Introduction to cFIREsim

crystal-ballAs you may be able to guess, I like to play with numbers and spreadsheets as much as the next dropout engineer.  One question that I know a lot of us looking for financial independence ask ourselves is “how much will be enough?”.

Many financial calculators spit out “the number” based on basic forward-looking assumptions, like an average return on investments. Sure, the U.S. stock market has returned an average of 10% since the mid-1920’s but it certainly hasn’t been a smooth ride.

I love the figure below to illustrate this.  Graphically, these are the returns of the U.S. stock market from 1926 to 2012.  The question at hand is this:  How many years did the market return the average plus or minus (+/-) 2%?  In other words, in the last 86 years how many of those years had “normal” 8-12% returns?

U.S.StockReturns 1926 to 2012You can see that in almost nine decades, a whopping 6 years were in that “average” range (highlighted in blue).  Not exactly consistent or comforting.  As I detailed in a past post, I’ll be watching my portfolio very closely the first couple of years into retirement because the returns in those opening years can really make or break us.

A Different Question

Most calculators are good for coming up with an initial educated guess as to the amount of money we’d need to reach financial independence.  I’ve shared our goal of reaching $1.25MM with no mortgage.

But there is another question that I believe is even more important.  How well would our retirement portfolio hold up over time given the historical returns of the stock market?

It seems to me if I had a portfolio that could weather anything thrown at us in the past 140 years, I’d feel a lot better about it going forward.  I know past performance has no bearing on future results.  But the U.S. stock market has seen many booms and busts in its history including a depression, multiple recessions, and world wars.  I expect the future to be just as volatile.

Introduction to cFIREsim

I want to tell you about a new tool.  It’s called The Crowdsourced FIRE Simulator (cFIREsim for short) and is the next generation of the venerable FIRECalc.  It takes inputs of your savings, spending, and length of your retirement and churns through as many cycles as possible to tell you the likelihood that your nest egg would last given different retirement dates throughout history.

cFIREsim is tricky to explain in words, so I’ll show an example given some numbers and assumptions.

  • Portfolio:  $1,250,000
  • Spending:  $50,000 per year
  • Retirement Year:  2014 (for illustration purposes)
  • Years to Model:  40
  • Spending:  Inflation Adjusted to Consumer Price Index (CPI)

cFIREsim - Simulation - GraphicalcFIREsim looked at 103 possible 40-year periods of the available data and concluded this portfolio would have failed 17 times for an 83.5% success rate.  You can see the individual lines that fall below the $0 portfolio water line.  Failure is defined as having run out of money before the end of the 40-year retirement.

It also spits out other portfolio-ending metrics like:

cFIREsim - Simulation - Metrics

Benefits of cFIREsim

“Big deal.  How is this different than FIRECalc?”

cFIREsim has a several advantages and new features over its counterpart.  My favorite being the ability to “investigate” the results.  In my current example that has a 83.5% success rate, I can get answers to the following questions with the click of a button:

Q1:  How big of a portfolio is needed to support $50,000 spending level with a 100% success rate?

A1:  $1.57MM

Q2:  What spending level is supported for a portfolio of $1.25MM at a 100% success rate?

A2:  $40,000 per year.

Another benefit cFIREsim has over FIRECalc is the ability focus in on certain dates or date cycles.  I can run simulations using all data available (from year 1871 to present) or I can enter in a custom year range (for example 1926 to 2012).  I can also execute a single cycle given a certain starting year (what does the curve look like if I had retired in 1960?).  The simulation cycle graph is interactive in that I can highlight a particular date range of interest and zoom-in on that data.

I’m just scratching the surface in terms of all the input variables at our disposal when using this tool not to mention the various Spending Methods there are to choose from to model different draw-down strategies.

I’d encourage you to have a look at this ground-breaking retirement simulator.  If you’re anything like me, you have to be careful as it can become a huge time sink running different scenarios to see how they fare.

I’ve been in limited contact with the creator and he has been very open to feedback so be sure to log any enhancement requests or questions you may have to the cFIREsim forum.

**Addendum – Click here to see my interview with Bo, the creator of cFIREsim.

6 thoughts on “Introduction to cFIREsim”

  1. Buck, great work here! I appreciate the good comments. I think one of the places where cFIREsim is starting to shine is the “Adjust spending” models, which give certain rules to your annual spending to make it flexible and more likely to succeed. Following these rules can make it easier to ER, because you might need a little less $ otherwise. Beware, though, some of the final tweaks to these models is still in the “development” version of cFIREsim. I’ll be moving it to the live version this week 🙂

    1. Hi Bo! Thanks for stopping by. Yes, I agree with you that the ability to adjust the spending models is a powerful feature and one that probably deserves its own post. Great work with this tool – I can only imagine how much work it has been to put it together. I certainly appreciate your efforts.

  2. Holy cow, what an awesome tool! Like you, I obsess with the numbers.

    One comment I’ve seen MMM making frequently is that 100% isn’t what anyone should shoot for. That is just unrealistic. I’m not sure what I will shoot for, maybe 90%?

    The older I get, the less I worry about failure. I could always go out and be a park ranger or just dial way back on the spending for a couple of years if things really took a dump.

    1. “100% isn’t what anyone should shoot for” – I agree with that sentiment but I think it is exceptionally cool that with this tool you can know what that metric is just by simply clicking a radio button.

Your Thoughts?