Know of any good web-apps or other easier to use programs for the layman to calculate these things? Thank you for mentioning these formulas too. These give a person something to argue with.

There's probably an even simpler model than that. Consider the pay cut you'd take to work at the startup... how much equity did the seed-round investors get for that much money? Multiply that by 1.5-2x (just a guess, perhaps someone has a better idea) to account for their liquidation preferences.

If that's less than your equity stake, then take the corporate job and use the extra cash to invest in startups.

Good idea, but then you are relying on what they told you the investors put in for that equity. If you can find another method to verify that info, all the better. Also, you are correct in saying that you should get at least that valuation matching, if not more due to the seat you put in.

How would they, as likely a non-accredited investor, "use extra cash to invest in startups?"

Your potential payout is your number of shares x the share price.

So let's say that comes out to $100K.

You have to discount that to present value. Money is worth more now than it is in the future.

Assuming an interest rate of 5% and there being a liquidity event in 5 years that is (1.05 ^ 5).

So $100K in 5 years at 5% is worth $78K now.

You also have to factor in risk. There are various models but I like to simply multiply by the probability of exit.

So if there is a 10% chance you will see something then I take the $78K and multiply by 0.10 which comes out to $7.8K.

Also, if you exercise early there is even more risk. Let's say you paid $10K for them. That $10K invested at 5% would have netted $10K * 1.05 ^ 5 = $12.8K.

So you are risking $12.8K to make $78K. But I would compare it as $12.8K vs the adjusted average expected return ($7.8K).

The amount (and degree) of variables is so large that the exercise is essentially worthless. How can you estimate the share price or probability of exit?

You can't really but if you were to think about it that's how I would go about it. You can test different values and probabilities that you think are (likely|conservative|optimistic) and go from there. That gives you a range and a ballpark for different scenarios.

Personally though, if I can't predict something with > 90% accuracy/reliability then I'm not interested in investing in it.