Yeah, I already have decided to press on for 2 or 3 more years before retiring, especially given the current collection of strange influences on the economy. Trouble is that FIRECalc suggests I might just make it if I pulled the plug right now.
Anyway, I'm intrigued by my little test. If I add the fractions to the whole years, FIRECalc seems to do what you'd expect when compared to the result just using the whole numbers - a "middle" of the year calculation results in larger high end and average $$'s than the whole year, and less than the values when using the next whole year.
Once an engineer, always an engineer I guess.