As we continue to evaluate Michel Therrien as a coach, and with thanks to Derek Zona and Rob Vollman for the concept: here is a player usage chart for the Pittsburgh Penguins in 2007-2008 -- the only season fully coached by Therrien for which the venerable, terrifying Gabriel Desjardins's invaluable behindthenet.ca resource has data:
(Click on the image for a full-sized version.)
This is every Penguin with at least 30 games played. I decided to leave Gill, Hossa and Dupuis in, even though most of this data would be from their Toronto and Atlanta games.
The bubble size is relative Corsi; blue is positive, red is negative, and the size of the bubble indicates by how much. X-Axis is incresing offensive zone start, Y-axis is increasing relative quality of competition, so the toughest minutes are on the upper left.
Immediately we see that this is not the usage chart of someone who pays no attention to this stuff. Therrien seems to have carefully handed out zone starts, relying heavily on Staal and Gonchar for defensive-zone faceoffs against tough opposition and giving both Malkin and Crosby, as well as then-rookie Kristopher Letang, some pretty offensive minutes. In fact there really seems to be a sizable gap between his offensive lines and his defensive lines. He is not quite Alain Vigneault, but he certainly seems to be someone who gives his lines some definite roles at even-strength.
We also note Therrien had the good sense to heavily shelter Laraque. Unsurprisingly, this still resulted in a giant red bubble. This is a bad sign in the sense that while Therrien clearly recognized Laraque's total uselessness as a hockey player, he still believed his extracurricular skills had enough value that he dressed him for 71 games and gave him 7:29 minutes a game. Therrien, it seems, believes in goons, which is certainly a strike against him.
Unrelated to Therrien: That big blue Dupuis bubble in tough-minute-land brings another illustration of how much Waddell was fleeced in the Hossa trade.
So in terms of personnel usage, this is actually very good news. Therrien certainly looks very aware of zone starts and picks his personnel for such accordingly. He doesn't seem to be quite as concerned about matchup difficulty, except for giving the toughest to Staal and Gonchar, but then again, his "exploitation" centres were guys named Malkin and Crosby -- guys who don't precisely need to be handled with as much care as David Desharnais, despite all the respect I have for the little guy. There's a sense here that Therrien understands personnel usage and has a good idea of who to send out.
Now for the bad news. They are actually not in the bubbles themselves, but hidden in the scales and the backing data. First to note is the scale of the X-axis. Excluding Dupuis and Laraque, the range of offensive zone starts go from 37.5% to 52.7%. So what we're seeing is not a Vigneault-style spread, what we have is a team that had a very high proportion of its faceoffs in the defensive zone. This is a sure sign of a team that spends a lot of time in its own zone.
The straight Corsi numbers certainly bear that out. This chart uses Corsi Rel, but if it used straight Corsi, you'd see all the bubbles in red, going from tiny to giant balloon size, with one, small blue bubble, belonging to some guy named Sidney Crosby. Outside of the best player in hockey, everyone on the Pens with 30+ games was a negative possession player.
Now I'm looking at this usage and possession data and I'm certainly not encouraged, but I also can't help but think about the 2009-2010 Habs. That team was awful in puck possession, but despite this, it was entirely visible that Martin had a definite plan about who did what on his roster (once he got familiar with his club). The next year, Martin's team was a very strong possession club -- a turnaround actually comparable to the difference the Pens experienced by going from Therrien to Bylsma. A turnaround that we've credited in large part to a much improved supporting cast; the point is, a weak possession club does not necessarily imply a weak possession coach.
Could the 2007-2008 Pens have been less talented than we gave them credit for? Given the almost-immediate turnaround following the coaching change the next year, I have my doubts. Nonetheless, once one moves past the big three centers and Gonchar, there are a lot of very unremarkable players on that chart. I'd be very curious to see Pittsburgh's post-deadline numbers, once they added the likes of Dupuis, Hossa, and Gill.
Also, while I'm bringing this up, it's important to remember that 2007-2008 was Therrien's second full year with the Penguins (he'd coached them a total of 133 regular-season games before then), and the 2008-2009 Pens were still horrible until Bylsma took over. Unlike Martin, Therrien never managed to make an outshooting team out of the Penguins.
Ultimately, however, it seems Therrien does know how to handle the crucial aspect of situational player usage. That's certainly a good sign (even if it's a bit in the "at least he didn't fail coaching 101" sense), but the fact remains that his teams were getting constantly outshot despite this. Whether this points to a weak tactical ability, or a team that had weak overall talent due to a subpar supporting cast, despite some awfully big names, remains to be seen.
We'll just have to wait and see. It's not like we have much choice!