Something I’ve always wondered about is what I’ll call sub-threshold successes. Some examples:
A stand up comedian is performing. Their jokes are funny enough to make you smile, but not funny enough to pass the threshold of getting you to laugh. The result is that the comedian bombs.
Posts or comments on an internet forum are appreciated but not appreciated enough to get people to upvote.
A restaurant or product is good, but not good enough to motivate people to leave ratings or write reviews.
It feels to me like there is an inefficiency occurring in these sorts of situations. To get an accurate view of how successful something is you’d want to incorporate all of the data, not just data that passes whatever (positive or negative) threshold is in play. But I think the inefficiencies are usually not easy to improve on.
Many of your examples are maybe fixed by having a large audience and some randomness as described by Robo.
But some things are more binary. For example when considering job applicants an applicant who won some prestigious award is much higher value that one who didnt. But, their is a person who was the counterfactual ‘second place’ for that award, they are basically as high value as the winner, and no one knows who they are.
My assumption is that many of these successes would tend to be widely distributed around some mean, rather than being narrowly concentrated at one point.
So if a joke needs to be 7⁄10 funny to get a laugh, but a comedian delivers what is actually a 6.5/10 joke, you’ll still get some subset of people who find it funnier than it is, such that it gets an appropriate amount of laughs.
Probably there’s some inefficiency, but because of this effect, the number of laughs/number of upvotes I think gives quite good information about the perceived quality of the joke/post.
Something I’ve always wondered about is what I’ll call sub-threshold successes. Some examples:
A stand up comedian is performing. Their jokes are funny enough to make you smile, but not funny enough to pass the threshold of getting you to laugh. The result is that the comedian bombs.
Posts or comments on an internet forum are appreciated but not appreciated enough to get people to upvote.
A restaurant or product is good, but not good enough to motivate people to leave ratings or write reviews.
It feels to me like there is an inefficiency occurring in these sorts of situations. To get an accurate view of how successful something is you’d want to incorporate all of the data, not just data that passes whatever (positive or negative) threshold is in play. But I think the inefficiencies are usually not easy to improve on.
A standard trick is to add noise to the signal to (stochastically) let parts get over the hump.
I agree this is an inefficiency.
Many of your examples are maybe fixed by having a large audience and some randomness as described by Robo.
But some things are more binary. For example when considering job applicants an applicant who won some prestigious award is much higher value that one who didnt. But, their is a person who was the counterfactual ‘second place’ for that award, they are basically as high value as the winner, and no one knows who they are.
My assumption is that many of these successes would tend to be widely distributed around some mean, rather than being narrowly concentrated at one point.
So if a joke needs to be 7⁄10 funny to get a laugh, but a comedian delivers what is actually a 6.5/10 joke, you’ll still get some subset of people who find it funnier than it is, such that it gets an appropriate amount of laughs.
Probably there’s some inefficiency, but because of this effect, the number of laughs/number of upvotes I think gives quite good information about the perceived quality of the joke/post.