Diversification is also important, though, and is really the biggest problem with startup employee equity: your job is also your lottery ticket, and you lose both at once if/when the company fails. Dilution and preferred stock deals just make it worse.
There are, of course, other reasons to do work than money. But what charity would you give $50k/year to?
Yes, some blind luck is needed to win at this game. However, the job market has been such that a solid programmer can get something lined up pretty quickly if their current job fails. I don't think the risk is as bad as people portray. Again, picture buying long term call options for a single stock with a proportion of income each year. Somewhat risky, but risk is capped and the payoff could be quite big. More diversification is almost definitely more efficient but carries less tail upside.
By "unfairness" I don't mean risk, I just mean unfairness.
I agree that job loss probably isn't all that bad... (Unless it lines up with a wave of companies closing, which is known to happen from time to time.)
I can also just save + diversify /more/ with faang salary and equity than I could at a startup. The $50k/year is a lower bound on what I expect the pay cut to be in moving to a startup. So from where I sit, there's really no financial argument for moving to a startup. The startup has rather lower pay, some probability of payout multiplied by a value that I just kinda expect will be chipped away by the suits, and some probability of a bad outcome (with some bounded range of potential badness). I look at it and just think it's not interesting unless I get really, really bored.
Joining a company that's relatively close to IPO is an interesting middle area.