[jonkelly]

PG offered a clear underlying model in this essay. His model is that a) founders need to sell well to succeed and that b) it's hard to sell well if you can't be easily understood. You have a "guess" that mundane traits are more important for success. He has a theory and data that support his theory.

[vasilipupkin]

That's fine just as long as what he did wasn't the following:

see some small number of founders with a fairly strong accent, say < 30 or < 50 or < 60, then observe that those startups happened to do poorly and then extrapolate a theory from those observations to explain what is happening. I am not necessarily disagreeing with his thesis, but in statistical data analysis you have to be very careful about drawing conclusions from small samples when the underlying distribution is non-normal, with fat tails

[tlb]

This is an epidemiological observation, and those rarely indicate a particular mechanism. Even with giant populations and teams of scientists we have only tentative and shifting theories for why, for example, Americans are fatter than Europeans.

Controlled studies of startup founders are not practical.