A friend of mine spent a weekend building a retirement spreadsheet. He was proud of it — colour-coded tabs, compound interest formula, the works. It told him he'd retire comfortably at 58. Three years later he ran the same numbers through a proper retirement calculator that accounted for sequence of returns risk, and the honest answer was closer to 63. Same salary, same savings rate, same average return assumption. Five years, gone. The spreadsheet wasn't wrong about the maths — it was wrong about the question.
Most retirement calculators make the same mistake — and it's one that even the best-reviewed ones get wrong when it comes to sequence of returns risk. They take your savings, apply an average annual return, and project forward in a straight line. It's tidy, it's legible, and it's misleading in a way that can genuinely derail your plans if you're not paying attention.
The problem with average returns — and why sequence matters more
Here's the uncomfortable truth about averages: a portfolio that earns +20% one year and -20% the next doesn't break even. It ends up down 4%. Your spreadsheet probably doesn't know that.
This isn't a niche academic quirk — it's one of the most practically important concepts in personal finance, and it has a name: sequence of returns risk. The order in which your returns arrive matters enormously, especially in the years just before and just after you retire. If markets crash in year one of your retirement and you're drawing down to pay your living expenses, you're selling units at exactly the wrong time. That damage doesn't recover even when markets bounce back, because you have fewer units left to recover with.
To put some numbers on it: imagine two retirees, each with a $1.5 million portfolio and each drawing $75,000 a year. Retiree A gets good returns in the early years and poor ones later. Retiree B gets the same returns in reverse order — poor early, good later. Same average return. Same withdrawal. After 25 years, Retiree A still has money in the portfolio. Retiree B ran out at year 18. That's sequence of returns risk in its starkest form, and most calculators — including most expensive ones — don't model it at all.
The other thing calculators routinely get wrong is the assumption that you'll earn the same return throughout retirement. In practice, most people shift from growth-oriented investments in their 40s and 50s toward more conservative allocations as they age. A calculator that uses one flat return rate for a 40-year retirement is describing a financial life nobody actually lives.
What a good retirement calculator actually needs to do
A retirement calculator worth trusting needs to do at least three things that most don't.
First, it needs to let you model different return expectations across different life stages — not just one average for the whole retirement. Your portfolio's risk profile at 55 is genuinely different from its profile at 80. A calculator that treats them identically is telling you a comfortable story rather than a useful one.
Second, it needs to stress-test the sequence, not just the average. What happens to your plan if you hit a bad market run in your first five years of retirement? That's the scenario that breaks portfolios, and you want to know your exposure to it before it happens — not during.
Third, and this one is underrated: it needs to model what you can actually spend. Most calculators tell you whether your planned withdrawal is sustainable. Fewer tell you what the maximum sustainable withdrawal actually is, given your specific situation. That's a different and often more useful question — especially for people who've been conservative and are quietly over-saving.
If you've never run a sequence stress test on your retirement plan, that's the first thing worth doing — before adjusting your savings rate, before changing your allocation, before anything else. I've found that people are often surprised in both directions when they run these numbers properly. Some discover they've been assuming they can spend more than the portfolio will realistically support under stress conditions. Others — probably the majority, in my experience — discover they've been far too conservative and have more latitude than they realised. Neither discovery is possible with a straight-line average-return model.
How to stress-test your retirement plan properly
The tool at The Artifacts Lab was built specifically to address these gaps. It's free to try and doesn't require an account.
The core of it is a five-period model — you set your expected return for each stage of your retirement separately. Early retirement, mid-retirement, late retirement, and so on. You can dial in a growth-oriented assumption for your 60s and a more conservative one for your 80s, which is much closer to how people actually invest.
On top of that, the sequence of returns stress test runs your plan in two orderings: as entered (your plan), and worst-first (the pessimistic scenario). If you want to know whether your retirement plan survives a bad run early on, this is the most direct way to find out. The gap between the two outcomes is your sequence of returns risk, expressed in actual dollars.
There's also a withdrawal optimizer that works backwards from your assets to tell you the maximum sustainable withdrawal given your inputs, your legacy target (if any), and your tax situation. In practice, this is the number that most tools avoid giving you directly because it requires doing the hard work.
The tool at The Artifacts Lab runs this five-period model and stress-tests the sequence automatically. You can try it free at theartifactslab.com/retirement-calculator — no account needed
If you've previously used a spreadsheet or one of the more basic free calculators and felt vaguely reassured by the result, it might be worth running your numbers again with something that takes sequence risk seriously. Not to scare yourself — but because an optimistic model that doesn't survive contact with reality isn't actually reassuring. It's just comfortable.
My friend eventually revised his retirement plan. He didn't panic, he didn't make dramatic changes — he just adjusted his savings rate slightly and shifted his asset allocation timeline. Small changes, made early, because he had an honest picture. That's what the right tool is supposed to do.
If you try it and find something that doesn't make sense — or if there's a scenario you wish it could model — I'd genuinely like to hear about it. The best improvements to this tool have come from people asking questions it couldn't yet answer.
risk — the factor that can silently wreck your plan. Here's what to look for in a better one.*