The brown M&M filter is good because you want people to read every single instruction in regards to something critical to crew and audience safety.

On the other hand, reading every single Peter Thiel has ever written doesn't seem like a good use of an entrepreneur's time.

On the other other hand, if you are Peter Thiel you can make people ask "how high?" when you say "jump." I can see the heady appeal of that.

An interesting point. Then there's the question of whether people who research their potential funding sources to such a level are more or less likely to be good investments. That's a lot less clear than in the Van Halen case - here, I see plenty of arguments in both directions.

What would be an argument for dis-correlation? I have a hard time seeing how someone better at startups would be less likely to do their homework on a potential investor, given that they'd think ahead in numerous other ways. It would have to be a situation in which think ahead for everything but this, or in which their happy-go-lucky style is miraculously good at making the right choices everywhere but here.

It's not that they would be less likely to be good relative to themselves-but-did-research. It's that someone focused on the sale is what Thiel is purportedly trying to avoid in dismissing those in suits. Someone focused on the sale of the company to investors more than on the company might be even more likely to do that research than the best bets, and that might cause dis-correlation.

Who is most likely to be doing through homework: somebody who is having great success and can expect multiple competing offers or somebody who isn't having much success and has a really hard time raising money?

Now by no means is it a given that amount of success now tells anything about amount of success tomorrow, but if they aren't at least positively correlated then VC have no business being in business.