overtime rules

[Texas Aggie]

It's not correlated with anything it doesn't cause if it's a random event.

This isn't true. Lots of random events have both correlations and causations with other events. Car accidents, in general, are a random event. Yet, there's a strong correlation between accidents and alcohol consumption by one or both drivers beforehand. A drunk driver in a huge parking lot with no cars or obstacles isn't going to wreck (assuming he stays on the lot) but on the road with barriers, lights, and other drivers, well, you know the result. A car accident, obviously, doesn't cause drunk driving and DD, while strongly correlated with accidents (and vice versa), doesn't cause them. (This is an academic discussion; don't drink and drive to prove me wrong; just debate the point. PLEASE.)

[youngump]

Originally Posted by Texas Aggie

This isn't true. Lots of random events have both correlations and causations with other events. Car accidents, in general, are a random event. Yet, there's a strong correlation between accidents and alcohol consumption by one or both drivers beforehand. A drunk driver in a huge parking lot with no cars or obstacles isn't going to wreck (assuming he stays on the lot) but on the road with barriers, lights, and other drivers, well, you know the result. A car accident, obviously, doesn't cause drunk driving and DD, while strongly correlated with accidents (and vice versa), doesn't cause them. (This is an academic discussion; don't drink and drive to prove me wrong; just debate the point. PLEASE.)

Car accidents are not a random event. They are caused by the things that lead up to them. Those things are causative. There's a strong correlation between drinking and accidents because drinking is a causal factor in accidents. Things will be correlated only if one causes the other or both share a common cause. Causes don't necessarily have to be exclusive.
________
Hyde Park Residence

[sseltser]

Originally Posted by youngump

Car accidents are not a random event. They are caused by the things that lead up to them. Those things are causative. There's a strong correlation between drinking and accidents because drinking is a causal factor in accidents. Things will be correlated only if one causes the other or both share a common cause. Causes don't necessarily have to be exclusive.

The coin toss is a random event. It is still correlated to the outcome of overtime. "Random" is not the opposite of "correlated".

This also doesn't imply that the coin toss is rigged. It only says that the winner of the coin toss (which is a fair game) is more likely to win the game.

[mbyron]

Originally Posted by sseltser

The coin toss is a random event. It is still correlated to the outcome of overtime. "Random" is not the opposite of "correlated".

This also doesn't imply that the coin toss is rigged. It only says that the winner of the coin toss (which is a fair game) is more likely to win the game.

That's right. And even though the coin toss is fair, it is a fair game of chance. Ordinarily we expect football to be a game of skill and the outcome to depend as much or more on skill as chance. Otherwise, we could just end the game with the toss and not play at all.

[bigjohn]

I would much rather see them play a 7.5 minute (half a regulation quarter) overtime period or 5 minutes, whatever. That way the game is not sudden death. If the game is tied at the end of overtime, declare a tie and go on. I don't understand why a draw is so bad!

[Robert Goodman]

Originally Posted by bigjohn

I would much rather see them play a 7.5 minute (half a regulation quarter) overtime period

Which is what the World Football League adopted for their inaugural regular season, 1974. That prompted the NFL to extend their existing OT rules to regular season games.

or 5 minutes, whatever. That way the game is not sudden death. If the game is tied at the end of overtime, declare a tie and go on. I don't understand why a draw is so bad!

If the NFL were really concerned about adding too much playing time, they could simply go back to before 1974 and not have any tiebreaker in any games except their playoffs and finals. OT added a minuscule amount of playing time per team per year before then; most teams had never played any OT games.

Of course that would still leave the problem for post-season play, where one could say it was most critical. I don't know why the pros adopted sudden death as their tiebreaker to begin with; it was decades that the NFL had it in the rule book before it ever needed it used. I don't know if they or any of the other pro leagues in the USA that had knockout games had contemplated such, although Canadian football had a tiebreaking provision (similar to the WFL's) that went way, way back. Even leagues that had a championship game, when they weren't big enough to require playoffs to get there, could have left a championship tied because there'd be no game to advance to. I don't know what provision the AAFC had, for instance.

One thing regular season tie games did for the NFL was reduce the need for standings tiebreakers. Since ties have become such a rarity, there's been greatly increased resort to various statistics to determine playoff standings. The "T" in W-L-T often decided things before. But there was still the question of whether to treat ties as half win + half loss, as the AFL did, or ignore them in computing W-L ratio as the NFL did, which got the Rams in and kept the Colts out one year.

Another factor in producing OT games has been the 2 pt. try. In college football it was used mostly to break ties, while in the AFL it was used mostly to produce them. Since the NFL adopted the 2 pt. conversion, I'm sure there've been considerably more games gone to OT because of a similar philosophy. Going for 2 is considered less than a 50-50 proposition, so they'd rather play for OT if they have a chance to play for the win in regulation, but meanwhile it's given teams a lot of chances to tie scores instead of being 1 pt. behind. In college football it was considered wimpy to play for a tie almost no matter how bad the percentages were; but now that they have tiebreaking frames I suspect they play to tie it in regulation much more often.

You know, I just realized the term "regulation time" or "regulation play" is archaic in NCAA & NFL games. When there are no more games that end in a tie without extra play, a regulation game is a game that goes to tiebreaker if needed. No more distinction in that regard between regulation and championship games, except that in the NFL they can play more than one OT period in the latter. So in the NFL, until they go to the 6th period, they're still in "regulation"!

Robert

[youngump]

Originally Posted by sseltser

The coin toss is a random event. It is still correlated to the outcome of overtime. "Random" is not the opposite of "correlated".

This also doesn't imply that the coin toss is rigged. It only says that the winner of the coin toss (which is a fair game) is more likely to win the game.

Agreed, my point was to respond to an earlier poster who claimed that showing a significantly increased win percentage didn't show whether there was causation or correlation. If the coin toss is random and correlated, then it must be that the result of the coin toss actually causes the correlation.
________
List of best vaporizers