Great question. An optimal strategy doesn't exist, because it is so circumstantial. The employee and founder (the seller) need a good idea about the price and p

Great question. An optimal strategy doesn't exist, because it is so circumstantial. The employee and founder (the seller) need a good idea about the price and probability of success. Making a buy/sell decision either way certainly effects outcomes, independent of the ultimate success. It really depends how the seller values the stock, and how that differs from the buyers valuation. Knowing a lot about (startup) stock valuation will certainly help. Probably its more political than anything, because the price of a 100M$ company can fluctuate wildly, and so being in similar situations should help as well. I don't know much about Silicon Valley, but there you have a lot of investor driven companies (people raising money to push the value, to raise more money, etc. without a real business seems much more common). The 10B$ outcome is exceedingly rare, so the employee should think really hard about the value (see Moskovitz talk about it here at 36:00 https://www.youtube.com/watch?v=CBYhVcO4WgI https://www.youtube.com/watch?v=CBYhVcO4WgI) Usually 1M$ will be much more valuable when going from 0 then the next 1M$. In the example it seems irrational not to sell for 1M$, if the employee doesn't have already a lot of money. But usually people are not given the choice and need a liquidity event (IPO or acquisition). If one assumes that shares could be liquid from day 1 the dynamic changes. Most likely options will have vesting. The real issue here however, in most cases, is that there is no liquid market until very later stages. So the company might be worth 100M$ in its entirety to sell, but employees can't sell options. Also simple options might actually not be the best instrument, but it is too costly to customly define an instrument for every employee. Say an employee could be an option which maximizes wealth for 0-1M$ (1%). If the total value hits 100M$ the other owners could buy out those options cheaply, and both sides win. Having only one strike price makes the potential value curve extremely steep, making startup shares more like lottery tickets. To consider an extreme case - Google. Page almost sold Google for 1.6M$ in 1997 Luckily he didn't. In many cases it really depends on the belief of the founders/employees/investors. For employees it seems that their problem is more that they can't sell earlier than the liquidity event, which is due to the fact that IPO's are so expensive. With Crypto-finance there will be the potential to create a market for stock at almost no cost, which will change the game, because it allows to float stock from the earliest stages when the most wealth is created. It will be much more common for employees to exit early and for the wealth to be more widely distributed. Getting to 10-100M$ is magnitudes of order more common than the 1B$+ exits.