the problem is that you don't know if it will last. In fact, you are almost guaranteed to be deadpooled. Whether by getting crushed by a bigger opponent(the company offering a buy out has money to burn, if they don't buy you, they'll buy your competitor) or by a savvy startup which gets lucky.

Where is Lycos? Where is Altavista? Where is Friendster? When you can't take a step without stumbling over a corpse of a rotting startup, gambling it all away in hopes that you are an exception just seems silly.

It's better to sell out early, secure your future, then once you can risk it all without ending up in a poor house, you can go all the way with your second startup.