[Itse]

Originally Posted by Cleveland Steam Whistle

There is no arguing, that there is 0 correlation between fighting and scoring a goal (obviously) because you can't score while fighting, but the belief or the creation of another emotional state for the team is all that matters when a fight occurs.

The above makes no sense.

I think the problem with the analysis of trying to correlate the effectiveness or lack of in fighting on a game, is there is no base line for what "good" is.

Actually sports is a rare place where we have completely universally accepted measurements for "good". Such as goals and wins. This is why sports analytics is so popular, it's so black and white.

The anti fighting crowd will look at some stats and say, less than half the time (for example), when a team that's down fights, it doesn't spark a comeback. To them that proves fighting and results are not correlated.

I would suggest however, regardless of where you sit on the fence of this debate, that, that is false logic. Fighting is a tool in a team or a players tool kit of things they can use to influence the game (supposedly). Let's say for a second that a correlation could be made that said, 20% of the time when a fight occurs, the team mounts a comeback (fake stat, don't know the actuals). Anti fighting crowd will go, 80% of the time it doesn't work, fighting doesn't impact. Like I said, in my opinion that's flawed logic.

What you are saying here is the rough equivalent of claiming that calculators are not good for discovering the sum of two numbers.

Studying these kinds of phenomenons is essentially what statistical analysis was developed for. If fights worked in the way you described, statistics would probably show it. Statistics should not only tell us that the correlation exists, but it also put some kind of a number on how common the connection is.


Also, again, I'm not arguing that fights never have an effect on games. Arguing against that is arguing against a straw man. I'm just saying that
1) they very probably don't ALWAYS rally the troops, or even close to that
2) it seems IMO likely that fights very, very rarely actually have an effect on games.


It's also worth noting that nobody is asking for anything even resembling definite proof. What people are asking is some evidence beyond the anecdotal that there's a possibility this phenomenon exists. This is an extremely low bar to set. It's like asking "could you show me the car before I buy it" and getting "no that's impossible, but trust me it exists" for an answer. Should be understandable that such an answer creates some scepticism.

(What I mean by "low bar" is that a statistical correlation doesn't actually prove cause and effect, it just implies that a connection might exist. Even if fights and comeback wins DID correlate, this might just be because both comeback wins and fights are often caused by teams being pissed off about being down in a game.)