Comment by garyrob
If you try 2N times to succeed in ventures which each have a 1/N chance of success, as N increases the probability of such as success quickly converges on about 86.5%.
(The limit is 1-e^(-2).)
So, if you have a LOT of chances to try things that are highly improbable but high upside, your odds are quite good.
> So, if you have a LOT of chances to try things that are highly improbable but high upside, your odds are quite good.
But that perfectly highlights why the "startup gamble" is a great bet for VCs but a horrible bet for most employees. Let's say N is, generously, 50 (i.e. 1 in 50 startups are a resounding success, which seems probably a bit over-optimistic but reasonable). VCs can easily spread investment around to 100 startups, but employees get a few swings at bat at most when it comes to where they work.
For most things in life, you often just don't have that many chances. E.g. most people don't date a hundred people before finding their spouse.