Plan Thresholds is a preview. The numbers come from the same engine as the rest of your plan, so they are as accurate as anything else here, but the page is new and how it frames its questions may still change. Treat a threshold as a guide rather than a figure to plan around on its own.

Every other projection here runs the same direction: you set assumptions, and the planner tells you what happens. Plan Thresholds runs it backwards. You choose the outcome, and it works out the assumption that outcome requires.

Two questions, in both directions:

• Required return - holding your inflation assumption where it is, what average return does this plan need?
• Maximum inflation - holding your return assumption where it is, how much inflation can the plan absorb?

Against four outcomes:

• The plan lasts - the money reaches the end of your plan. Measured in years of margin, so it also tells you how early it would have run out, or how many years of spending you would have finished with.
• Essentials funded and full lifestyle funded - the two funded ratios.
• Legacy goal, if you have set one.

Which outcome goes with which question

This matters more than it sounds, and it is the one thing worth reading before using the page.

A funded ratio barely moves when you change the return you assume. That is not a limitation of this page, it is what a funded ratio is: it compares what you have now, plus income that does not come out of your portfolio, against what you owe. The growth rate of your portfolio appears nowhere in that comparison.

The effect is stark. On one plan, taking the assumed return from four points below its current setting to two points above moved the resources side by nothing at all, while the money went from running out early to finishing with over three hundred thousand dollars. The funded ratio actually drifted slightly down over that range, because higher returns earn more taxable interest and tax counts as an obligation.

So use the plan lasts or legacy goal for a return question, and the funded ratios for an inflation question, which they track strongly and in the right direction. Pairing return with a funded ratio is not blocked, and the page will say plainly that it found no threshold rather than inventing one.

What "return" means here

Your plan holds three return assumptions - stocks, bonds and cash - and an allocation across them. The axis moves all three together by the same amount, so your allocation and the spread between asset classes are left exactly as you set them. The figure quoted is the blended result at your own after-retirement mix.

Shifting only stocks would quietly change the shape of your portfolio as the number moved, and you would be reading two changes as one.

What "inflation" means here

It moves general inflation, medical inflation, and any per-expense rate you have set, all by the same amount, and it moves the Social Security COLA with them, because Social Security is inflation-linked in reality. The next-year COLA that has already been announced does not move, since a scenario should not change a figure that is already published.

Two things deliberately do not move, and both make the answer cautious rather than optimistic. A pension with a fixed COLA keeps that COLA, which is what the contract actually says and is part of what higher inflation does to a plan. And tax brackets are not re-indexed, so a high-inflation scenario applies today's bracket widths to tomorrow's income and charges more tax than reality would. Read a high-inflation threshold as a conservative one.

Why this is a single projection, not the simulation

Finding a threshold means running the plan twenty or so times, narrowing in on the answer. Doing that with the full 2,000-run simulation at every step would take a very long time, so each step is one deterministic projection: the same engine, the same taxes, RMDs and withdrawal ordering, with returns held at their average instead of varied.

So a required return here is what the plan needs on average, not a probability. It is a different question from the success rate, and the two answer different things. A plan can need a 5.8% average return to last and still show a 90% success rate, because the success rate is asking how often a varying market gets you there rather than what the average has to be.

The outlook map

Underneath the answer is the whole plane: every combination of return and inflation your plan might meet, coloured by how it comes out. A threshold is one number on one axis, and the map is the surface that number sits on. Its slope is the useful part, because it shows which of the two assumptions your plan is actually sensitive to, which no single figure can.

Your own assumptions are marked with a ring labelled "You are here". It usually sits between gridlines rather than on one, because the axis is laid out on round numbers and your plan is wherever it is.

The colour scale is centred on the line that matters, so the middle colour always marks exactly break-even: the money lasting precisely to the end of the plan, a ratio of exactly 100%, or a legacy landing exactly on its goal. Green is better than that line, red is worse.

Obligations are discounted at one fixed rate across the whole map, stated underneath it. That is deliberate: the discount rate is derived from your bond assumption, so letting it drift as the return axis moves would change the answer for two reasons at once and the picture would show neither.

Zones

The zone view replaces the shading with four bands, in the order needs, wants, legacy:

• Legacy goal reached - everything below it is covered too.
• Full lifestyle funded, legacy missed.
• Essentials funded, lifestyle short - the interesting one. A plan that works if it has to and disappoints if it does not, which no single number tells apart from either neighbour.
• Essentials underfunded.

A plan with no legacy goal simply never reaches the first band, rather than being given a fictional one.

When there is no single answer

Three cases say so rather than quoting a number:

• Already met everywhere - the plan clears the target across the whole range tested. Reported as headroom instead of a requirement.
• Out of reach - even at the top of the range the target is not met. Returns alone will not close that gap, and a number would imply otherwise.
• No single crossing - the outcome moves both ways across the range, so no one threshold describes it. This is rare, and usually a plan carrying a lot of inflation-linked Social Security alongside a fixed pension.