Because just like with lotteries, there are a lot of blanks, few small wins and a tiny amount of never-work-again wins: Many startups don't net the investor much, some make quite a bit after being acquired, and very few - such as facebook - turn out to be really successfull.

Your phrasing suggests that you compare Facebook (now) with lottery as a system. You should rather see Facebook as a lottery ticket that turned out to be the jackpot.

Every lottery ticket has the same odds of winning, no matter the "strategy" employed in picking numbers, location of buying the ticket, etc. Facebook is not in any way a analogue, unless you are arguing that its success was entirely random, and would have stood an equal chance of happening if I had started it at a community college, then put the headquarters in Alaska, which is false.