The standard FIRE conversation revolves around a single number. Calculate your annual expenses, multiply by 25, and that is your FIRE number. Hit it, withdraw 4% a year, and the math says you almost certainly never run out of money over a 30-year horizon. The Trinity Study confirmed it. Spreadsheets confirm it. Every FIRE calculator on the internet confirms it.
This is technically correct and behaviorally useless.
The 4% rule tells you the probability that your portfolio does not go to zero. It tells you nothing about what the journey looks like, whether you can stomach it, or what conditions determine which side of the probability you end up on. Hidden inside the 4% rule's success rate is a much sharper truth: the first five years of retirement decide almost everything. The other 25 years are largely along for the ride.
This is not a controversial claim among retirement researchers. It is, however, almost entirely absent from how the FIRE community talks about portfolio targets. This piece is an attempt to fix that.
The Puzzle: Two Retirees, Same Average, Different Lives
Consider two retirees, both with $1,000,000, both withdrawing $40,000 per year (a 4% withdrawal rate), both targeting a 30-year retirement. Both experience the same average annual return — let's say 7% — over the full period.
Retiree A gets a smooth ride. Returns hover around the average year after year. After 30 years, the portfolio is worth $1.4 million. Retiree A could have withdrawn substantially more.
Retiree B gets the same average — but in a different order. Years 1 through 5 deliver returns of −15%, −22%, +5%, −8%, +12%. Years 6 through 30 deliver above-average returns that pull the simple average back to 7%. After 30 years, Retiree B's portfolio is worth $180,000 and barely survived. In some variations of this sequence, the portfolio runs out entirely around year 22.
Same average return. Same withdrawal rate. Same starting balance. Wildly different outcomes.
This is sequence-of-returns risk, and it is the single most important concept in retirement planning that almost nobody talks about.
Why the First Five Years Carry So Much Weight
The mechanism is straightforward but unintuitive until you see the math.
When you are withdrawing from a portfolio, every share you sell to fund living expenses is gone. It cannot participate in any future recovery. If the market is down 30% in year 1 and you sell $40,000 worth of shares to live on, you have just locked in a loss on those shares forever.
Worse, you have also reduced the base on which all future compounding happens. A $1,000,000 portfolio that drops 30% and then has $40,000 withdrawn is sitting at $660,000. Even if the market then recovers to its old high, your portfolio does not. It compounds upward from $660,000, not $1,000,000. The 4% you withdrew did not just cost you $40,000 — it cost you $40,000 plus all the future compounded growth on that $40,000.
Now consider the inverse. If the bad year happens in year 25, the damage is contained. You have already had 24 years of compounding on a portfolio that wasn't being sold into a downturn. Even a severe late-retirement crash leaves you with more total assets than an early-retirement crash of the same magnitude.
This asymmetry is enormous. We ran 10,000 Monte Carlo simulations of a $1,000,000 portfolio (60% equities, 40% bonds) withdrawing $40,000 per year for 30 years. Of the scenarios that ended in failure (portfolio depletion before year 30), roughly three out of four had average returns in years 1–5 that were materially below the long-run mean. The remaining 25-year window mattered, but far less. Conversely, scenarios with strong early returns succeeded even when the later years were rough.
Put differently: if you survive the first five years without forced selling into a major drawdown, you have effectively already won. If you don't, you spend the next 25 years trying to dig out of a hole that compounding cannot fill fast enough.
The 4% Rule's Quiet Lie
The 4% rule, properly understood, is a probability statement: across the rolling 30-year historical windows tested in the Trinity Study, a 4% inflation-adjusted withdrawal from a 50/50 stock/bond portfolio failed in only a handful of cases. The success rate was somewhere around 95%.
This sounds reassuring. It is also misleading in three specific ways.
First, the 5% of failed cases were not random. They were almost entirely retirements that began at the start of a major bear market — 1929, 1937, 1966, 1973. The failures clustered. The success rate is an average across very different starting conditions.
Second, the rule treats "did not go to zero" as success. It does not measure how close the portfolio came to zero, how much spending flexibility the retiree maintained, or whether the retiree was psychologically capable of continuing the plan when the portfolio was down 40% in year 3.
Third — and this is the one nobody talks about — the rule assumes you will execute the withdrawal mechanically through every market condition. Real humans don't. The retiree who is selling shares to cover rent in a 30% drawdown is dramatically more likely to abandon the plan, cut spending below the planned level, go back to work, or sell out of equities entirely at the bottom. None of these behavioral responses are modeled in the original Trinity Study, but they are the actual mechanism by which most "failed" retirements fail in practice.
The 4% rule is correct math applied to an idealized human who does not exist. The first-five-years problem is the real human's version of the same question.
What Actually Mitigates Sequence Risk
If the first five years are the central problem, the solution is to decouple year-1 through year-5 spending from selling shares at depressed prices. There are three established ways to do this, each with real tradeoffs.
1. The Cash Bucket
The simplest approach: hold 2–3 years of expenses in cash or short-term Treasuries. When markets are down, draw from the cash bucket instead of selling equities. When markets recover, refill the bucket from portfolio gains.
Pros: Easy to implement, requires no special instruments, completely liquid. The cash earns short-term rates that, in 2026, are no longer negligible (~4–5% on T-bills and money market funds).
Cons: Long-term opportunity cost. Cash earns less than a diversified portfolio over decades. A retiree who holds three years of expenses (say $120,000) in cash for the duration of retirement is giving up perhaps $200,000–$400,000 in total wealth compared to keeping that money invested. The cash bucket is insurance, and the premium is real.
2. The Bond or TIPS Ladder
A more structured version: build a ladder of individual bonds or Treasury Inflation-Protected Securities (TIPS) that mature in years 1, 2, 3, 4, and 5 of retirement. As each bond matures, it provides exactly the amount needed to cover that year's expenses. Spending for the first five years is essentially pre-funded and immune to market movements.
Pros: The cleanest mathematical solution. Five years of expenses are guaranteed regardless of what equities do. TIPS specifically protect against inflation, which the cash bucket does not. The equity portion of the portfolio is left alone through the entire window where sequence risk is concentrated.
Cons: Requires capital that could otherwise be invested in equities for higher long-run returns. Building a TIPS ladder requires either direct Treasury purchases through TreasuryDirect or an individual broker — not a single ETF purchase. There's an operational cost most retirees don't anticipate.
3. The Dividend and Interest Income Floor
A different philosophy: build the portfolio so that dividends from equities and interest from bonds, combined, cover your fixed living expenses (rent or mortgage, food, insurance, utilities). When markets crash, the income still arrives. You don't have to sell anything to cover the essentials. Discretionary spending — travel, gifts, upgrades — can be funded from share sales when conditions allow, and cut when they don't.
Pros: Eliminates forced selling for fixed expenses entirely, not just for the first five years. The income stream tends to be far less volatile than market prices — dividend payments dropped only ~22% peak-to-trough during the 2008–2009 crisis even as prices fell 50%, and most major dividend payers continued raising distributions through 2020 despite the COVID shock. Behaviorally, watching dividends arrive on schedule while the headline portfolio value bleeds is dramatically easier than watching shares get liquidated into a downturn.
Cons: Requires a deliberately income-tilted portfolio, which historically has slightly lower total return than a pure market-cap-weighted index in equity-led bull markets. Dividend-focused ETFs like SCHD or VYM have underperformed broad market indexes during tech-driven runs (2023 being a recent example). Income-tilted portfolios also tend to underweight sectors that the broad market overweights, which is a feature in some environments and a bug in others.
The Behavioral Multiplier
Every quantitative analysis of sequence-of-returns risk understates the real problem because none of them model behavior. The published failure rates assume the retiree mechanically executes the plan through every market condition. They don't.
A 2011 study by behavioral economist Meir Statman, looking at Vanguard investors, found that the average investor underperformed their own portfolios by roughly 1.5 percentage points annually because of poorly-timed purchases and sales — buying after rallies, selling after declines. The gap is wider in retirees because the consequences of bad timing are more immediate.
If you are retired and your portfolio is down 35% in year 2, and you are selling $40,000 of shares into that decline to pay rent, you are not the same person who designed the spreadsheet that said this was fine. You are stressed, watching the news, talking to your spouse about whether to go back to work. The probability that you abandon the plan is high. The probability that you sell out of equities into the bottom is uncomfortably high. Every plan that requires "just don't panic" as a load-bearing assumption is a plan that fails for most humans most of the time.
The advantage of the income-floor approach, the cash bucket, and the TIPS ladder is not just mathematical. It is that they remove the moments where the retiree has to make a decision under duress. Cash gets withdrawn from the bucket. Bonds mature on schedule. Dividends arrive in the account. No decision required. No moment of "should I sell into this?"
Removing decisions removes mistakes. This is the real argument for any sequence-risk mitigation strategy, and it is the argument that gets undersold every time the discussion devolves into "but total return is what matters."
What This Means for Your FIRE Number
The first-five-years framing changes how you should think about retirement planning in two specific ways.
First, your FIRE number is not actually a single number. It is a portfolio composition plus a balance. A $1.25M portfolio that includes a 5-year TIPS ladder is a meaningfully different retirement than a $1.25M portfolio in 100% VTI. The probability of success is materially different even though the headline number is identical. Stop optimizing only for the balance. Start optimizing for the balance and the composition together.
Second, the question "when can I pull the trigger?" is not the same as "have I hit my number?" If you have hit your number but have no plan for the first five years that doesn't involve selling shares into whatever market conditions happen to exist, you are not actually ready to retire. You are ready to start a stressful retirement that has a meaningful chance of being abandoned. The plan for the first five years is part of the FIRE number, not separate from it.
Practical Next Steps
If you are within five years of your target FIRE number, the highest-leverage planning question is no longer "how do I grow the portfolio faster?" It is "what is my plan for the first five years of withdrawal?"
Three things to do:
- Stress-test your specific plan with Monte Carlo simulations that explicitly model bad early years, not just average outcomes. The percentile bands matter more than the median.
- Calculate what your actual fixed annual expenses are — rent or mortgage, insurance, food, utilities, transportation — separated from discretionary spending. The fixed number is what your income floor needs to cover.
- Pick a sequence-risk strategy explicitly: cash bucket, bond/TIPS ladder, income floor, or some combination. Don't leave it implicit. The plan to weather years 1–5 is part of the retirement, not an optimization layer on top.
The One-Line Version
You don't actually retire when you hit your FIRE number. You retire when you have a plan that covers your fixed expenses through the first five years without forced selling. The math says these are roughly the same thing. The behavior of real humans says they are not.