Reading your results

Your results page answers one question: will my money last? It answers honestly, as a range of outcomes rather than a single verdict. This page explains every number and chart on it, in the order they appear, so you know exactly what each one is telling you and what it isn’t.

A note on tiers: the headline numbers and the fan chart are free; the deeper diagnostics (sensitivity, failure causes, counterfactuals, the spending breakdowns) are premium, and are marked below.

Am I OK?

The top of the page gives you three numbers and a chart, the quickest read on whether the plan holds. If you set an inheritance target, a fourth number joins them.

Success probability

The headline figure: among the simulated futures in which you reach retirement, the share where the money lasts. “Lasts” has a precise meaning: in that future you funded your survival floor (essential spending, housing, healthcare, and any care costs) every year, right up to the last death in the household. A future still counts as a success even if the last euro runs out in your final year; the metric measures whether you made it, not whether you died wealthy.

Futures where you die before reaching retirement are left out of this figure entirely rather than scored as wins. That distinction matters most for underfunded plans: if you’ve saved almost nothing, the only way a “money lasts” test can pass is to not live long enough to spend it, which would inflate the number into something meaningless. By measuring only futures that actually reach retirement, the figure answers the question you care about: if I get there, does the plan hold? Dying during retirement with money still in the account does count as a success, because your money genuinely lasted as long as you did.

Read it as “in 87% of futures, the plan holds”, not as a grade and not as a guarantee. 87% also means roughly one in eight futures fell short, which may be perfectly acceptable or not, depending on how much risk you want to carry.

Terminal wealth

Labelled What’s left at the end · median (today’s €). It’s the median amount remaining at the last death, in today’s euros so it’s comparable to money now: half of futures end with more, half with less. The small band on the tile shows the 25th-to-75th-percentile spread.

The tile follows the today’s-or-future-money switch in the top bar. In future money it shows the balance as it will actually read at the last death, before inflation, which the engine works out life by life: every simulated future ends in its own year at its own price level, so there is no single conversion the page could apply to the today’s-money figure. A result saved before the engine reported that figure stays in today’s money on this tile, and the label says which it is showing.

Futures that ran out of money are counted as zero here, so a low median is a signal in itself. It usually means a meaningful share of futures ended empty, dragging the middle down.

For homeowners, the tile also shows a “plus home equity ≈ …” line: the median value of the home at the last death among futures where it was never sold, in today’s euros. It’s shown as context, not added to the figure, which counts only liquid portfolio money. And when the home-equity fallback fires (see the technical reference), the headline sentence reports how often: “in X% of futures the home is sold to keep the plan funded (median age Y)”, and the median sale age is listed under the fan chart. It is not drawn as a line on it: that age is a median across only the futures where the home was sold, so there is nothing at that point on the bands, which are taken across every future still going.

Inheritance goal

Only shown if you entered an inheritance target. It reports the share of futures that end with at least that much left, in today’s euros.

It is deliberately kept separate from success probability. Running out of money at 82 and dying with €380k against a €400k goal are not the same kind of failure, and averaging them into one number hides which one you’re actually looking at. So your target never moves the success figure. It only fills in this tile.

The tile shows two readings. The big number is the share of all futures that hit the goal. The line underneath gives the share of futures where your money lasted that also hit it, usually the more useful of the two, because it separates “does the plan survive?” from “does it leave what I wanted?”. If the first number is far below the second, the gap is futures that ran out of money, not futures that merely fell short of the goal.

Saved at retirement

Labelled Saved at retirement · median (today’s €). It is the size of the pot you’ve built by the time you stop working. It’s the median portfolio balance in the year you retire, in today’s euros so it’s comparable to money now: half of futures reach retirement with more, half with less. The small band on the tile shows the 25th-to-75th-percentile spread. This is the closest single answer to “how much do I need to have saved?”, though the real test is whether that pot lasts, which is what your chance of the money lasting measures.

In future money the tile uses the median price level at your retirement age, the same one the fan chart uses at that point, so the tile and the chart’s retirement line say the same number. The question mark beside the switch puts that price level into one sentence from your own plan: what a hundred at your retirement age buys today.

The fan chart

The chart traces your portfolio balance over time across all the futures at once.

The centre line is the median outcome. The inner band holds the middle half of futures (25th to 75th percentile); the outer band stretches to the 5th and 95th. A wide fan means high uncertainty; a fan whose lower edge dives toward zero means a real chance of running dry in poorer outcomes. Values are in today’s euros. At each age the bands describe only the futures where the household is still alive, and the chart stops at the age where almost no simulated futures remain alive rather than trailing off on a handful of extreme lifespans. This is the same picture the whole approach is built on; see a range, not a number.

The numbered marks above the plot are moments in your plan: retiring, a pension starting, a mortgage paid off, a period of family support. The list under the chart carries the same numbers with the age and what happened, so you can read every mark without pointing at anything. A filled mark is something your plan sets; a hollow one is the single mark that describes the simulated futures themselves, the age past which half of them have ended. Hover a mark with a mouse to see its label; on a phone, tap it and the label stays open after you lift your finger, placed away from where you tapped, until you tap somewhere else. Dragging a finger across the plot follows the age readout and leaves it where you let go. Moments the chart lists but does not draw (a median age at which something happens across only some futures) appear dimmed and unnumbered, because there is nothing at that point on the bands for a line to claim.

What’s most likely to break?

This panel ranks the inputs that most drive your chance of the money lasting, which are the levers your plan actually hinges on. Each bar has two parts: the first-order effect (how much that input moves your odds on its own) and the total effect (including its interactions with everything else). A tall bar means small changes there swing your outcome a lot, so that’s where attention and effort pay off. The full method is in the technical reference.

How bad is ‘bad’?

Success probability tells you how often the plan holds; this section tells you how it fails when it does.

Success if you live long

Your success probability recomputed over only the futures where you live to an old age (the top of your lifespan range). If this is much lower than your headline number, longevity is your binding risk: the plan works unless you live long, which is precisely the case you most need it to cover. If the two numbers are close, long life isn’t what threatens your plan.

Median shortfall in failures

Shown inside the “How large is the shortfall in the failing futures?” expander. Among the futures that failed, this is the typical size of the failure: the survival-floor spending (essentials, housing, healthcare, care) that went unfunded, summed over the remaining years of life, in today’s euros. A small shortfall means failures are near-misses (you came up a little short, late). A large one means that when the plan breaks, it breaks badly. Always read it alongside your chance of the money lasting: 90% with a small shortfall is a very different risk profile from 90% with a large one. The expander also shows the worst-5% figure, the same sum in the deepest one-in-twenty failures, a tail of the tail.

Lifestyle preserved

Labelled Worst-5% lifestyle preserved. It is how much of your planned spending you actually got to fund, in the unluckiest one-in-twenty futures, added up across retirement. 100% means even your worst futures funded the lifestyle you planned; a lower number means that in bad outcomes you had to trim, and by roughly how much. It’s the gentler cousin of the shortfall figure: shortfall measures outright failures, this measures belt-tightening short of failure.

Years funded

Labelled Worst-5% years funded, and deliberately pessimistic. “Years funded” counts how many years of retirement were covered before the money ran out (or before the end of life, whichever came first). The tile reports the worst 5%: in the unluckiest one-in-twenty futures, your retirement was funded for this many years. It answers “if things go badly, how long am I actually covered?”

Planned vs. realised spending

Two median lines, one for what you planned to spend and one for what you actually funded after any forced cuts, with the gap shaded between them. Where the lines diverge is where lifestyle cuts begin to bite. A gap that opens late and narrow is mild belt-tightening; one that opens early and wide is a plan under real strain.

What’s breaking your plan

When plans fail, this breaks down why. It replays each failed future with one risk switched off at a time and attributes the failure to the risks that, once removed, would have saved it. The longest bars are therefore your dominant failure causes (outliving the money, a market crash, a care episode, and so on). The method is in the technical reference.

What could you change?

Each card tests one concrete change to your plan (saving more each month, retiring later, shifting your investment mix, insuring against care costs) against the same set of futures, and shows how much it moves your chance of the money lasting (along with your end-of-life wealth and your worst-case years funded). The changes shown are chosen for what matters most in your plan, so the list is personal rather than generic. The method is in the technical reference.

A free run gets one of those cards, computed exactly the same way. Which change takes the slot is chosen by the engine from the ones that apply to your plan, and it rotates with each run, so running the same plan again tests a different change. Saving the run keeps the card it got.

What tax costs you

The headline is what your plan pays in tax across every retired year, added up and shown in today’s money. It is the middle outcome: of every simulated life, half pay more and half pay less. Where the tax system also pays money back to you, the headline says both figures and never nets one off the other.

That last point is the whole shape of this section. Four different kinds of money can end up in a tax calculation, and one of them runs the other way: a refundable credit is the state paying you, not charging you. Adding it to a charge gives a number nobody can name. So Premium splits the bill into three figures that are never summed together:

• Tax you pay. Income tax to your national government, anything your region charges on top, any charge on the savings themselves, and any penalty for taking retirement money out early. Nothing paid back to you is taken off this, and it is never below zero.
• What you get back. Money the tax system pays you rather than takes. Today that means help with the cost of health insurance before you turn 65, in the United States. It is shown against the cost it pays for and never subtracted from the tax above, because your plan is billed the full cost of that cover in your spending.
• Cost of withdrawing. What taking money out of your savings added to the bill, including any help you stop receiving because your income rose. This is the figure your plan can move, and where your savings sit changes it more than any rate does.

The three belong to one simulated household, the one whose lifetime bill sits at the middle rank, which is why the tables and the charts add up to them exactly rather than approximately.

Premium also breaks the bill down by the kind of charge your country levies, and by which government charges it. What appears there depends on the country: several tax savings income on the same scale as everything else, so there is no separate line to show, and the panel says so rather than showing a zero. The country models are documented in tax by country.

Your tax year by year

Opening Where your tax comes from shows the same bill three ways: itemised, over time, and split by why it was owed.

Where the tax system pays you something, the first chart draws two lines, both money and never a negative: what you were charged that year, and what you were paid that year. The years on each add up to the two figures above them.

Beneath it, the range of what a year could cost, across every simulated life, with one life drawn over it as a line. That line is the same household the figures above describe. This chart is net, so where a year’s help is bigger than its tax the range sits below the line at zero.

The bottom of the range needs reading with care. Late in the plan it is mostly lives whose savings have run out: there is nothing left to take money from, so the bill falls. A low figure down there is not good news, and it is the same set of futures as the bottom of the portfolio chart near the top of the page. The line stops at the age that household died, which is why it ends before the range does.

Last, the same life split by why the tax was due. It is drawn as separate lines rather than as stacked bands because a line can legitimately dip below zero: in a few countries an investment account pays back more in credits than it costs you in a given year, and a stack has no honest way to show that.

Moving money into your Roth

For a United States plan with money in a 401(k) or traditional IRA, Premium adds a panel on whether moving some of it into your Roth each year pays off. The move is taxed as income in the year it happens, and in return that money is never taxed again. Whether the trade wins depends on rates you cannot know yet, so the panel shows evidence instead of a recommendation.

The chart shows what the next dollar you convert would cost, in your first retired year, with prices and thresholds at the middle of your futures. The cost counts everything that reacts to your income that year: federal income tax, your state’s income tax, and any help with marketplace health cover that shrinks as income rises. It is not a smooth slope and it does not only go up: it steps at the points marked on the chart, and a cheap stretch can sit above an expensive one. That is why “fill up your tax bracket” can stop at exactly the wrong place: the bracket edge is often not where the real steps are. Each mark is named below the chart: where income tax starts, where Social Security starts being taxed and where that stops getting worse, where investment income moves through its taxed band, where the surtax on investment income switches on and off again, where help with health cover starts shrinking, jumps, changes pace and ends, and plain bracket edges. A step that makes the next dollar cheaper is never called a bracket. Two of the marks are not rates at all but money charged in one go: the health-cover cliff, which forfeits the whole year’s help on a single dollar, and the small jump in your share of the plan’s price at 133 percent of the poverty line. The chart draws those as marked amounts rather than as spikes, so a $200 bump and a$9,000 loss do not look the same. Medicare premium surcharges are not in this line, because they arrive two years after the income that caused them; the plan itself does charge them.

The dashed line is the other half of the question. A dollar you do not convert stays in the account and comes out later, taxed at whatever rate that year’s income puts you in. The line shows that rate for the first year required withdrawals apply, with your pensions, Social Security and the required amount itself coming in, priced the same deterministic way at the middle of your futures and at today’s split between your accounts. Dollars below the line cost less to convert now than they would cost coming out later; dollars above it cost more. It is a reference, not a verdict: it does not know which future you will get, and the comparison below does. Plans where required withdrawals already apply in the first retired year, or never apply within the plan, get no line.

The comparison takes the yearly amount you set on the accounts screen and runs your whole plan four times on the same simulated futures: converting nothing, half your amount, your amount, and half more. Because every level faces identical markets and lifespans, the differences between rows are the conversions doing the work. The last column answers the question that actually matters and that a fixed-date calculator cannot: in what share of your possible futures did converting leave more behind than converting nothing. Futures where the money is needed early, or where a life ends soon after the tax was paid up front, count against it honestly.

What the model charges and what it leaves out: the five-year wait on converted money taken out before the access age is charged; the year’s required minimum comes out before anything converts; conversions raise the income that health-cover help and Medicare surcharges are judged on. It assumes the whole 401(k) or IRA balance has never been taxed (no after-tax contributions sitting inside it). State income tax on the conversion is charged at the state your plan names, in the conversion year and in the later year the dashed line prices alike; a household that moves states between the two is priced as if it had not. The best row in the comparison is the best of the levels tried, not a promise that no other amount does better.

The full picture

Income and expense breakdown

Two stacked charts: where your money comes from (pensions, rental, portfolio withdrawals, lump sums) and where it goes (the eight expense buckets), shown as median real euros per year from retirement onward. They make it easy to see how much you lean on portfolio withdrawals versus guaranteed income, and when costs like healthcare ramp up or discretionary spending tapers off.

On a plan that gave one spending total rather than the separate categories, the spending chart says so: the total was split the way a typical household in your country spends (the same split the whole simulation ran on, see the assumptions), and each part then followed its own path with age. A Premium plan sets each part itself.

One thing to read carefully on the income chart: where your plan says a stream is taxable, it is counted before tax. So the stack adds up to more than you actually get to spend, and the difference is your tax bill, which has its own section below. The eighth expense bucket, “tax on savings”, is not that whole bill: it holds only the charges some countries levy on the savings themselves, such as a wealth tax.

Sustainable spending floor

A single figure that answers the planning question in reverse: the most you could spend each month while still hitting your target success probability. If it sits above your current spending, you have room to spend more; if below, it’s the level that would bring you back to target.

It covers essentials, discretionary and healthcare only, not housing. Your mortgage or rent, upkeep, insurance and property tax sit on top of this figure and are held fixed, because they aren’t costs you can cut by deciding to: a mortgage is contractual and upkeep tracks the house, not your lifestyle. The figure your current spending is compared against covers those same three buckets, so the two are like for like. The method is in the technical reference.