Hacker News Reader

Unless your definition of sucess is a bit more flexible than the one of the article... If that happens there are small improvements to your chances.

But then, it's a dificult calculation. The more capital you apply at the company, the more money you must take out of it.

My probability theory is a bit weak so correct me if I'm wrong. Since each company you start is an "independent event" doesn't your probability still stay at 6e-5?

No, the probability I calculated is the probability that at least one of the companies will succeed if you start N of them, or 1-(1-0.00006)^N

only if success (here defined as creating what becomes a billion dollar company) is iid. its probably not. if you've failed at starting 5 companies, the probability of succeeding at the next one is probably less than 0.00006%

Your conclusion does not follow from the premise. Couldn't the experience gained from starting the first 5 result in a greater chance of "success"? If failure is defined as anything less than 1 billion, it is particularly meaningless as an indicator of future success.