Applied Physics

[Triple Zero]

For those who didn't read the article -

If performance at higher up on the hierarchy is correlated to performance at lower levels, you should promote your best people.
If performance is uncorrelated, then you should promote your worst people.  (If Alice is a great burger flipper and Bob is a shitty burger flipper, you should promote Bob over Alice because either is just as likely as the other to be a good manager, but if you promote Alice you're out one good burger flipper.)

Random advancement is only the least bad strategy if you need one that works in both universes.

They haven't proven anything about actual promotion strategies.

thanks for summing it up, I hadn't read the article yet.

so in short, this wouldn't work for electing admins, cause admins are still active posters, so promoting one to admin doesn't give us one less active poster.