I was puttering around on NHL.com, which has a pretty nice stats page actually, though due to the nature of the sport no stat page ever produced by anyone associated with the NHL will ever approach Baseball-Reference.com. Just looking at the numbers and the logical conclusions you can draw is an interesting exercise because of how often those conclusion defy common sense, both our own and those of the team/players/coaches/media.
"We have to get pucks to the net," "They outshot us badly, that's why we lost," "Man, the Leafs SUCK, they've been outshot 20-2 so far!" Each is a phrase we've heard endlessly and, in the case of the last one, will continue to enjoy hearing as long as that defence is in place. But is outshooting the other team a determining factor in winning hockey games? Common sense and anecdotal evidence support this, as this equation shows: more shots=more scoring chances=more goals, right? Plus, how many overtime games have you seen that were decided by a fluke goals, lucky deflections, and rebounds off scrums? Exactly: ALL of them. Yet if you look at team shooting percentage records, it's a little less clear.
Shooting percentage is a limited stat of course, it doesn't account for shot quality or frequency, but over the course of a season its reasonable to think shot quality will even out and that breakaway chances, counting for one shot, will even out with dump-in shots than happen to hit the net. So, assuming shot quality does even out over the course of the year, here's the top 5 winning percentages for teams outshooting their opponents, with their overall league points ranking in brackets, as of March 11/08:
1. Atlanta .750 (27)
2. Detroit .707 (1)
3. Dallas .667 (2)
4. Anaheim .667 (5)
5. Pittsburgh .654 (6)
So that's pretty representative actually, except for Atlanta who got lost and wandered to the top of the list. The Thrashers are actually third last in the NHL in terms of shots against average per game, so yeah, they should shoot ALL the time at anything that moves. What's interesting isn't the rankings, the list tends to more or less reflect actual league standings, but the actual winning percentages. So far, the mean winning percentage in the NHL this season for teams outshooting their opponents is .532, a coin toss in other words, with a median of .538 and a standard deviation of .108. That's a pretty small deviation, meaning that most teams are pretty damn close to the average winning percentage of .532. THAT means if your team outshoots the other teams, it's basically 50/50 if they'll win. What's amusing is the team with the worst record on this list, the Leafs of course. When outshooting their opponent, the Leafs have a winning percentage of .333. When outshot, it's .552! What the hell?! What a great team. Not at hockey or scoring or playing defense or getting traded or anything, but certainly at putting up funny stats. Oddly, the team with the worst winning % when outshot is the Red Wings. Here's the top 5 winning %'s when teams are outshot, just for interest:
1. New York Rangers
2. Minnesota
3. Toronto
4. Calgary
5. Washington
Exactly, three bad teams and two good teams. We've learned nothing. The mean winning % for teams outshot is .469 with a standard deviation of .107, once again showing that most teams are pretty close to average. Further proof then that any given game in the NHL, the team with more shots did themselves no favors. Overall though, the average NHL point percentage* is .555, standard deviation of .060, which means the average team wins only slightly more than they lose, and is pretty close to a coin flip too. Nothing was learned today except that outshooting your opponent means very little over the course of the season, unless you're Atlanta. However, the team that scores more goals than their opponent in a game wins about 65% of the time.
Having said all this, I still want my team to outshoot their opponent every single night because I am a stupid irrational fan.
*The percentage of points derived from the number of games they've played.
No comments:
Post a Comment