I think that's a good point, but it's just unlikely to actually happen with startup options because of the typical 90-day exercise clause and limited information.
The 90-day exercise clause and options instead of shares effectively forces you to buy the coconut, meaning that instead of a 50/50 chance of 0/200, if you plan on leaving before the event that resolves 0/200, then if your strike price on a coconut is $50, it's more like 50/50 chance of -50/250 because you expect to have to exercise when you leave. You can play with the exercise cost, but it does change your math quite a bit regardless. Now if startups gave actual stock instead of options or if they don't place a 90-day time limit on exercising vested options, this wouldn't be an issue. But almost all startups do it this way.
The second point is that the company often doesn't look dead until much later. You're just not going to be able to tell one year in, or two years in, or three years in. So you won't be in a good position to capitalize on volatility like you would on the public markets.
Then you have things like liquidation preferences that will skew your EV calculation, except they might happen after you join, but you may or may not ever find out that those gotchas are there.
As it stands, all of this just creates an incredibly inefficient market that requires employees to take badly informed, high risk bets, where they often don't have deep enough pockets to absorb the risk.