Playing in the Offensive Zone

In the last several weeks, Derek has used something called "zone-start adjusted Corsi" to help us in our evaluation of players and coaches. As with so many of the helpful statistics kicking around, the methodology goes back to (I believe) Vic Ferrari and JLikens. Both are very smart cookies to whom we're all indebted for our better understanding of the game of hockey. Anyroad, when it comes to the Oilers, these adjustments haven't generally been very flattering with regard to the Oilers' young players. That got me thinking about whether or not this adjustment is fair for players at the extremes, and I decided that it might be helpful to compare these young Oilers to other players who have played in a similar role in one of the last five seasons.

So what counts as a similar role? I started with position, so we'll only be looking at forwards here. I then moved on to zone-start, and plucked all of the players with at least 60% of their end-zone starts in the offensive zone in a given season (min. of 40 games except for 2011-12 which had a min. of 20 games). I quickly realized that included a whole lot of plugs and fourth liners, and that's not the role that Jordan Eberle, Taylor Hall, and Ryan Nugent-Hopkins have been playing. As such, I also made sure that all qualifying players received at least 12 minutes of five-on-five ice time per game. That makes for guys playing in the top six or top nine depending on how the coach runs his bench. With that criteria I was left with 61 qualifying player seasons, and that seemed like a good place to stop. So are all of these guys battered by the adjustment?

Nope. The average shows a positive adjusted Corsi, so even after letting the air out, these players are generally able to move the puck forward. Among the trio of Oilers, just Taylor Hall comes out above the average. Eberle and Nugent-Hopkins, by contrast, have a long way to go to catch up with their peers.

Another interesting note is the presence of Tom Renney on this list, both with the Oilers this year, and with the Rangers in 2007-08 and 2008-09. If Tom Renney is here for the long run, this strategy of pushing certain players with highly favorable zone starts is probably something we can expect to see going forward.

The other column I decided to include is points/60 because there are some interesting results. Evgeni Malkin, for example, got decimated in terms of possession in 2008-09, but scored a boatload of points. Johan Franzen was the opposite in 2007-08, doing very well on the shot clock but with nothing to show for it on the scoreboard. After running a quick correlation between adjusted Corsi/60 and points/60, I found that, in these 61 player seasons, there's actually a very slight negative correlation (-0.09). Another quick calculation reveals that the correlation between offensive zone starting percentage and points/60 (again, in just these 61 player seasons) is robustly positive (0.43).

This is odd. The most likely explanation, to me, is that there's a sample size issue. Because of the criteria I decided on, the zone start range isn't even that large, so I wasn't expecting such a strong correlation, and we know that there is a strong correlation between Corsi and zone time, so it doesn't make much sense to have a negative correlation between scoring (which usually takes place in the offensive zone!) and Corsi. With a larger sample, I expect that we'd see Corsi and points come together given that the majority of goals are scored in the offensive zone. Still, it's an interesting little twist.

Of course, it's an interesting little twist that saves Eberle more than it saves Nugent-Hopkins, who's below-average in that category too (despite having the 12th easiest zone start percentage even in this sample). He's obviously being compared to some pretty darn good offensive players, but it just goes to show that, as fun as he's been to watch so far this season, there's still plenty of room for him to improve.