I tried FIRECalc for the first time, and it's giving me a 100% success rate based on my assumptions. For me, the critical assumption is projecting the value of my investment portfolio in 10 years at the time early retirement "starts." (My wife claims she has no interest in early retirement, so my "safe" assumption is her working 10 more years to age 51, and because she makes more than we spend, our portfolio wouldn't be touched the next 10 years.)

I've been putting in an investment portfolio double the current value because that's the most statistically likely outcome at a 7% growth rate. That gives me the 100% success rate, but even assuming 0% growth between now and then, that comes out at 100% too.

The issue is how to avoid double counting the worst case market scenarios? If the market drops by 50% in the next 10 years and I put in an investment portfolio value 50% of the current total, how can I discount the "failures" in FIRECalc that assume the worst historical periods?

In other words, we're not (likely) going to have a 20-year period resulting in a 75% market drop. And because the sequence of return risks always seem to focus on the first few years of retirement, avoiding double counting the "worst time ever" seems important.

Thoughts? I'm not interested in modeling sky-is-falling assumptions or 2% safe withdrawal rates or anything like that. I'm trying to make an informed decision about what to do based on most likely outcomes (which includes my wife working--but that's a separate assumption that I can model).

I appreciate the help.