The calculator
Presets: named crash years from this site's S&P 500 price-return tape. Each fills the crash field, and you can still type any other return.
Identical 5.0% years and one -30.0% year, in opposite orders, differ by $427,546 at the end.
Why the order matters
Two people can earn the same average return and finish in different places, because a withdrawal takes a larger share of a smaller portfolio. Selling after a fall locks the loss in; the same fall at the end hits a pile that has already compounded. On this calculator's defaults, $1,000,000 paying $40,000 a year for 20 years ends with $415,472 when a 30% crash comes first and $843,018 when it comes last. The glossary entry on sequence of returns risk is the definition; the sequence of returns risk guide is the pillar, including what a crash in the year you retire does to a whole plan.
Why sequence risk matters when withdrawing, not while saving
During the accumulation phase, with no money going in or out, the order of returns makes no difference at all. $1,000,000 left alone through nineteen 5% years and one 30% crash ends at $1,768,865 whether the crash comes first or last, because multiplying the same returns in a different order gives the same product. While you are still contributing, an early crash even helps: new savings buy shares cheaply.
Withdrawals break that symmetry. Each year you sell a fixed dollar amount, so after a fall you sell more shares to raise the same cash, and those shares are not there for the recovery. That is why sequence of returns risk is a retirement problem, and why it peaks when the portfolio is largest: the years just before and just after you stop working.
Sequence of returns risk example
The same $1,000,000 portfolio, $40,000 withdrawn each year for 20 years, and 5% in every year but one. The one bad year is a real S&P 500 calendar year, placed first or last:
| Crash year | Return that year | Crash first | Crash last | Gap |
|---|---|---|---|---|
| 1931 | -47.1% | $1,461 | $637,485 | $636,024 |
| 1974 | -29.7% | $422,311 | $846,413 | $424,102 |
| 2002 | -23.4% | $576,407 | $922,914 | $346,507 |
| 2008 | -38.5% | $209,620 | $740,824 | $531,204 |
Every row has the same average return and the same withdrawals. Only the order changes, and the order alone moves the ending balance by hundreds of thousands of dollars.
Sequence of returns in real retirements since 1928
The table above isolates order with one crash. Real markets deliver a whole sequence, so we replayed every 30-year retirement since 1928: $1,000,000 in a 60/40 stock and bond portfolio, 4% withdrawn in year one and raised with actual inflation every year. Balances are in start-year dollars.
| Retired in | After 5 years | After 10 years | After 30 years |
|---|---|---|---|
| 1929 | $735,000 | $821,000 | $392,248 |
| 1966 | $750,000 | $475,000 | Ran out in year 26 |
| 1973 | $565,000 | $388,000 | $95,620 |
| 1949 | $1,319,000 | $2,164,000 | $2,021,851 |
| 1975 | $932,000 | $974,000 | $2,530,072 |
| 1982 | $1,733,000 | $2,084,000 | $3,477,593 |
The worst sequence was not the biggest crash. Retiring in 1929 meant the Great Crash, yet prices fell too, so inflation-linked withdrawals shrank and the money lasted. Retiring in 1966 meant no crash in the first years, just flat stock prices and rising inflation, then the 1973 to 1974 bear market on top. That combination ate the portfolio, and it ran out in year 26. The 1982 retiree took exactly the same withdrawals, started into a long bull market, and finished with more than three times the starting value after inflation. Overall the plan lasted 30 years in 94.1% of start years, with every failure starting between 1965 and 1969. The historical backtest runs your own plan through every one of these years.
Pick a retirement year
Presets for historical sequences: choose a start year, or type any year from 1928 to 1995, to see the same $1,000,000 plan through the returns and inflation that actually followed.
Retiring in 1966 with $1,000,000 in a 60/40 portfolio and a 4% inflation-adjusted withdrawal, the money ran out in year 26.
The first five years: the retirement red zone
Planners call the five years either side of the retirement date the retirement red zone, because the portfolio is near its peak and withdrawals are starting, so a loss there does the most lasting damage. The replay above shows why. 6 start years were worth less than half their starting value, in real terms, ten years in, and every retirement that ran out of money was one of them. A bad first decade does not guarantee failure (1973 dropped that far and still lasted 30 years), but every start year still worth more than half after ten years lasted the full 30.
The practical point is timing. The defences below cost something, either lower expected returns or lower spending, and they are worth paying for mainly in those first five to ten years. After a decade of withdrawals that went well, most of the risk has passed.
How to reduce sequence of returns risk
- Spend flexibly after bad years. Rules such as Guyton-Klinger guardrails and Vanguard dynamic spending cut a little when the portfolio falls, so fewer shares are sold at the bottom. The retirement guardrails calculator helps choose a rule.
- Hold a cash buffer. A few years of spending in cash and bonds means stocks do not have to be sold after a crash; the bucket strategy calculator sizes it.
- Build a bond tent. Raise the bond share in the years around retirement, then let stocks drift back up over the first decade. Wade Pfau and Michael Kitces called this a rising equity glide path (2014): it holds the most bonds exactly when a crash would do the most damage, and less once the red zone has passed.
- Start lower. A first-year withdrawal below 4% leaves more room for a bad start; the safe withdrawal rate calculator shows the trade-off, and the retirement withdrawal calculator shows how many years each withdrawal lasts.
- Keep some income early. Part-time work in the first years shrinks the withdrawals that a crash can lock in; the Barista FIRE calculator shows how much.
Assumptions
One crash year and a flat return everywhere else: enough to isolate order, not a forecast. No fees, no taxes, withdrawals at the start of each year. A full plan needs a distribution of paths, not two: the Monte Carlo retirement calculator. Spending rules that cut after a bad year are the Guyton-Klinger calculator and the Vanguard dynamic spending calculator. The first-year cheque is the safe withdrawal rate calculator.