View Single Post
Old
06-14-2011, 11:22 PM
  #80
MathMan
Registered User
 
Join Date: Jan 2006
Posts: 17,025
vCash: 500
@Lafleurs Guy, I appreciate your attempt to bring goal-differential as a yardstick of team strength, but you need to understand how percentages relate to goal-based metrics.

Goal-based metrics are the confluence of three things: shots (for and against), shooting percentage, and save percentage. Simple enough: goals-for = shots-for x shooting percentage, and likewise goals-against = shots-against x (1 - save-percentage). Intuitively this makes sense, but those stats are not created equal.

1- shooting percentage is transient and not something teams have any real control over; it heavily regresses to league average over time, and teams have little to no ability to maintain a high or low score year over year.
2- goaltending is more sustainable than shooting percentage, as goaltending is a real skill, but it is also very volatile and heavily regresses to career average
3- generating/preventing shots is a much more sustainable skill. This corresponds to controlling the play, possessing the puck, and generating/preventing scoring chances.

This means that goal-differential, while normally a decent yardstick of team strength, is like all goal metrics subjects to the vagraries of chance and percentages -- and that extreme percentages will heavily skew goal metrics, in ways that don't represent actual team ability.

As it turns out, both Boston and Montreal had extreme percentages this year.

Consider the Boston club of last year. They were much like the Habs of this year, in that they had the worst 5-on-5 shooting percentage in the game. This created the illusion that they were extremely weak offensively, when in reality they were only unfortunate. They took an unusual swing this year in that they went right up to the 4th-best shooting percentage in the league, creating the illusion that they became an offensive juggernaut over a single offseason with essentially the same roster. In practice, this is not entirely untrue as they did generate significantly more chances/shots (at the expense of defense, as it turns out; Boston was a very high-event club at both ends). But the majority of their increase in goal-scoring has been due to changes in percentages.

Those percentages were the primary factor that turned Boston from the 28th 5-on-5 scoring club into the highest.

(One of the many reasons I don't see the Bruins as elite.)

Montreal, this year... is pretty much were Boston was last year. 2nd-worst shooting percentage in the league (comparable to last year's Boston; only New Jersey was worse). This created the illusion that they did not improve their 5-on-5 game, even though they both increased shots for and cut down on shots against at even-strength (their turnaround at both ends of the ice was nothing short of phenomenal).

Unfortunately, they couldn't buy a goal. And that led people who are obsessed with goal totals (and apparently rely entirely on stats and don't watch games ) to assume the Habs hadn't improved their offensive game, which is incorrect. The nature of the Habs' game has changed profoundly: instead of being a terrible possession club that wouldn't score because the other guys always had the puck and they couldn't generate chances, they became an excellent possession club that wouldn't score because even though they had the puck more often than not and would generate chances, they could not finish.

I think it's unlikely that the Habs' finishing talent took a sudden nosedive relative to last year wouldn't you say?

A swing like the Bruins had, going from a very low to a very high shooting percentage year over year is possible, but unlikely. It's much more likely that it'll return to average. Nevertheless, should the Habs' shooting percentage indeed return to the level it was last year (league average, slightly below even) they will be a top-third offensive club. Their improvement in puck possession and chance generation was that good.

MathMan is offline   Reply With Quote