Week 9 Snapshot: Scratching, Benching, Winning

Bruce Bennett - Backstrom

“I’m an anteater!” “I, also, am an anteater!” (Photo: Bruce Bennett)

Some time in the last seven days, Adam Oates became unsatisfied with his lineup. Perhaps not winning a game in regulation since November 17th was the reason why; no matter. On the 23rd, Oates scratched Martin Erat, who had requested a trade, from the Toronto Maple Leafs game. In Erat’s spot was Eric Fehr, who played every game this week and recorded three points. On the 29th, Oates restored Jay Beagle, who had not played since October 16th but apparently is in awesome shape, to active service. And with Brooks Laich still laid up with a lower-body injury on Saturday, Martin Erat got his sweater back and reclaimed the spot where he started the season: the 4th line.

That’s more moving and shaking than we’ve seen all season, and our numbers reflect it. It seems that the Capitals are now motivated to make changes, and some of those changes have already paid off. The losing streak is over, and the even-strength Capitals are stronger than we’ve seen all year.

These are the numbers as of noon on Sunday, December 1st. My sample doesn’t include power play or penalty kill situations– just 5-on-5 play while the score is within one goal in the first two periods and tied in the third. That eliminates the effects of blowouts and comebacks, when teams are protecting a lead or desperate to get a goal back. Stats of note are highlighted in a powderpuff pink and discussed below.

See previous snapshots: week 1, week 2, week 3, week 4, week 5, week 6, week 7, week 8

Forwards

Player Pos GP TOI GF GA SA% Sh% Sv% PDO ZS%
Brooks Laich C 25 194.5 4 7 40.3% 5.6% 93.8% 99.4 47.6%
Marcus Johansson C 27 227 8 9 48.2% 7.3% 92.7% 100 57.7%
Mikhail Grabovski C 27 216.7 12 11 47.2% 11.1% 91.5% 102.6 50.6%
Nicklas Backstrom C 27 237.9 10 8 49.3% 7.4% 94.5% 101.8 62.1%
Alex Ovechkin L 25 228.9 7 8 49.3% 5.3% 94.4% 99.7 63.9%
Jason Chimera L 27 209.9 12 14 47.6% 10.9% 88% 98.9 46%
Eric Fehr R 18 129.9 4 10 51.4% 6.9% 83.9% 90.8 49.3%
Joel Ward R 27 199.7 9 13 50.5% 8.7% 87.3% 95.9 48.1%
Martin Erat R 24 162.1 6 6 49.6% 8.1% 93.1% 101.2 52.8%
Tom Wilson R 27 116.9 4 4 43.6% 8.7% 93.3% 102 55.8%
Troy Brouwer R 27 208.8 6 8 40.1% 8.7% 93.4% 102.1 50.2%
Michael Latta C 16 63.1 3 4 48% 10.7% 87.9% 98.6 48.4%
Aaron Volpatti L 20 85.7 2 3 43% 5.6% 93.2% 98.7 52.8%
Jay Beagle R 7 39.5 0 2 40.4% 0% 88.9% 88.9 58.8%

Defense

Player Pos GP TOI GF GA SA% Sh% Sv% PDO ZS%
Nate Schmidt D 23 237.6 9 9 51.6% 7.3% 92.7% 100 56.9%
Alexander Urbom D 20 145.1 6 8 42.1% 10.3% 91% 101.4 50.8%
John Carlson D 27 268.7 11 14 44.1% 8.9% 91.4% 100.4 53.3%
Karl Alzner D 27 275.7 12 11 46% 9.5% 92.7% 102.2 51.7%
Mike Green D 24 269.6 8 11 54.1% 5.8% 91.4% 97.2 57.5%
Steve Oleksy D 20 162.3 10 8 46% 12.3% 91.1% 103.5 48.4%
Dmitry Orlov D 1 13.7 0 0 36.8% 0% 100% 100 33.3%
Tyson Strachan D 8 66.5 0 2 33.3% 0% 95.8% 95.8 38.3%

Observations

  • The Caps’ team-level even-strength shot-attempt percentage in close games (i.e. puck possession) rebounded from 45.91% to 47.22%, the best we’ve seen in a snapshot so far. There’s a bunch of reason why this may have happened: weaker opponents, better effort, better players, better combinations. Despite the team’s inability to win in regulation lately, they’re becoming stronger at even strength– which is a good thing since the special teams to which we ascribed their early wins have all but dried up.
  • After a stint at 3C, Brooks Laich left the lineup after Wednesday’s loss to Ottawa. Laich missed more than 80% of last season due to injury, so we have reason to be worried for him. On the ice, however, Laich’s absence may prove to be a good thing. It makes a reunion of the Black Hole Line (Laich, Brouwer, random unsuspecting victim) an impossibility and allows a guy like Eric Fehr (51.4% SA% to Brooks’ 40.3%) to get a sweater.
  • Fehr, by the way, is still plagued by some really crummy goaltending, a team-low 83.9% Sv% inside our sample. That’s unfortunate, but not necessarily a reflection of Fehr’s defensive abilities. Fehr had two goals and an assist this week. He’s a top six player. Tell me he’s not.
  • On the flip side is Troy Brouwer, who was on for three goals-against in our sample this week and still enjoys one of the highest on-ice save percentages on the team: a stout 93.4%. Brouwer’s possession stats (SA%) ticked up ever-so-slightly this week, suggesting that simply being apart from Brooks Laich may not be enough to solve what’s ailing Troy.
  • While we’re on the topic of lucky and unlucky save percentages, the Caps’ top line (Ovechkin, Backstrom, usually Johansson) has enjoyed great goaltending (94.4%, 94.5% Sv%). While Alex Ovechkin leads the league in individual shot attempts, he’s also allowing a crazy ton towards his own net. I don’t recall Ovechkin ever being an exemplary two-way forward, but now that the renaissance is in full swing, I’d like to see more attention paid towards breaking up D-zone time.
  • Last week I said this about Marcus Johansson: “[while he] seems to be enjoying the high esteem of Caps fandom  lately, [Johansson] still needs to shoot more– not only to get himself some goals, but also to make the Caps’ top line a truly multi-dimensional threat that is harder to defend.” Johansson had eleven shots on goal this week, and the percentage of shot attempts going the right way (SA%) jumped up from 44.3% last week to 48.2% this week. Very, very encouraging.
  • Alex Urbom earned a benching and a scratch after allowing Danny Briere to harangue Braden Holtby on Friday (~50 seconds into the video below). Urbom has been the Caps’ weakest defender, and– sad to say– the team is stronger while he’s eating nachos in the press box.

  • For the record, here is a player usage chart for the Capitals defense (minimum 10 games). The y-axis shows how tough their competition is (higher = hard), the x-axis shows zone starts (right = more starts in the O-zone), the bubble color tells you if the team is stronger or weaker when that player is on the ice, and the bubble size tells you by how much.

Player Usage Charts

  • With Urbom out, the Capitals brought in Dmitry Orlov– friend to truckers along US-15; EZ Pass wielder; first-time caller, long-time listener of sports talk radio. Orlov got a shot on goal on Saturday night, and while he kept the puck out of the net, he was stuck in his own zone for most of the night (36.8 SA%)– though it’s possible his D partner, Tyson Strachan (not my favorite player), might be a factor there. No one thought Orlov would immediately fix the defense, but I do think he’s an improvement over Urbom. And while most of the D corps improved this week (except for Carlson, who got smoked by Tavares on Saturday), there is still work to be done, and I think that work will require George McPhee’s intervention.
  • But maybe Strachan shouldn’t be taking 60% of his non-neutral zone starts (ZS%) in the defensive zone? Just a thought?
  • The fourth line had been impressing me a bunch this season, but not so much lately. Their goal on Saturday was a great anecdote of an energy line getting rewarded for tough play, but the larger pattern isn’t so sunny. Aaron Volpatti and Tom Wilson had been relative bright spots in possession, but in the last five games they’ve dragged down the team’s possession between 13 and 30% (Corsi Relative). I don’t know what changed, but let’s keep an eye on how the grinders grind this upcoming week.

Glossary

  • GP: Games played
  • TOI: Time on ice
  • GF: Capitals goals for which the player was on the ice
  • GA: Opponent goals for which the player was on the ice
  • SA%: Percentage of shot attempts (from both teams) that went towards the opponent’s net, excluding blocked shots
  • Sh%: Capitals’ shooting percentage while the player was on the ice
  • Sv%: Capitals’ goalie save percentage while the player was on the ice
  • PDO: The sum of Sh% and Sv%, a number that regresses closely to 100 in larger samples; a proxy for luck, in a sense– i.e. high ≈ lucky
  • ZS%: The share of shifts the player started in the offensive zone, excluding neutral-zone starts; data not limited to close games.

Big ups to the excellent ExtraSkater.com for furnishing these stats.

No related posts.

Tagged with:
 
  • Chip

    Well…we won two games in a row…time to go back to not playing Erat, Fehr or Beagle.

  • Jack Conness

    I love these weekly snapshots. You guys do a great job with all of this information.

    I think we all agree…we all want to see Erat-Grabo-Fehr manning the 2nd line. That would be great to see.

    My biggest concern is Mike Green. I don’t know if he is still hurt but he has been a total bum all season. He can’t play any defense and he has had a countless number of turnovers on offense. He has to figure it out. He is hurting more than he is helping.

    I mentioned this before but I have fallen in love with Grabo. He is MUCH better than Riberio. He is great with the puck, sets up his guys, and he always plays at 100%. I always see him finish every check. The guy is great. He needs an extension.

    Braden Holtby continues to wow everyone. If only he had a D in front of him, he would be the best goalie in the league. He just played back-to-back nights with stellar goaltending. He and Ovi have to be co-MVPs. The Caps wouldn’t win a game without these two.

    I hope the Caps continue improving. I hope Fehr, Beagle, Erat, and Orlov can stay in the lineup but Oates’ obsession with Laich will change that.

  • Andrew Stafford

    Do you still think D will be a big issue once Hillen and Erskine return?

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    I’ve been worried too, but Green’s been making steady improvement. Blue line is possession, red is PDO.

    Green is totally unlucky for not having scored on 62 shots, but I’m hoping he’ll turn it around. http://hkref.com/tiny/BBF3j

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    Great Q. I don’t think there’s a timeline on either coming back, but I’d be worried if GMGM or Oates think they’re better off putting those guys back in the lineup rather than making a roster move. The team desperately needs a new top-4 defenseman– and maybe a 5th as well.

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    Also, thanks for the kind words!

  • Owen Johnson

    This is totally off topic. But December 14th is ugly sweater day at my work. If money weren’t an issue, I would totes spring for and Islanders alternate sweater. Thems ugly.

  • Jack Conness

    Unlucky? I feel like you know the numbers better than I do but when I watch, I feel like when he shoots the puck isn’t even close to the next. I always think it ends up way wide. Just from what I’ve seen.

  • Josh Carey

    Good work again, guys!

    Has anyone else been really surprised by how Beagle is playing? Sure he might not have the best possession in the world, but I really like him on the third line getting a few more minutes. He’s hard to knock off the puck and seems to do well with Chimmer and Ward although I think Chimmer and Ward would make me look good with the way they’re playing.

    I hope Orlov’s possession numbers are in part due to him starting in the defensive zone a lot. I thought he played well and I think if Binky goes in instead of Strachan, after a few games together, our D could pick it up.

    Not necessarily related to this post, but did anyone else notice on the Isles’ second goal yesterday who was in front battling (or something) trying to prevent them from scoring? Hopefully we see more of that out of Ovi, except with him getting out of the way of slappers.

    Holtby is amazing.

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    Green averages 8.5% shooting over his career– above average for a defender. On 62 shots we’d expect him to score 5.27 goals, but it’s not unheard-of for him to be running this cold.

    He is shooting more than he had in recent years, so maybe he’s putting more low-quality shots on? I dunno.

    Either way, I bet he’ll get the scoring touch back, though I doubt he’ll jump all the way back up to 8.5% on the year.

  • Bench23

    I love this blog.

    I hate this guy: Tyson Strachan.

    No, I don’t have lots of fancy graphs, but I do have my butt-in-a-seat-for-every-home-game data that says this guy just doesn’t play smart hockey. Lots of turnovers, even more bad passing and poor decision making skills.

    As an aside, Peter et al, do you guys have any commentary on the following:

    1) The Dale Hunter dump-n-chase game being back, AGAIN?
    2) The trademarked Oates line change (at the worst possible times, always)
    3) How awesome have Beagle/Fehr been of late?

  • Graham Dumas

    Plus ones for Grabo’s performance and your concerns about Green. Peter, the numbers on that graph below are definitely encouraging, but what about Green’s apparent propensity to make really big mistakes on ice, resulting in goals against? He seems awfully high in that category (though not as high as Carlson, who, I think, has been playing very well). That pinch last night on the PP seemed a bit amateurish to me (obviously an amateur).

  • Graham Dumas

    I know these numbers are about non-special teams play, but I’ve been wondering: when do we start worrying about the PP? Seems like Blaine may have some work ahead of him, no? Or is Oates really correct when he says the only problem with the PP is the lack of shots from dudes not named Alex?

  • Jack Conness

    I feel like they are low quality shots IMO. I hope he gets out this awful funk, offensively and defensively, he is in.

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    They’re still top-10 in power play shots(http://www.extraskater.com/teams/on-ice?sort=sf&sit=pp&type=rate ) , but they’re dropping– due in part to Ovi getting shadowed. They’ll have to make a tactical adjustment I think.

    And Oates isn’t entirely correct– Carlson, Backstrom, and Green all shoot a bunch. Always could use more!

  • Graham Dumas

    I wonder how far Oates et al. will be willing to go in making that adjustment. One thing that comes to mind when I compare our play to other teams is that our PP seems very static.

    A static PP seems to have both advantages and drawbacks: it’s easier to catch the D out of position with a pass than it is with foot-movement (i.e. you can get the puck to the open man faster than you can open yourself or your linemates up with movement); but it doesn’t work the D as hard, and doesn’t force them out of position as badly as a cycle game does.

    Clearly we do cycle sometimes, but the preferred setup is the static 1-3-1, as far as I can tell. Perhaps it’s time to add a bit more player-movement into the mix, without abandoning the 1-3-1 that’s worked so well thus far. I think that might help to free up the diamond guy for a quick stuff- or tap-in shot.

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    The 1-3-1 has taken over the NHL in the last 6 months– I’d bet more than 20 teams use it. The rumor had been that Oates had a whole bunch of adjustments he would make once other teams got used to his version of it. I don’t think we’ve seen those adjustments yet, but I’m eager to.

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    1. That would be zone entry data, which the NHL doesn’t collect, and I don’t think anyone has it for the Hunter era Caps.
    2. Not a clue.
    3. Fehr has been awesome. Still too early to say for Beags, but I like him!

  • brian!

    Out of curiosity, is there any way to determine via fancystats et al whether the shallower nets have had any impact on the game as a whole? I’d like to think the extra room has aided playmaking or somewthing.

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    You’re absolutely right that he’s made some tactical goofs, and he deserves some blame for his low sv% (though perhaps not a lot). I’m thinking that those problems can be fixed. I don’t think he’ll ever be 2009 Green again, but he can still be real good. I hope.

  • Graham Dumas

    Fingers crossed as to this and your comment below. LET’S @#$!& GO CAPS!

  • JenniferH

    I agree about Braden Holtby; I like that it seems he finally seems to be getting more widespread credit for his stellar goaltending play. He’s pretty dang awesome. And would be getting even better stats (and his are pretty dang good) and quite a few more shutouts with better D in front of him.

    Obviously we wouldn’t be where we are without Ovi, but right after him, there’s Holtby. Without him, we wouldn’t be where we are either.

  • JenniferH

    I noticed that with Ovi too and was very impressed. Seeing Ovi make that kind of defensive play was surprising and heartening. Go Ovi!

    And, yes, I agree… Holtby is amazing.

  • Brendan Maltese

    Thanks for defining “when the score is close” this week!

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    Haha, thanks. Always improving, that’s my motto.

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    Maybe. I dunno. It’d be hard to isolate its effect from stuff like hybrid icing and edicts to refs on how to call plays that we’re not privy to.

  • Anon

    Partial collected by Flyers blog Broad Street Hockey, posted at Japers http://www.japersrink.com/2013/2/25/4025552/bruce-boudreau-dale-hunter-and-zone-entries

  • Owen

    if only erat wanted to stay/oates would do this:

    90-19-8
    16-84-10 (dream line)
    25-43-42
    24-83-46

    27-74
    52-88
    61-81

  • TheShameOfFife

    Hi big fan of the blog and of your coverage of the stats, but I have a question/observation about the use of possession stats. I completely understand the usefulness of SA% in showing how the team/players/lines are doing overall, and that having more possession equals more wins. However, as the Caps in general, and certain players in particular, are struggling in this area, what I really want to know is why & how – that is why are certain players/lines struggling and how can this be improved. Looking solely at the possession stats I used to just think ‘player X should be scratched/demoted and replaced with player Y as he has better possession stats’. Essentially that player X is bad and player Y is good. That is a very simplistic interpretation of how a team operates though, and doesn’t offer much hope for team improvement if the majority of your players are on the wrong side of 50%. Basically, do you think it’s possible to use stats to understand the WHY of possession?

  • Sausaged

    I just realized how much of a smoke Shania Twain is….Thanks for opening my eyes.

  • http://www.russianmachineneverbreaks.com/ Peter Hassett

    Yeah, absolutely.

    I like to think of all player data as a stack. At the top of the stack are goals and assists– basic performative stats. They’re awesome retrospectively but they don’t predict or tell us why something happened. Below that stack, we used to just have adjectives– stuff that sounds like a scouting report (“a hard skater who isn’t afraid to go to the crease”) and qualitative assessments of how those goals and assists came to be.

    After that, we had zone time, or zone time as proxied by shot attempts– i.e. puck possession stats. Now the question is: what’s below that stack? What informs possession in the same way that possession informs goals and assists?

    For the most part, it’s still adjectival and anecdotal: “Alex Ovechkin carries the puck in, uses the defender as a screen, usually his shot gets blocked.” or “Jason Chimera cycles the puck, Joel Ward fights a board battle, clears it to Grabovski in the slot, who gets a shot off.”

    Zone entries are another way of doing it, as is the “Passes” stat (a function of shooting percentage and primary assists that isn’t popular yet), but like all things (including Corsi/Fenwick), it’s just a transitional form.

    Eventually we’ll have data on the level of touches– who had the puck, when, where, and what happened next. Then we’ll know that player X handles the puck 1.2x more often and with 1.5x the duration of player Y, and player X completes more successful passes that lead to shots from dangerous areas. All that and a lot more. I could go on for hours about what could be done with that data on the pattern level.

    That data is *already* manifested in possession stats, but without granularity or high resolution. We don’t know *why* a player is good at possession (or at least, we can’t prove it scientifically yet), so we rely on the higher, more general stack of data.

    The touch data will happen eventually. It’s happened in other sports, but it requires league-mandated motion-tracking data-collection that will be made public.

    Great Q.

  • pixiestix

    i don’t understand the anteater caption. Will someone please explain that to me?

  • http://www.russianmachineneverbreaks.com/ Peter Hassett
  • pixiestix

    thanks! I’ll get it from now on.