Ignoring time considerations and otherwise oversimplifying, [Expected return] = [Probability of a non-zero return] * [Expected return | Return > 0] VCs tend t

Ignoring time considerations and otherwise oversimplifying, [Expected return] = [Probability of a non-zero return] * [Expected return | Return > 0] VCs tend to think that the first factor will always be low, so they want the second one to be high. So far, so good. But now let's complicate things a bit more. 1. Actually, the set of all possible outcomes is partitioned into at least three sets: {Zero or very low payouts} {Decent but not great payouts} {Huge payouts} Especially in the middle case, VCs' interests may not be aligned with founder/employees', because of the preferred/common stock distinction and some onerous terms that VCs impose on deals. Also, VCs' benefits aren't just cash, but also reputational, which is another reason why their interests aren't perfectly aligned with companies' ... ... yet past the earliest stages, VCs tend to control the board, and run things for their benefit. 2. It's possible to have other kinds of interest than the classical VCs'. For example: A. If you lend against genuinely good collateral, you have a good chance of getting your money back, and can be more restrained in what piece you take of the upside. B. Seed investors who offer notes with unclear conversion prices often are doing something similar, but with worse odds. Often, they're hoping for a chance to invest in the next round of the best companies, which makes sense only if they assume such an investment will be very beneficial to them. 3. Dave McClure at 500 Startups trumpets the idea of factoring "singles and doubles" into his return calculations.