Tuesday, September 22, 2009

Season Preview: Pittsburgh Penguins

In terms of regular season results, the 2008-09 Penguins do not compare favorably to the majority of recent Stanley Cup champions. In fact, outside of the 2005-06 Hurricanes, the only teams that had similarly mediocre numbers were, perhaps fittingly, the Lemieux Penguins of the early 90s.

For reasons articulated in this post, the Lemieux Penguins were better than their regular season results would suggest. I would argue that the same was also true of last year’s team.

The Penguins 2008-09 season can essentially be divided into two distinct parts: that which preceded the coaching change, and that which followed it. As has been discussed on this blog before, the team’s performance improved substantially following the coaching change. While it would be inaccurate to award all of the credit for the turnaround to Bylsma – as the acquisitions of Guerin and Kunitz, as well as the return of Gonchar from injury, surely had an impact as well – it’s nevertheless apparent that the team markedly improved under his tenure.

It’s difficult to overstate how much better the Pens were after the coaching change. The shot for and against numbers are especially telling. For example, consider the difference in Corsi differential per game between the Penguins under Bylsma (2.8) and Therrien (-3.39). That’s a considerable difference. To put those numbers in context, that’s equivalent to the difference between the 4th best team (the Capitals, at 3.05 per game) and the 2nd worst team (Atlanta, at -3.21 per game) last season. In other words, the Penguins basically went from being one of the worst teams in the league in terms of territorial play at EV to one of the best.

The difference is even more remarkable if one considers that the intra-season split-half reliability for EV shot differential at the team level is huge. For example, among NHL teams last season, the correlation between EV shot differential in the first half of the season and EV shot differential in the second half season was 0.82. Suffice it to say that few teams experience large swings in shot differential from one part of the season to another. Fewer still experience swings of the magnitude described above.

In terms of forecasting, few teams had underlying numbers as good or better than the Penguins did down the stretch and, given that their current roster is similar in composition to last year’s team, I think it's sensible to assume that the same will be true this season. The results may not be present to the same degree -- the Penguins EV shooting percentage with the score tied was almost twice as good as that of their opponents during the stretch in question, which is simply not viable in the long run. Nonetheless, they’re probably the best team in the East at this point in time and I suspect that that they'll finish as a top four seed in the conference. While I don’t necessarily expect them to win the cup again – as they’re not obviously better than any of the other potential contenders, a repeat is certainly conceivable.

No comments: