Clayton Blog: Competition Committee recommends OT changes

[Concord]

I like the way College does it.

It's even and fair.

 

[AbeBeta]

There is a potentially infinite data set. Just like the coin flip situation. There is no difference. You can flip the coin 350 times , and if the results are 52% heads, then that abnormal result may or may not be statistically significant, meaning it may or may not be sufficient to tell us there is some flaw in the coin causing it to land on heads.

Yes, that is exactly what we're asking. Whether this is a large enough sample of a potentially infinite data set to tell us whether the results here are statistically significant.

Maybe. Maybe not. Obviously there should be some influence. The influence may actually be greater than the stats shown here. I don't think the sample is large enough to really make any meaningful determination one way or the other.

We are just going to have to agree to disagree here. I believe you are completely missing the boat on what a population is and isn't and what a sample is and isn't and when you would use an inferential statistic and when you wouldn't. You are completely right IF you define populations and samples as you are doing. But completely wrong if you define populations and samples as I've (correctly) done.

 

[Doomsday]

Im glad they are looking at more subtle changes to the overtime format, instead of looking at the college format

 

[kmd24]

Again, this isn't a sample. It is data from all the games that were played. Inferential statistics focus on drawing inferences about populations based on samples. This is not a sample. It is all of the games that went to OT. That's the population. We know exactly how correct the 52% is because it is based on the entire population.

You have a sample, and you are using inferential statistics (i.e., trying to reach conclusions that extend beyond the immediate data). You have a sample because the population is the infinite set of games that could be played under this rule. You are inferring because you are using the results from past games to infer the fairness of the coinflip rule in all games, including future games - games that haven't been played and can't possibly be in your data set.

Think about it this way. If you limit your population to include only games that have already been played, then you can make statements with certainty about only those games. If you want to extend your population to all games that could ever be played (the question of fairness), then you need to consider computing the likelihood that a fair process could generate the results we have to date.

I think you are letting the fact that you have a reasonable number of data points cloud your judgement on this matter. Would you still argue this strongly if there were only ten results and the coinflip winner had only won four of the games?

I don't understand why you think this problem is any different than the problem of determining whether a coin is a fair coin from the results of a given number of flips.

 

[burmafrd]

If you cannot win a game in 60 minutes, why all the heartburn about OT?
As was pointed out, one possession wins happen only 29% of the time. Out of how many games a year? When it affects SO little, why change something that has been in the NFL for 50+ years?