> Yet, in investor/VC/pitch/roadshow decks, I usually see "this is how many MIT and how many Stanford grads we have"... Trust me, it's not nearly as relevant as people think, but it must work.

VCs are risk-averse pattern matchers. Their training patterns include knowing that startups with a high percentage of MIT and Stanford grads do well. Despite the fact that correlation isn't causation, they live off of that fallacy, so they would rather invest in a bunch of MIT and Stanford grads than not.

It actually makes a lot of sense for a VC to do that. They have a huge incentive to have a lot of false negatives and avoid false positives, so it's much better for them to use criteria that will eliminate good companies than allow bad ones.