I was curious what this could mean in reality so I did some back of the envelope math. (As a disclaimer - I have no inside knowledge of the performance of YC's portfolio, so all this math could be wrong.) YC Summer '17 had 294 founders at 124 startups. Let's say there is an Airbnb ($30 billion to common at liquidity) and three smaller but still substantial exits ($1 billion, $500 million and $250 million respectively, same terms). The rest of the class is a wash after expenses and fees which leaves $31,750,000,000 to split up. On a side note, every time I do this math I'm reminded how the ten figure exits really carry the rest of the valley along for the ride; whether there is one in your class or not is a roll of the dice.
YC keeps 4% for themselves (out of which they generally fund expenses and management fees) and puts 3% (with no carry) into the common class pool, divided equally by startup (not founder, to avoid perverse incentives around cramming). In other words, the YC Summer '17 class collectively owns 3% of every member company. After ten years the 3% has been diluted down by follow on rounds to 1.5%. At liquidity, the fund returns 1.5% of $31.175 billion: $467,625,000. Divided among 294 founders equally (which it wouldn't be, as mentioned above - but for easy math) that's about $1.5 million per founder.
If you miss the class with an Airbnb, Uber, or Snapchat and end up with (merely) a few traditional unicorns, the returns decline 90% to a couple hundred thousand dollars per founder, a.k.a. not that exciting financially. So it seems like a gamble.
Still, I like the spirit of the whole idea. As a lifelong entrepreneur who hasn't quite pulled the trigger on applying to YC this might push me over the edge. It would definitely feel like being part of a grander experiment of some kind.