. What would be a reasonable expectation for him in that situation?

    5. Stevezie
      July 8, 2013 at

      Great article.

    6. Pingback: Puck Daddy’s 2013 NHL Free Agent Frenzy Report Card | YO Status ->YO Status ->

    7. TheOtherJohn
      July 9, 2013 at

      Ference’s contract was not frontt loaded whicjh is surprising and shows how completely out of his depth Rick Olczyk is in helping his inexperienced GM. Front loading this contract would not be a huge issue to Ference.

      The price is not, for me, an issue. It’s the term.

      I think the bigger issue with this contract will be how a slightly undersized combative D holds up at age 37 & 38.Impression is that older/slower big body D age better because their size does not go away. That impression may be completely wrong. Big bodies in the NBA age much much better than smaller athletic players. Do not think fact that heis a fitness nut will matter to extend his effectiveness.

      Love Oiler fans that say Ference is better than Whitney. Stupid comment. Suspect we have 3 guys in OKC that you could say the same about. He will have to be in our top 4,and think he will fill easily that roster spot but will love to see how Ference’ level of plays trends as he ages

      For all of the stories about him coming home (MSM really really love simplistic narratives) understand Ference does not live in Edm in summer. Nor did he bring the Stanley Cup here when the Bruins won in 2011

    8. Pingback: End-to-End Rush: We Interrupt Your Regularly Scheduled Programming | Bolt Statistics

    9. Pingback: The Edmonton Oilers still have too many ifs and maybes on defence | Edmonton Journal

    10. Pingback: Jesse Joensuu: The Bjässe |

    Leave a Reply

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