info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Hockey Cliches, Conventional Wisdom, and Statistical Abuses That You're Tired Of
17 views
Skip to first unread message
Jim Bauch
Jun 14, 2017, 6:52:42 PM6/14/17
to
Hey, it's the offseason, and I can only talk so much about the expansion draft.
I've got three that are bugging me:
1. "A two-goal lead is the most dangerous lead in hockey." I literally have no idea what this is supposed to mean. It can't possibly be intended at face value. Nobody actually thinks that a one-goal lead is preferable to a two-goal lead. Players don't intentionally avoid shooting with a one-goal lead lest they inadvertently make it a "dangerous" two-goal lead. Is it supposed to mean that "teams that are trailing by two goals are more likely to score the next goal than in other score situations"? Because that seems unlikely, too, plus I've never seen anyone attempt to back this up with actual data. It just seems to be one of those things that people say because somebody once thought it was clever and nobody's really giving it any actual thought any more. Can't we please retire it?
2. "This game may be over, but they're sending a message for Game Two." Used in playoff series to justify a team gooning it up late in a hopelessly lost game. What, pray tell, is the "message" being sent? "We're poor losers and will react with pointless violence if you beat us?" Perhaps that's the actual message, but since commentators generally seem to approve of this message-sending, that's probably not what they mean. "You may have beaten us tonight, but we've got plenty of fight left in us and will come out playing hard in the next game!"? This is probably it. But why would you want to send that message? Wouldn't it be better to just come out playing hard in the next game, without warning the opponent that you intend to do so? Why not let them get overconfident? The only rational reason I can think of for doing this is to send a message to your TEAMMATES and coach that "hey, I'VE still got some fight in me, what about the rest of you?" Of course, I suspect the real reason isn't rational at all, it's just blowing off some steam since (a) you're pissed at losing; and (b) you hate the other team or at least a particular player, and now's a good time to take your shot when you don't care about giving up a power play. Which is a perfectly good explanation -- why try to pretend it's some strategic thing?
3. "Team A wins blah% of the time when Player X scores a goal." Usually said with the implication that (1) "blah%" is awfully high, (2) that this is meaningful, and (3) that this shows that X is a clutch player who's really important to the team's success (bonus points for saying "he's the straw that stirs the drink." Point (1) is often misleading, because almost every player's "blah%" should be higher than his team's record. In today's NHL, there are a lot of shutouts. The Habs were 47-26-9 last season, for a 62.8 points percentage. Take away the six regulation and one OT/SO loss in which they were shut out, and they were 47-20-8, for 68.0%. That means that, on average, the Habs won 68% of the points when ANYONE scored. That percentage will be higher for some players, lower for others, but on a weighted-by-goal basis, it's 68%. So the percentage for any particular player better be significantly higher than that to even be worth commenting on. Point (2), well, that's our old friend sample size. Even a 20-goal scorer only scores in at most 20 games, which is not a big sample, so the team's record in those games could easily be coincidence. And as to Point (3): couldn't you just as easily say that "Player X gets a lot of easy goals in games his team is dominating"?
Sorry -- that's three long rants for one post, but I couldn't justify starting three separate threads on my pet peeves.
Jim
I've got three that are bugging me:
1. "A two-goal lead is the most dangerous lead in hockey." I literally have no idea what this is supposed to mean. It can't possibly be intended at face value. Nobody actually thinks that a one-goal lead is preferable to a two-goal lead. Players don't intentionally avoid shooting with a one-goal lead lest they inadvertently make it a "dangerous" two-goal lead. Is it supposed to mean that "teams that are trailing by two goals are more likely to score the next goal than in other score situations"? Because that seems unlikely, too, plus I've never seen anyone attempt to back this up with actual data. It just seems to be one of those things that people say because somebody once thought it was clever and nobody's really giving it any actual thought any more. Can't we please retire it?
2. "This game may be over, but they're sending a message for Game Two." Used in playoff series to justify a team gooning it up late in a hopelessly lost game. What, pray tell, is the "message" being sent? "We're poor losers and will react with pointless violence if you beat us?" Perhaps that's the actual message, but since commentators generally seem to approve of this message-sending, that's probably not what they mean. "You may have beaten us tonight, but we've got plenty of fight left in us and will come out playing hard in the next game!"? This is probably it. But why would you want to send that message? Wouldn't it be better to just come out playing hard in the next game, without warning the opponent that you intend to do so? Why not let them get overconfident? The only rational reason I can think of for doing this is to send a message to your TEAMMATES and coach that "hey, I'VE still got some fight in me, what about the rest of you?" Of course, I suspect the real reason isn't rational at all, it's just blowing off some steam since (a) you're pissed at losing; and (b) you hate the other team or at least a particular player, and now's a good time to take your shot when you don't care about giving up a power play. Which is a perfectly good explanation -- why try to pretend it's some strategic thing?
3. "Team A wins blah% of the time when Player X scores a goal." Usually said with the implication that (1) "blah%" is awfully high, (2) that this is meaningful, and (3) that this shows that X is a clutch player who's really important to the team's success (bonus points for saying "he's the straw that stirs the drink." Point (1) is often misleading, because almost every player's "blah%" should be higher than his team's record. In today's NHL, there are a lot of shutouts. The Habs were 47-26-9 last season, for a 62.8 points percentage. Take away the six regulation and one OT/SO loss in which they were shut out, and they were 47-20-8, for 68.0%. That means that, on average, the Habs won 68% of the points when ANYONE scored. That percentage will be higher for some players, lower for others, but on a weighted-by-goal basis, it's 68%. So the percentage for any particular player better be significantly higher than that to even be worth commenting on. Point (2), well, that's our old friend sample size. Even a 20-goal scorer only scores in at most 20 games, which is not a big sample, so the team's record in those games could easily be coincidence. And as to Point (3): couldn't you just as easily say that "Player X gets a lot of easy goals in games his team is dominating"?
Sorry -- that's three long rants for one post, but I couldn't justify starting three separate threads on my pet peeves.
Jim
Chuck
Jun 14, 2017, 7:15:57 PM6/14/17
to
A two goal lead is dangerous when a team stops going for a 3 goal lead! Coasting playing offence free hockey while protecting a two goal lead, if a goal is scored, the momentum created by that goal often leads to a quick tying goal.
Chuck
Jun 14, 2017, 7:20:11 PM6/14/17
to
2. The game is out of reach, the losing team's knuckle staggers get the chance at icetime they'd never would see in a close game. So they use that icetime to showcase the only skill set they have.
Chuck
Jun 14, 2017, 7:24:12 PM6/14/17
to
Hate autocorrect the post should read knuckle draggers
Gerry
Jun 14, 2017, 8:00:21 PM6/14/17
to
Give me one of those nice writeups about historical records vs. teams, or records in Building X, dating back the last waaaay-too-many-years-to-be-remotely-meaningful. Like it matters that some version of the Habs beat some version of the Leafs 10 years ago, when like 3 skaters in the whole game are even on the same team anymore (and for that matter are 10 years older and either shadows of their former selves or have blossomed from when they were baby-faced rookies).
Last season's record? Ok, for a division rival you played many times last season, I'll accept it. But the roster turnover is too significant even over a few years for most of these records to have any significance.
l8r,
Gerry
Last season's record? Ok, for a division rival you played many times last season, I'll accept it. But the roster turnover is too significant even over a few years for most of these records to have any significance.
l8r,
Gerry
0 new messages