A couple weeks ago, I made a post with a couple graphs showing how the players on the Hurricanes were used last season in terms of zone starts and quality of competition. Charts like that are excellent for seeing how coaches use their players but they don’t give you the best idea of how they performed against them because raw corsi and corsi relative numbers do not take context into the situation. Hockey statisticians all over the Internet are looking for ways to fix this and one system that has been introduced is Adjusted Corsi or “Corsi corrected for shift location.”
I believe that I stated this in a few of my articles, but where a player starts his shifts will have a huge impact on his corsi number because if a player starts most of his shifts in his own zone, he’s likely to be on ice for more shots and his corsi number will suffer from it even if he plays well defensively. To correct this, the writer at Objective NHL determined that one offensive zone start is worth 0.8 in corsi based on Vic Ferrari’s study that an offensive zone start is equal to 0.6 in Fenwick and Fenwick is just Corsi without blocked shots factored into it.* If you look at the graphs that I made for the Hurricanes, you’ll notice that there were a lot of players who had mostly defensive zone starts and I attempted to show their performance by color coding the bubbles, but corsi relative numbers alone doesn’t reflect how well they performed here so I thought this project would be good to do. I used the formula in the Objective NHL article posted above to determine these numbers and found their zone start percentages and corsi numbers at Behind The Net’s stat site.
*I prefer to use Corsi because I feel that blocked shots should be accounted for here. If a player in his own zone allows the opposing team to get a shot away then he should be penalized for it. However, shot blocking as a skill should be taken into account when signing players.
DZ = Defensive zone starts, OZ = Offensive Zone Starts, OZ% = OZ/(DZ+OZ), Corsi/60 = Corsi On number from BTN, Adj Corsi/60 = (Corsi + (DZ – OZ)* 0.8), Corsi Rel QC = Corsi Relative to Quality of Competition
I color coded this chart to show how how each player performed relative to their zone starts and to show what kind of competition they faced. The green means their adjusted corsi number was higher than their regular corsi and red means that it was lower. If I left a player’s cell white, it means that their adjusted corsi showed little difference or their sample size was too small for me to make a judgement.
In the Corsi Rel QoC column, the deep red means they faced tough competition, light red indicates slightly above average competition, white means average, light blue is below average and deep blue is weak competition.
- Brandon Sutter looks a lot better here than he did in the scoring chance graph. He was used primarily as a defensive player and didn’t play as bad as his terrible as his low corsi and scoring chance rating suggest. However, the fact that he wasn’t able to break even shows that he wasn’t as good offensively as he was in 2009-10 and I’m not sure if he’ll ever get to that level again if he stays in his current role.
- If Chad LaRose had just a little more of an offensive skill set then I would love to see him on the first line…but he’s not that kind of player. Everything I’ve seen shows he’s a damn solid top-9 player, though.
- The line of Jussi Jokinen, Jeff Skinner and Tuomo Ruutu were given more offensive zone starts compared to most of the team and performed only decently when you consider that they were sheltered more than other players. However, this is a perfect line to assign easier draws and competition to because they are much more dangerous in the other teams zone than in their own end and it’s possible that they aren’t good enough defensively to play with Staal and receive the tough competition that he draws. That’s just a theory, though. I think a big problem is that the team was outshot so badly as a whole last year that Paul Maurice couldn’t shelter certain players as much as he would have liked to.
- Erik Cole and Cory Stillman played frequently on Staal’s line according to Dobber Hockey and as you can see, they faced the toughest competition as well (although there’s a good chance Stillman’s data is skewed from his time in FLA). Stillman looked slightly better when factoring in zone starts while Cole stayed the same. Let’s see what our first line can do with some wingers who can drive possession more.
- Drayson Bowman was given a lot of defensive zone starts against extremely weak competition while Jerome Samson received the same treatment except he got mostly offensive zone starts. Now it become clear why the latter had a better scoring chance rate. The Canes appeared to have three lines that got tough zone starts but it would be nice to have at least one more get sheltered minutes and use either Samson or Bowman in that situation. Or we could do that with Zach Boychuk instead. Yeah, that sounds like a good idea.
Only one regular defenseman started the majority of his shifts in the offensive zone, sort of reinforces what I said earlier about the team being outplayed too much at even strength for Maurice to give certain players easier minutes.
- Gleason and Allen look a lot better here than they did going by scoring chances and the numbers they put up here actually are not bad at all when you consider the tough assignments that they received. When thinking of defense pairings for next season, pairing one of them with Joni Pitkanen would be something to consider. Gleason played most of his minutes with Corvo last year and they weren’t a bad pairing when you consider the competition they faced and pairing a stay-at-home defenseman like one of them with an offense-first kind of guy like Pitkanen would be a nice balance. However, this would mean Pitkanen would have to take on tougher competition and he didn’t fare too well against it last year. Another option is putting Allen and Gleason together to give the team a “shut-down” pair and free up some easier zone starts for Kaberle and Pitkanen but I tihink that may cause some problems later down the road because Gleaosn and Allen are not your ideal “shut-down pair” even if I feel the two take more flack from fans than they deserve.
- Derek Joslin and Jay Harrison are going to be fighting for playing time next season and Harrison may have the inside track because he had more time with the team and played well but he saw the weakest competition among defensemen by a longshot. It really shows that he’s a good 5th-6th defenseman and nothing more than that. Joslin, on the other hand, played slightly tougher minutes but had easier zone starts and he seems to contribute more at both ends of the ice when considering his scoring chance data so this could be an interesting decision. I feel that Joslin’s upside is higher but I’d like to see how Harrison does against some tougher forwards.
- The addition of Tomas Kaberle probably means that Pitkanen will have to take on some tougher minutes this year because Kaberle faced similar competition as him in Boston and Toronto and Corvo isn’t here anymore. All I can is hope for the best with that because Pitkanen didn’t exactly play that great last season when you consider that he was sheltered a bit compared to other defensemen. This is why I think my earlier idea of pairing him with Allen or Gleason may work because his most frequent partner last year was youngster Jamie McBain, who was a rookie and is more of puck-mover than a physical, defensive player. Giving Pitkanen a more defensive-minded partner to help off-set some of the problems in his own zone may be a good call even if he has to take on tougher competition than he’s used to.
- I have a feeling that McBain may get top-4 minutes again next season and I’m not 100% sure that he should when you see that he only faced middling competition and didn’t exactly outplay it. He didn’t play horribly against his competition but I would rather have someone who can outplay his assignments rather than do enough to get by if that makes sense.
There’s also the concept of Balanced Corsi Numbers, which also take zone starts into consideration and I will quickly have the numbers for those in a day or two and see how they compare.