No one is suggesting FIRECalc is predictive. We all already know it’s not. It only shows how one's assumptions would have stood the test of time. My example shows that so far, I’ve hit a favorable period. My results are within the range of possible outcomes.I don't think anyone would expect that the next 30 years' ups and downs will be the same ones as we experienced in the last 30 years. This is why none of these calculators can credibly claim to be predictive. FireCalc does not.
From a career in science and engineering, my view is that attempting to assign a precise probability and then use it as if it were predictive is a classic garbage-in, gospel-out scenario. 100% is probably the same as 90% and maybe even the same as 80%. @MichaelB's experience illustrates. If I were to write one of these planners it would probably have three possible outputs: "Doesn't Look Good," "Pretty Marginal, be ready to adapt to immediate circumstances," and "Looks Pretty Good, but luck is always at the table." Really, for @MichaelB 85% is working out to be more like 100%, probably buoyed by the recent good market. For the next retiree, 100% may turn out to be more like 85% if the market gives them a bad dose of SORR.
Clint Eastwood/Dirty Harry's famous line pertains to retirement planning: "...you've got to ask yourself one question: 'Do I feel lucky?' Well, do you, punk?"
George Box said it best, “all models are wrong but some are useful”. If used as intended, FIRECalc can be useful.