Why the average path is not enough

Asha and Vikram retire with the same corpus, ₹3.00 Cr. Each takes out ₹18.00 L a year. Over the next ten years their investments earn the same ten yearly returns, with the same average of 9%. The only difference is the order. Asha gets the two bad years first. Vikram gets them last.

If neither of them had taken anything out, they would have ended with the same amount. Multiplication doesn't care about order. Withdrawals do. Money taken out after a fall is gone before the recovery arrives, so the recovery has less to work on.

This has happened

William Bengen, whose research produced the 4% rule, lists the worst stretches for American investors. In 1973 and 1974, shares lost 37.2% while prices rose 22.1%. Someone who had just retired was selling shares at low prices to pay bills that were rising fast. Bengen found that the damage reached back to people who had retired years earlier.

What an average hides

A plan built on one steady return describes the middle path. About half of the futures you could get are worse than the middle. A plan that works on the average path fails about half the time. That is not a plan most people would accept for their only retirement.

What the calculator does about it

It tries 400 futures. In each one, every month's return is the return you entered, moved up or down by chance. How far it moves depends on how bumpy you said your investments are. Bad months come in stretches, as they do in markets, so a future can hold a fall that lasts more than a year.

For each future the calculator works out the corpus that future would need on the day you retire. Then it asks how sure you want to be, and picks the corpus that is enough in that share of futures. That corpus is your FIRE number.

Each step up in certainty costs more than the one before. The last few futures are the worst ones, and covering them takes a lot of extra money. You are choosing how much of your working life to trade for protection against futures that may never come.

When you get there

Markets also decide how long it takes to reach the number. In the futures where your investments do well early, you get there years sooner. In the ones with a bad stretch in your forties, you get there years later. So the calculator gives a date and a range. The date is for a future of average luck. The range holds 8 of every 10 futures. The number is fixed. The date moves with the markets.

How to read the chance

A chance of 80% does not mean the plan is 80% good. It means that in 80 of every 100 futures tried, every month was paid and the legacy was left. In the other 20, something had to give: spending, the legacy, or the retirement date. A person living through one of those futures could see it coming and spend less for a while. The model doesn't do that. Once you retire, it assumes you spend the same amount whatever happens.

That makes the chance a strict test. If you would cut back in a bad stretch, your own chance is better than the one shown. The stretches in the model follow figures this site chose, and are not a record of any market. The 4% rule and what it was measured on covers the studies that used real market history.

See the range of futures on the timeline