Which Flyers defensemen improve their teammates the most?

In the last day or two, I’ve been posting some of the findings from the with-or-without-you scripts I put together to look at how pairs of players perform together.

Yesterday I questioned the reported lines and in particular the reported decision to put Nodl with Carter and Giroux. Nodl didn’t end up playing on that line, so either the Inquirer reported the lines wrong or Lavy changed them before the game started (maybe he saw my article).

Earlier today I showed why the Carcillo-Richards pairing that was used a fair amount last year has been abandoned.

Now it’s time to take a closer look at the defensemen, after the jump.

With-or-without-you analysis on the defensemen is simpler in some sense because there are way fewer pairings to look at, but it’s also tougher because the pairings don’t change as much. Here are all of the pairings that played together for at least 200 shot attempts (roughly 100 minutes of ice time):

I’ll walk through the first line to make it clear how to read the table. When Carle and Meszaros play together, their Corsi is 54.8% — that is, the Flyers get 54.8% of the shot attempts when those two are both on the ice. (I like Corsi because it correlates well with zone time, puck possession, and goal scoring.) That 54.8% number is 4.1% above Carle’s overall average and 4.7% above Meszaros’ average. A few assorted observations from this data:

• When Pronger is out, we go from Carle-Pronger (50.4%) to Carle-7th Dman (42.9%). Yet another way of saying we need Pronger back.
• Meszaros gets at least a 4/5 here. Carle-Mez is at 54.8% vs Carle-Pronger at 50.4%. O’Donnell-Mez is at 47.5% vs O’Donnell-Coburn at 41.9%.
  • Carle-Mez was used primarily against Staal vs Pittsburgh and against both Bergeron and Krejci vs Boston in the two games I checked. So this isn’t a quality of competition thing where players got easier assignments when they were with Mez.
  • Meszaros has one of the lowest offensive zone start percentages of any defenseman, so it isn’t a zone start effect either. He seems to be genuinely driving the play.
• I guess it’s what you’d expect from the 6th defenseman on a team with five very good ones, but it’s startling to see how O’Donnell drags down everyone he plays with.

To that last point, here’s a chart like the ones I’ve made recently, showing how each player improves (or doesn’t) when they play with O’Donnell. (Minor clarification: this table compares Corsi with SOD to Corsi without him, whereas the above table compares Corsi with SOD to overall total Corsi, so the numbers aren’t quite the same.) It’s, erm, not good.

I guess I didn’t need a chart to tell me SOD wasn’t as good as the top five defensemen, but man, that red makes my eyes hurt. At least Carcillo likes playing with the old guy.

And finally, I found it interesting that it appears that certain lines have done better with certain defensemen behind them. It looks like the Hartnell-Briere-Leino line loves having Pronger backing them up:

Whereas the Carter-Giroux line really prefers having Timonen behind them:

This isn’t just a small sample size issue either — Timonen has spent more ice time with Giroux and Carter than any other forwards, whereas Hartnell, Briere, and Leino rank 1, 2, and 4 on Pronger’s most-played-with list. We’ve said all year that both HBL and Pronger have been deployed in primarily offensive situations while CG_ and Timonen have been in defensive situations. Now we see that not only is that true, but that it was the correct usage.

So to sum up, WOWY analysis gave us another look at Pronger’s impact, made a case for Mez as being quite impactful, showed just how big the dropoff to O’Donnell is, and gave us a neat look into how the top defensemen have been strategically deployed.