Curious Effects
Tyler Dellow has a fascinating post in response to this from Eric Duhatschek:
Generally, what sets the NHL apart from the far more predictable (and parity-free) NBA is the curious effect that team play and chemistry, coaching and camaraderie can have on results.
Tyler – using some Tom Tango work – makes a compelling case that the curious effect is due to random variance, not any of the reasons cited by Duhatschek. In other words, luck is a smaller factor in the NBA than it is in the NHL. There is less parity in the standings in the NBA not because there is a larger spread between the best team and the worst, but because bad breaks have a smaller impact and the better team is upset less frequently in basketball.
The legendary baseball writer Bill James sees things the same way and suggests that the information can be used to actually improve leagues. I’m not sure that is true because I’m not sure what constitutes improvement. Basketball is the sport where the best team is the most likely to win. Football is the sport where luck and randomness has the most influence. Baseball and hockey fall between the two extremes. On principle, I’d prefer skill to win out almost universally but as James points out, that would not make for a very interesting league. On the other hand, which of us wants to accept that the breaks can make the difference between a successful season and an unsuccessful one for hockey teams?
The best answer for the fan – and Eric Duhatschek – is to decide that Tyler is full of shit. All we have to do is believe in the curious effect team play and chemistry and coaching and camaraderie can have on results in hockey. (I think leadership is really, really important too.)
I don’t really care if people decide that I’m full of shit on the hockey point. What kills me is when they transfer the kind of thinking to stuff that actually matters, like economic or criminal policy. Not to turn your blog political – and they’re the only party I’ve ever voted for, although their grasp on my vote is EXCEEDINGLY tenuous – but when you look at the Tories’ crime policy, they basically acknowledge that they have no sound basis what they’re doing with things like mandatory minimums, such as research into whether they work or even whether they are stiffer than the time that the crime attracts in the absence mandatory minimums. When challenged they say things like:
“That was a good thing for us politically, in that sociologists, criminologists and defence lawyers were and are all held in lower repute than Conservative politicians by the voting public. Politically it helped us tremendously to be attacked by this coalition of university types.”
I mean, that’s insane. It’s probably right – but it’s not a good thing. Politics is a dirty business but these people prey on the general ignorance of the public. I hammer away at the thinking about hockey that exists but, god help us, the thinking elsewhere is no better and (oddly) the non-MSM political sites don’t really seem to be the least bit helpful; it’s all just partisan screaming.
On the other hand, which of us wants to accept that the breaks can make the difference between a successful season and an unsuccessful one for hockey teams?
I think that there’s an excellent case to be made that this happens more than we think. Columbus, to pick an example, looks to me like that’s absolutely what’s happening there. I don’t know why that makes it a bad thing though; I guess it depends on your outlook on things. I tend to think that we’re all subject to the vagaries of fate, regardless of our skills and aptitudes, ultimately it’s all in the hands of the gods. One day you’re touted as a future premier of Ontario, the next day you’re charged with the death of a bike courier. Why shouldn’t the fates have a place in hockey? Why is that bad?
I don’t really care if people decide that I’m full of shit on the hockey point. What kills me is when they transfer the kind of thinking to stuff that actually matters, like economic or criminal policy.
If I was not absolutely clear, I don’t think you are full of shit. I think you are absolutely right. I don’t really follow the math that was used to quantify it, but the logic of the position is unassailable in my view. I remember when I came across an article on the new (at the time) work being done by Kahneman and Tversky. It was like an epiphany. I thought their work would clearly change public policy since at least half of it was clearly wrongheaded. Didn’t happen. I don’t think it ever will.
Life in the short run is luck, but we have this illusion of control so powerful we blow on the dice and talk to them even though we know it can’t possibly affect the outcome. We don’t reason well at all. Experience frequently misleads us. We invent rational explanations for chance occurences.
It is everywhere.
I think that there’s an excellent case to be made that this happens more than we think. Columbus, to pick an example, looks to me like that’s absolutely what’s happening there. I don’t know why that makes it a bad thing though; I guess it depends on your outlook on things. I tend to think that we’re all subject to the vagaries of fate, regardless of our skills and aptitudes, ultimately it’s all in the hands of the gods. One day you’re touted as a future premier of Ontario, the next day you’re charged with the death of a bike courier. Why shouldn’t the fates have a place in hockey? Why is that bad?
I think it happens all the time, too. Every year probably. The difference between the last four or five teams in the playoffs and the four or five teams that barely miss is usually nothing. There is no good reason to explain why one team missed and the next did not.
Fate does have a place in hockey. Everybody will acknowledge that. What they will not acknowledge is that fate makes a big difference. Fans do not want to believe that so they do not believe it. Hockey writers can’t be explaining things with “shit happens” all the time even though that is usually the story. So they say that the breaks even out or they claim that you make your own breaks by working hard. They dismiss injuries as excuses. They tell the narrative about the plucky band of lads who rise to the occasion and overcome great odds to win. And fans lap it up. The narrative is part of the package that sells the sport. It is part of the appeal. It doesn’t have to have either truth or substance.
In other words, luck is a smaller factor in the NBA than it is in the NHL. There is less parity in the standings in the NBA not because there is a larger spread between the best team and the worst, but because bad breaks have a smaller impact and the better team is upset less frequently in basketball.
I agree that luck plays a far larger role in hockey, but that doesn’t mean that it increases pairty. While pairity would definitely increase if only the bad teams caught lucky breaks, there’s essentially an equal chance that luck will cause good teams to perform better than they otherwise would, decreasing pairity. Overall, the two effects should essentially cancel each other out, with no effect on pairity.
I agree that luck plays a far larger role in hockey, but that doesn’t mean that it increases pairty. While pairity would definitely increase if only the bad teams caught lucky breaks, there’s essentially an equal chance that luck will cause good teams to perform better than they otherwise would, decreasing pairity. Overall, the two effects should essentially cancel each other out, with no effect on pairity.
This cannot be true. Skill influences the result and luck influences the result. You are right that luck will tend to even out over the long run (keeping in mind that even a season is not the long run) so the luck factor is a 50/50 proposition. The point is that the skill is not a 50/50 proposition. If there is a change made that increases the luck factor – shootouts, say – then skill becomes less important. Which team benefits more when skill is less important? The lesser skilled team. That team gets better results and the better team gets worse results. Result? More parity.
The better team wants luck to have zero influence. If both teams are healthy and play their best and the lucky breaks even out, the better team will always win. The worst team would prefer to see the game decided on a lucky break. When that happens they have a 50/50 chance to win.