My new voting policy is to downvote every single comment that makes baseless inferences from raw karma, starting with this one. It’s much less informative than you seem to think it is. At minimum, in general, most comments don’t even receive enough votes for the karma balance to be statistically significant. (Otherwise, the variance in karma would be significantly higher, assuming reddit-like distributions.)
If by “lack of local sanity” you are referring solely to people who have voted on your previous comment, then do notice that you have very little information from which to inform a prior on how many people have read your comment, how many people have voted upon it, and the distribution of votes thereof. Whatever alien calculus you have that converts raw net karma to a measure of sanity seems horrifically flawed and horrendously underinformed.
At minimum, in general, most comments don’t even receive a statistically significant number of votes. (Otherwise, the variance in karma would be significantly higher, assuming reddit-like distributions.)
This seems to be an incorrect application of the concept of statistical significance.
My new voting policy is to downvote every single comment that makes baseless inferences from raw karma, starting with this one. It’s much less informative than you seem to think it is. At minimum, in general, most comments don’t even receive enough votes for the karma balance to be statistically significant. (Otherwise, the variance in karma would be significantly higher, assuming reddit-like distributions.)
If by “lack of local sanity” you are referring solely to people who have voted on your previous comment, then do notice that you have very little information from which to inform a prior on how many people have read your comment, how many people have voted upon it, and the distribution of votes thereof. Whatever alien calculus you have that converts raw net karma to a measure of sanity seems horrifically flawed and horrendously underinformed.
This seems to be an incorrect application of the concept of statistical significance.
You should thank the people who explained your comment for you, where you should have yourself. Fixed… perhaps?
paper-machine is currently winning by a large margin, but the competition will continue for another day or so!