A Radical Suggestion for the Boston Bruins Power-Play

Right now, the Boston Bruins of the National Hockey League, are statistically the best 5-on-5 team in the league, and given that hockey is a 5-on-5 game, one would assume that the Bruins of Boston would be one of the best teams in the league. At 3^rd in the Eastern Conference, first in the Northeast Division, and almost a lock to make the playoffs, they're doing alright. But. They've had their struggles. Two, three, and four game losing streaks. Stretches where they only get the overtime point. And they can't seem to beat Montreal (one likely first round playoff opponent) or Buffalo (another likely first round playoff opponent.) As statistically the best team, for the most common situation in the sport, you would think they would be dominant. But even though they're good, they're not as good as they could be, and I for one, and not assuming a Stanley Cup this year.

So, if they're the best 5-on-5 team in professional hockey, why aren't they the best team in professional hockey? Basically, their power-play sucks. It's not the worst in the league, but it is plenty bad enough to bring their overall game down. They can go weeks it seems without scoring a goal when they have five players on the ice and the other team only has four. Their biggest trade this season was for defenseman Thomas Kaberle, primarily to improve their power-play. They've rotated many players through the power-play units and they've tried different strategies, but nothing's worked. They grabbed a 5-on-3 goal last week, (and that was a relief) but I can't remember the last power-play goal before that.

In watching their power-play, all season, what might be the most frustrating, (head-coach Claude Julien must be going out of his mind) is that, as a power-play unit, it doesn't look that bad. Passes are usually crisp. Shots are pretty good. Rebounds are produced. Their entrances into the attacking zone could be better, but they're not atrocious. But, if they want to win the hands-down best trophy in sports, they'll need at least an average power-play to do it. So, here's my radical suggestions to improve the Boston Bruins power-play.

Don't run a power-play. That's right. Instead of two power-play units, roll the same lines used during even strength. Don't run any 5-on-4 specific plays. Don't do anything different from even strength. Dump and chase with an aggressive forecheck. If a defender backs off the half-wall, don't try to set up a diagonal cross-ice path; attack the net as you would if it were 5-on-5. Instead of high-low passing to set up a shot from the point with men in front, set up the low zone forwards cycle.

The different plays power-plays run are designed to take advantage of the extra man. The team on the power-play can take more risks, they can hold the puck longer to look for passing lanes, and they almost always have more room on the periphery to make passes. Power-plays are designed to take advantage of these, well, advantages.

But right now, the Bruins aren't. But they score more and allow fewer goals than any other team when playing 5-on-5. It stands to reason that if they play the same way when they are on the power-play, they'll start scoring more goals. Even if they don't, there's still time before the playoffs to try something else. It could act almost like a reset button, completely wiping clean whatever their power-play is now so they can get something more effective ready for the playoffs, whatever it may look like.

In a somewhat related vein, Jack Edwards says a lot of ridiculous stuff on the air, but he's right when he points out the inadequacy of the power-play statistics. Power-play and penalty kill statistics are calculated as basic percentages, with the higher the better. The problem is all power-plays and penalty kills count the same in the statistic. Killing a two-minute penalty counts the same in the calculation as killing a five-minute or a 30-second penalty. It's not an entirely useless statistic, but it really only gives a sense of a team's quality one way or the other.

My thought for a more accurate statistic: Goals per ten minutes. It's pretty simple, the amount of time killing penalties is added up, the amount of goals scored during that time, and, then reduce the fraction. The more G/10s (or something, ask Bill James) the worse the penalty kill and the better the power-play. It's a simplification that would, I think, reflect more of the complexities of the events it attempts to describe, which is what the best stats do. There. Two problems solved. What's next? Concussions? No way. That's an entire other essay.