FireCalc-same input, different results?

[sandman1959]

Hi all,

Just joined the forum and this is my first post. I have been using firecalc but I noticed that when I input the exact same input, I get different results. For example, I enter Start here - Spending-$122,000, Starting amount $2,183,443, 40 years. Other income - SS of $28,536 starting 2026. Not Retired - Retire in 2026. Spending models-Inflation using CPI, Constant spending power. Your Portfolio - Expense ratio of 0.7%, Random performance with portfolio return of 7.89%, SD of 15.56%, Inflation at 3%.
I have gotten the following "success" results with this input - 83.9%, 53.6%, 94.8%, 90%. What the heck am I doing wrong? Thank you in advance.

[REWahoo]

Quote:

Originally Posted by sandman1959 Random performance...

I think that's your issue.

Try allowing FIRECalc to use actual history instead of a Monte Carlo calculation - use the Total Market or Mixed Portfolio option.

[sandman1959]

Quote:

Originally Posted by REWahoo I think that's your issue. Try allowing FIRECalc to use actual history instead of a Monte Carlo calculation - use the Total Market or Mixed Portfolio option.

ReWahoo,

Thanks, your suggestion now gives me a consistant output. But, your reply then causes me to ask why the Random Input (monte carlo) even exists if it gives you a different percentage success every time? I have used other financial calculators utilizing monte carlo and they do not give a different result every time. What am I missing here?

Thanks

[clifp]

Well I am not sure how the other calculators work. But you can imagine that your probabilities of success are going to vary greatly depending on when you get a 1995-2000 bull market, a 1929 crash, or a 2008-2009 crash.

Instead of historical results Firecalc will randomly generate a portfolio return which 2/3 of the time will be between -8% and +23% for each of the hundred odd years

[REWahoo]

What clifp said.

Using the random option bypasses the historical sequence of market returns that is the core of how FIRECalc works. (Take a look here and scroll down to "How about describing FIRECalc step by step?")

When you use the random option some runs may have multiple bad years following downturns (think year after year of 1929 market crashes) or a string of 2013 market booms. End result - highly variable rates of success.