> It takes what is easy to quantify without any consideration of whether it's at all a useful proxy for the thing it is supposed to measure. The measures it us

> It takes what is easy to quantify without any consideration of whether it's at all a useful proxy for the thing it is supposed to measure. The measures it uses all seem reasonable proxies for "popularity", which doesn't have a rigorous definition. Unlike some other attempts (eg. Dice), they at least combine multiple proxies. > So, naturally, managers love it. Well, naturally. By definition, most people are mediocre at their jobs, and managers are no exception, so naturally the main de-risking strategy is "what is being used most". Why would you find the popularity of popularity contests surprising?