[naterator]

Shouldn't there be some kind of formula for this kind of thing? Some kind of way to gauge how much salary/year you should give up based on how likely you think they will be successful. For example,

[% annual salary haircut] = [likelihood success in 10 years] * [estimated stock price] * [expected salary sans options]

... or something. Obviously it's far from a science, but I think there should be some kind of general way to arrive at reasonability. Currently start-ups essentially rely on how poorly people gauge the probability that a company will actually be successful vs. actual numbers. Or, put another way, how willing the hiree is to gamble/risk.

I only bring this up because, after personally seeing or being on the losing end, I would never accept the salary haricut for options again unless I could somewhat reliably gauge them, probabilistically. Maybe that's something I'll look into for the future. People do this kind of thing in the Insurance industry and on wall street all the time, I suppose.

Sidenote: Paul, thanks for all the really cool stuff you've done over the years.

[lutusp]

> Shouldn't there be some kind of formula for this kind of thing?

It's not like science or math. There are too many subjective issues at work -- a relatively new company in a fluid marketplace, issues of personal psychology, guessing how applicants will respond to the prospect of options. Any fixed formula or policy would likely be invalidated by events in a short time. It reminds me of the old saying, "The first casualty of battle is the battle plan."

> I only bring this up because, after personally seeing or being on the losing end, I would never accept the salary haricut for options again unless I could somewhat reliably gauge them, probabilistically.

Yes, and this is because you have experience. A younger applicant might not understand the issue as well, or might have heard a statistically improbable but dramatic story of someone becoming rich on options.

From the management side, they might guess at the age and experience level of the applicants, and adjust their pitch accordingly.

> Paul, thanks for all the really cool stuff you've done over the years.

You're most welcome. Here's my latest article: http://arachnoid.com/IPython

[dllthomas]

If you treat it as a gamble, you can look at expected value (to determine whether it makes sense to play at all) and something like the Kelly Criteria (though my understanding is that it doesn't apply directly) to determine how much it makes sense to be betting. All of that, as Paul said, must be adjusted for your personal risk preferences as determined by your personality and situation, and you should keep in mind that your assessment of the odds is likely wrong (or you should be a VC).