The driving dynamics of VC are a) power laws and b) saying no >95% of the time
Exit of Twitch, at the time, was greater exit value than that of all other accerator exits put together; Facebook roughly had higher exit value than all other start-ups founded since as of 2014
VCs say no so often because of 1. risk management (they write 7-8 figure cheques) 2. portfolio dynamics (can only make 1 big bet per category per fund; can't bet on 2 car marketplaces) 3. partnership time
Y-Combinator doesn't face any of those constraints. They sit on no boards. Only putting $20k, $150k into those startups (it's now bigger). Being wrong is trivial. What really matters is saying no to something that becomes huge.
Hence Y-Combinator can dominate accelerators like no other - winner takes all market. Not the same for VC, Sequoia has to say no, therefore wins are more distributed
In the long run, hard to return money - best companies go to YC. Not to the median, unlike VC.