Flawcastr
Solve
Leaving everything else as it is, what would bring your investment assets as close to $0 as possible at the end of the modelling period?
Click below to see what level of retirement spending, annual savings, or retirement age (to the nearest whole year) would do it. If you like the result, Add scenario creates a new scenario with the solved value and leaves the one you are editing unchanged.
This uses your entered return assumptions only, not the Monte Carlo lines. While this panel is open, only the scenario you are editing is drawn on the chart. Do not rely on this calculation. It is provided as a reference only.
Savings and retirement age can't be solved because you're already retired.
Monte Carlo settings
Your main projection applies the same investment return every year. Monte Carlo draws extra lines that vary the returns year by year, so you can see how much the order and size of returns matters.
The lines are drawn for the scenario you are editing only. Only one method is used at a time.
For this method, fifty lines are drawn. Each line runs through your projection one year at a time, with the same savings, spending and one-off items as your main projection. The only thing that changes is the investment return in each year.
Each year, instead of applying your return rate exactly, a line applies a random return that is centred on your rate but can come in higher or lower. If your balance sits across your investment threshold, the centre is the blend of your two rates, just as in the main projection.
How far a year's return can stray from your rate follows a bell curve: most years land close to your rate and fewer land far from it. The curve is drawn so that its width (one standard deviation) is twice your rate, and never less than 2 percentage points. For example, with a rate of 3% the mean is 3% and the standard deviation is 6%. With a rate of 0.5%, the mean is 0.5% and the standard deviation is 2%, because twice the rate would be only 1% and the 2 percentage point minimum applies instead.
The randomness is in each year's return, not in the ending balance. Each line is simply what happens when one particular run of good and bad years is added up.
This is deliberately simplified. It is not a statistically accurate model of investment volatility. Real markets do not follow a bell curve, and the two-times-rate rule is a convention chosen for this tool, not a measured figure. The point is only to show that these are flawcasts, not forecasts: year-to-year variation alone can change your outcome substantially.
Each line replays actual year-by-year US market returns from 1871 to 2025, already adjusted for inflation.
This is based on the publicly available long-run series compiled by Robert Shiller (Yale University), covering the S&P Composite share index and 10-year US Treasury bonds, published at shillerdata.com. It is not a prediction, and you should not rely on it.
US data is used because New Zealand market history does not go back far enough.
One line starts in 1871, the next in 1872, and so on, with one line for every year in the data. Each line runs through your projection with the same savings, spending and one-off items as your main projection, but with each year's return taken from history in the order it really happened. Lines starting too recently to cover your full modelling period are shorter and simply stop early.
Your main projection line still uses the investment return percentages you entered. The Monte Carlo lines ignore them: history supplies their returns instead.
Your investment threshold still applies to those lines. It decides how much of your balance is treated as defensive (bonds and cash) rather than as shares:
- The balance up to the threshold is treated as entirely defensive. Defensive money is split equally between bond returns and cash (earning 0% after inflation).
- The balance above the threshold is treated as 80% shares and 20% defensive.
- The mix is reset to these proportions at the start of each year.