Network News

X My Profile
View More Activity
On Twitter: kcarrera and PostSports  |  Facebook  |  E-mail alerts: Sports and Redskins  |  RSS
Posted at 7:54 AM ET, 01/24/2011

Statistical analysis: Who is the Capitals' shutdown pair?

By Neil Greenberg
Statistical analysis

Washington's defense has been a pleasant surprise this season, allowing only 2.54 goals against (entering Saturday's game) after surrendering 2.77 goals per game last year. So when it comes to being the Capitals' "shutdown pair," who is doing most of the heavy lifting?

The defensive pairings have seen far less tinkering than the forward line combinations, so we are able to limit it to the six pairs who have spent a bulk of time together both on the penalty kill and during even strength. The definition of a "shutdown pair" is murky, so for purposes of this article I will define it as follows:

  1. They must play against the opposition's best lines as measured by Corsi, which is a proxy for puck possession. A higher Corsi number means those skaters are seeing more pucks directed at the opposition's net.
  2. They must see a majority of scoring chances go in Washington's favor when on the ice.
  3. They must keep the other team from creating scoring chances and from scoring off those chances.
  4. They must perform well on the penalty kill together.

First we'll look at even strength, and as one would expect, John Carlson and Karl Alzner lead the pairing in time on ice together, followed by Mike Green and Jeff Schultz.

Defensive pair EV TOI Scoring chance% Scoring chances against per 20 min Goals against per 20 min Avg Corsi of competition
Alzner-Carlson 483.5 54.9% 4.88 0.58 0.985
Green-Schultz 345.4 47.8% 5.50 0.87 1.183
Green-Hannan 171.9 53.3% 5.70 0.58 0.426
Carlson-Erskine 113.5 52.8% 5.99 0.53 -0.464
Erskine-Poti 97.4 44.0% 5.75 0.62 -0.079
Poti-Schultz 86.8 45.8% 5.99 1.15 2.548

As most Caps fans know, it's the youngsters (Carlson and Alzner) who are stealing the show. With more than half the scoring chances going in Washington's favor, they still manage a minuscule half a goal against per 20 minutes - against very stiff competition.

Now let's look at the penalty kill:

Dpair PK TOI PK scoring chances against per 2 min PK Goals against per 2 min
Green-Schultz 53.4 1.01 0.22
Alzner-Carlson 51.4 0.89 0.23
Green-Hannan 28.2 1.28 0.00
Carlson-Erskine 19.6 1.53 0.20
Poti-Schultz 15.2 1.19 0.26
Erskine-Poti 11.8 0.85 0.17

The pairings of Green/Schultz and Carlzner play almost half of the penalty-kill minutes for Washington, but Carlzner allows fewer chances and goals against per two minutes of penalty-kill time -- getting the nod as best pair when skating 4-on-5 or 3-on-5.

So there you have it: Carlson and Alzner see more minutes against better competition and keep those lines off the scoreboard during even strength and on the penalty kill. What was thought to be a liability coming into the season has thus far been the team's best defensive shutdown pair.

By Neil Greenberg  | January 24, 2011; 7:54 AM ET
Categories:  John Carlson, Karl Alzner, Statistical analysis  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Matt Hendricks on his highlight-reel breakaway against the Maple Leafs
Next: Boudreau doesn't expect injured Capitals back until after all star break


Seriously do we need yet another stat analysis for something as obvious as this??

I have a feeling that if Greenberg was doing a football analysis, he wouldn't have anything positive to say about Big Ben's performance last night by simply looking at his stats.

Sometimes eye ball tests give you more insights than this. No matter how hard these stat geeks try to make sports into science, they are much of an art than a science because you can't always measure all the intangibles needed to win.

Posted by: joek443 | January 24, 2011 8:15 AM | Report abuse

and this is just going to give vermontcaps more ammo as he kneels down to service Erskine.

Posted by: jmurray019 | January 24, 2011 8:24 AM | Report abuse

Poti/Schults might be allowing twice as many goals against, but the Corsi ratings of their opponents are MORE than twice the corsi ratings of Carlzer's opponents. So I think that it more than evens out.

So Green/Hannan allow ZERO goals on the PK, and Carlzner is a better combo?????????? Yes, Carlzner is allowing fewer scoring chances, so you could chalk the goals against up to Neulamov's (why not?) performance. But maybe Green and Hannan are allowing lower-quality scoring chances.

Posted by: robbie1299 | January 24, 2011 8:40 AM | Report abuse


Keep the team's pt. streak going w/DJK in the lineup tonite. The RAGS ARE GOING DOWN!!

Posted by: vermontcaps | January 24, 2011 8:43 AM | Report abuse

The analysis raises an interesting question as to whether Hannan or Schultz is a better partner for Green. It looks like the performance with Hannan is better, but against weaker competition. But I assume that the Hannan data includes the time when he was new to the team and still learning the system. As I recall he had a pretty weak +/- for a while, until he settled in. That "learning phase" could be split off from the rest of the data, but that might make the data set too small to draw conclusions.

The good news is it is reasonable to assume that Alzner and Carlson are only going to get better. They are still kids.

Posted by: zmega | January 24, 2011 8:57 AM | Report abuse

While we're at it... Can I get a statistical analysis of the correlation between points in the standings and games in which the Capitals outscore the opponents?


Posted by: GFisher1 | January 24, 2011 9:00 AM | Report abuse

Look, the Caps last two wins are against teams down in the standings, that's a fact. The other fact is the Capitals are keeping their GA down and Ovi and Backs are getting their scoring touch back. Ovi has points in his last nine games or so and I'm not sure how many Backs has banked lately. They are mostly assists for Backs, but his confidence with the puck is starting to show. Think about this. Ovi and Backs have gone through arguably the worst scoring droughts of their careers and Ovi currently sits 9th in scoring and Backs at 14th. These last 30+ games should be interesting. Semin will probably be back the 1st game after the All-Star break as well as Brads. Things are looking up.

Posted by: fanohock1 | January 24, 2011 9:07 AM | Report abuse

"Look, the Caps last two wins are against teams down in the standings,"


Last 3 :)

Ottawa stinks too.

The Caps' last three wins have been a veritable "who's who" of the worst teams in the East.

Posted by: VTDuffman | January 24, 2011 9:19 AM | Report abuse

this is just one of those kinda stats that dives way into one category which doesn't result into anything thats guaranteed

The only stat that should matter is the final score - so go out and trade for a vet goalie and/or pick one of your existing three! Win baby!

Posted by: Nats1924 | January 24, 2011 9:30 AM | Report abuse

you guys ought to lighten up on the stats guy. Everyone complained for years there wasn't enough coverage of Caps hockey here in Washington ( ie as there is in say Toronto ) Now we have multiple persons with new looks and you're still complaining. If you don't like the column, just move on.
Thanks for another angle Neil.

Posted by: billd2 | January 24, 2011 9:37 AM | Report abuse

I got $10 saying "bild2" is Neil lol

Posted by: Nats1924 | January 24, 2011 9:41 AM | Report abuse

This was in a recent article by Bruce Garrioch and a possible Mike Fisher trade. He was talking to two league executives.

-“I just don’t think they can get enough for him to warrant that kind of move,” said another executive. “That’s a drastic move and they’ve got to play (well) next year. They need to have some players.”

So, what would be enough to move the veteran Senator?

“You’d have to get two pretty good young players — and I mean prospects — and a top draft pick,” the executive said.-

Also, according to the article, Stephen Weiss is on the trade block. What do people think of him as a possible 2nd line center. He only makes $3.1M/yr also.

Posted by: sgm3 | January 24, 2011 9:51 AM | Report abuse

I guess I am the only one that sees that Ovi and Backs are starting to get their swagger back. Confidence, regardless of what level athletes compete, is second only to skill. Confidence coupled with skill is deadly. The 3rd and 4th liners along with better defense has kept the Capitals in the mix. The scorers starting to become scorers again is a good thing. I don't mind seeing the Caps grind out wins so they can begin to peak when we all want them to peak, right at the start of the real season. So many critics in here want to bash the Capitals because they aren't at the top of the standings when the overwhelming response prior to the season is that the season doesn't matter as long as the Caps qualify for the playoffs. They seem to be winning the "right way." The "right way" has turned some people off because personal stats are down and the "right way" is often very boring. At times I have been a little worried and still feel they need to turn the corner before I feel the playoffs are a lock but Ovi and Backs boost in confidence over the last couple of weeks sure makes me feel better. My kool-aid glass remains half full rather than half empty.

Posted by: fanohock1 | January 24, 2011 9:55 AM | Report abuse

I agree with billd2...lighten up. I find the statistical analysis interesting. It's nice to apply metrics to the game; it either confirms observations or in some cases, gives an objective measurement of strengths or weaknesses that are not being realized by observation.

Posted by: Blueline | January 24, 2011 9:57 AM | Report abuse

Maybe I don't understand the Corsi rating, but couldn't the reason the Corsi is so much higher for Poti/Schultz, or even for Green/Schultz in comparison to Carlzner, is that they aren't able to regain possession and clear the zone? Or is the Corsi rating independent of how the lines are faring during that particular game against the Caps?

Either way, agree this trying to answer a pretty obvious question.

Posted by: gmAndAguinness. | January 24, 2011 9:59 AM | Report abuse

I like the stats posts.

Even if something is "obvious," it's nice to have actual numbers behind it.

Posted by: VTDuffman | January 24, 2011 10:03 AM | Report abuse

I agree with billd2...lighten up. I find the statistical analysis interesting. It's nice to apply metrics to the game; it either confirms observations or in some cases, gives an objective measurement of strengths or weaknesses that are not being realized by observation.

Posted by: Blueline |

Agree completely. Some people hate it when statistics do not support their observations that they already convinced themselves were correct.

Posted by: sgm3 | January 24, 2011 10:04 AM | Report abuse

and this is just going to give vermontcaps more ammo as he kneels down to service Erskine.

Posted by: jmurray019


whatever it takes to keep the big guy going.

"Some people hate it when statistics do not support their observations that they already convinced themselves were correct."

Posted by: sgm3

statistics can definitely be valid in supporting an argument. Unfortunately, too many people rely solely on statistics to MAKE the argument.

Posted by: cstanton1 | January 24, 2011 10:23 AM | Report abuse

Inferential statistics need to be more rigorous than this sham job. Data collection needs to be as complete as possible, and of course, paper stats always drop a ton of information that actually happened physically on the ice. The analysis itself needs a confidence interval, significance test, etc.

Posted by: slurpeesarecool | January 24, 2011 10:31 AM | Report abuse

definitely also pleased with Alzner's progress. He's added an edge to his game. He finishes checks in the def zone that in the past he would simply choose not to engage a player. Had 2 really nice effective hits along the boards v the Leafs.

Even Schultz who wasn't credited with a hit played decently in front of his crease v the Leafs.

Things are always in flux but I wouldn't be surprised to find Schultz as the odd man out as time moves on. Last yr it looked like he was entrenched in the Caps plans based on his gaudy +/- totals. Now, as the young kids are improving he may be expendable in a trade to bring back another element. And Green's game has gotten better overall in the def zone (yeah i know he's good for one bad turnover a game but if he does the other stuff well i can live with it).

Posted by: cstanton1 | January 24, 2011 10:33 AM | Report abuse

Some things you can just tell relatively easily by just watching the game. I didnt need a whole bunch of random numbers to tell me what Neil detailed. If someone actually watches the game and pays attention to it, they can tell whats going on and who is facing who. But I guess some people just love numbers too much, they need them to support everything.

Posted by: ThePat | January 24, 2011 10:34 AM | Report abuse

i like Erskine / Hannan combo

Posted by: doughless | January 24, 2011 10:56 AM | Report abuse

Joe Reekie was a guy with nice numbers on the +/-

But I wouldn't want him on the ice at the end of a game holding onto a lead. He made some horrific turnovers ala Poti. Esp in playoff games. His early yrs with the Caps were decent, his later years not so much.

Posted by: cstanton1 | January 24, 2011 11:12 AM | Report abuse

The numbers have their limitations. They are not the end-all, be-all of knowledge.

Of course they won't be statistically significant, it is just analysis from this season.

But, let the man talk numbers. It gets the discussion going. That is all.

If you look at the numbers and expect to find the meaning of life, you will be disappointed.


Posted by: CF11555 | January 24, 2011 12:08 PM | Report abuse

we can all see this and we will see further improvement in the upcoming future. as bart scott likes to say "CANT WAIT!"

Posted by: skinfan2k | January 24, 2011 12:23 PM | Report abuse

After they started to get used to playing with each other I think Green and Hannan has been a really good pairing both 5 on 5 and on the PK. I don't like messing with defensive pairs that are working and even though Green/Schultz was good last season, I think they should stick with Green/Hannan instead of sticking Hannan on the 3rd pairing. If you didn't include Hannan's adjustment phase in these ratings (more representative of their expected play the rest of the year) I think their numbers would be very very good.

Posted by: Stu_c | January 24, 2011 12:40 PM | Report abuse

I said as much last night, but just talking out of my a**. Good to see the stats confirm it. Carlzner gets the job done.

Posted by: seth_malaguerra | January 24, 2011 1:31 PM | Report abuse

I agree, this is good stuff - well done Neil.

If the coaches are generally putting Carlson - Alzner out against the other team's top line [esp. at home with the last change] you can be pretty sure, that's who they see as our best defensive pairing. Bodes well as they're both very young and should continue to grow.

Posted by: Timbo_1 | January 24, 2011 1:59 PM | Report abuse

But I wouldn't want him on the ice at the end of a game holding onto a lead. He made some horrific turnovers ala Poti. Esp in playoff games.

Why do I feel Hannan could be inserted here come springtime - good defensively, solid shot-blocker but the outlet passes look like they could be a problem. Hopefully just part of the 'adjustment phase' however.

Posted by: Timbo_1 | January 24, 2011 2:03 PM | Report abuse

I have always said that Jeff Schultz is not a top 4 defenseman in the NHL. He has a lot of trouble clearing the puck out of the defensive zone and is prone to errant passes/turnovers in the neutral zone. He contributes zero in the offensive zone.

Posted by: RoyHobbs4 | January 24, 2011 2:06 PM | Report abuse

No argument here on that one!

Posted by: Timbo_1 | January 24, 2011 2:59 PM | Report abuse

The team isn't stretching the ice as much this year as last - so players are not out of their position as often. Players like Schultz do well when they get in to retrieve the puck b4 the forwards are in there...well this year most of the time the D is already back and not as tired with shorter shifts and all. I think in the right system with good shift lengths any team could generate these improved numbers - now as for our scoring......

Posted by: yesisaiditfirst | January 24, 2011 3:32 PM | Report abuse

Little help please. Want to make sure I am understanding all columns correctly. Scoring chance % is % of chances for the caps right? Not so sure about the last one. I understand what Corsi is but am not following this number. Does this # only apply to playing time against the caps? And he says a higher number means more pucks directed towards opponents net. but is it the defensemans competition or the opponents? And this means Poti/Schultz have played against competion that is vastly better than everyone else?

Posted by: jtf11 | January 24, 2011 9:57 PM | Report abuse

Post a Comment

We encourage users to analyze, comment on and even challenge's articles, blogs, reviews and multimedia features.

User reviews and comments that include profanity or personal attacks or other inappropriate comments or material will be removed from the site. Additionally, entries that are unsigned or contain "signatures" by someone other than the actual author will be removed. Finally, we will take steps to block users who violate any of our posting standards, terms of use or privacy policies or any other policies governing this site. Please review the full rules governing commentaries and discussions.

characters remaining

RSS Feed
Subscribe to The Post

© 2011 The Washington Post Company