[Roberton3]

Or that's just random noise.  I didn't calculate any test statistics, but given how small the sample size is, I doubt that the difference in average finishes between the preseason #1 and #2 is statistically significant.

I am no stats guru. But the pre-season #1 was the "champ" in 2 seasons while the #2 was the "champ" in 5. Seems like a pretty big difference.

I just ran a quick statistical test to compare the finishes for the preseason #1 and #2 teams, and the difference isn't even close to being statistically significant.  The test looks at a hypothetical: if the preseason #1 and #2 have exactly the same odds of finishing in any given position, what's the probability that the difference in actual results (in a sample this size) will be at least as big as we see in the data.  Or in other words, if it doesn't matter whether you're the #1 or #2, what are the odds that we would get a result like this just by chance.  The answer was roughly 40%.  The standard threshold for calling a result statistically significant is that it would occur by chance less than 5% of the time.  So this wasn't close.

Of course, this doesn't prove that there's no difference between being ranked #1 or #2.  It just says that we don't have enough data to tell whether there's a difference or not (or which way such a difference would go).


[For the stats geeks out there: I just ran a simple t-test.  There are a lot of reasons why this isn't exactly right for this case, but it should be fairly close, and I didn't feel like taking the time to find a test that would be more strictly correct (if one exists).]

[Roberton3]

Or that's just random noise.  I didn't calculate any test statistics, but given how small the sample size is, I doubt that the difference in average finishes between the preseason #1 and #2 is statistically significant.

I am no stats guru. But the pre-season #1 was the "champ" in 2 seasons while the #2 was the "champ" in 5. Seems like a pretty big difference.

I just ran a quick statistical test to compare the finishes for the preseason #1 and #2 teams, and the difference isn't even close to being statistically significant.  The test looks at a hypothetical: if the preseason #1 and #2 have exactly the same odds of finishing in any given position, what's the probability that the difference in actual results (in a sample this size) will be at least as big as we see in the data.  Or in other words, if it doesn't matter whether you're the #1 or #2, what are the odds that we would get a result like this just by chance.  The answer was roughly 40%.  The standard threshold for calling a result statistically significant is that it would occur by chance less than 5% of the time.  So this wasn't close.

Of course, this doesn't prove that there's no difference between being ranked #1 or #2.  It just says that we don't have enough data to tell whether there's a difference or not (or which way such a difference would go).


[For the stats geeks out there: I just ran a simple t-test.  There are a lot of reasons why this isn't exactly right for this case, but it should be fairly close, and I didn't feel like taking the time to find a test that would be more strictly correct (if one exists).]