Nice article, but drastically oversimplified. Paul ignores two critical issues: Risk, and non-linear utility-of-money functions. These two factors become critical when there is a tradeoff between probability of success and the payoff of success. Suppose, as a simple example, that I have a startup which I think has a 50% chance of succeeding and being sold for $1M, and a 50% chance of failing and being worthless. Now…
How does that mesh with the fact that a failed startup is probably worthless (in the literal sense that you can't make money from it), and most startups probably have >90% failure rate?
I know there is a learning experience in startups and that working hard on something fun is valuable, so worthless is really just talking about immediate money here.