The Parameters tab controls the assumptions that shape every cycle: how many years the simulation runs, when retirement starts, how returns and inflation are sampled, the tax jurisdiction, and the withdrawal strategy that decides where money is pulled from each year.

Core fields

• Final Year. The last calendar year the plan runs through. The simulation starts this year, so picking 2055 in 2026 runs 30 years. Each option shows everyone's age in that year.
• Retirement Year. Anchor for spending start and retirement-linked timing.
• Inflation Settings. Constant Inflation, Historic Inflation, or Monte Carlo Inflation (a Pro feature: randomized paths fit to historical CPI; works with Historic or Monte Carlo returns). Optional: Offset by a percentage. In Monte Carlo mode the Monte Carlo page manages the same setting (Monte Carlo or Constant), and the Monte Carlo CPI fit accepts an optional percentage or offset modifier configured there.
• Returns Settings. Historic Returns (with an optional data year range that limits which historical years cycles may draw from), Constant Returns (equity / dividend / bond / cash rates), or Monte Carlo (a Pro feature: randomized paths for the four built-in classes, with the controls on the Monte Carlo page rather than in this modal). With Constant Returns and Constant Inflation every cycle would be identical, so the plan runs as a single cycle and the results page shows no cycle picker; with Historic Inflation the cycles differ only by the inflation path each start year draws, and the picker says so.
• Tax Jurisdiction. Controls which tax country and account types are available. United States, Canada, United Kingdom, Spain and the Netherlands. The Netherlands is experimental and is currently offered to a small group of testers, so it may not appear in your picker.

Withdrawal strategy

The Withdrawal Strategy section decides where the simulation pulls money from each year to cover spending and taxes. One Strategy selector picks between four options, and only the selected one's editor appears below it:

• Standard order (the default). Your phases run every year, unchanged.
• Bucket Strategy. Three time-horizon buckets, managed for you.
• Sequence-risk defense. Spend your safe accounts in a bear market.
• Rebalance via withdrawal. Spend whichever class is overweight.

Bucket Strategy and the two sourcing strategies both decide where spending comes from, so only one can run at a time. Picking one clears the other in the same save - there is nothing to reconcile by hand.

Bucket Strategy

The Bucket Strategy divides your accounts into three time-horizon buckets, each with its own role in funding spending and weathering market downturns:

• Bucket 1 (Cash/Safety). 1–3 years of spending. All withdrawals come from here first.
• Bucket 2 (Income). 3–10 years. Bonds and moderate-growth holdings.
• Bucket 3 (Growth). Long-term equity growth. Refills the other buckets over time.

Dividends from Buckets 2 and 3 are automatically swept into Bucket 1. In bear markets, the strategy stops selling equities and lets Bucket 1 draw down. Protecting the growth bucket until markets recover.

Refills and dividend sweeps only deposit into taxable accounts (checking, savings, brokerage). Money cannot be moved into a retirement or education account from outside, so an IRA, 401k, HSA, 529, or inherited account placed in Bucket 1 or 2 is skipped as a deposit destination; the editor notes which accounts this applies to. Those accounts still count toward their bucket's balance and can still fund spending and refills. The editor also warns when Bucket 1 holds pre-tax money and retirement starts before age 60, since those withdrawals pay a 10% early-withdrawal penalty on top of income tax.

To place accounts, click Accept All Suggestions to use the suggested placement, or Manually Select to start with every account in Bucket 1. From there, use the left/right arrows on each account card to move it between buckets.

Every account has to be placed, and each of the three buckets has to hold at least one account. Until both are true the Run button stays disabled, the editor names what is missing, and the same reason appears as a row in the setup drawer's Needs attention section — so you can see it from any step, not only this one. If your plan has fewer than three accounts, add one on the Accounts tab or use Standard order instead.

The phases (Standard order and both sourcing strategies)

The phase editor lets you dictate the exact order accounts are drawn down. Use the left/right arrows on each account card to move it between Phase 1, Phase 2, and so on. Phase 1 accounts are drained to $0 before the simulation moves to Phase 2, and year-end rebalancing respects that: it never refills an account you placed in an early phase. Any cash your target allocation calls for is held in your investment accounts instead. While a phase is active, the rebalancer tries to keep contributions inside that phase unless it has no choice. Early-withdrawal-penalty avoidance is always prioritized within a phase.

Sourcing strategies

A sourcing strategy sits on top of your phases. It watches your portfolio each year and, when its rule matches, changes where that year's spending comes from. Your phases stay exactly as you set them and become the default order that runs in every year the rule does not fire. Pick Standard order to keep today's behaviour.

• Sequence-risk defense. When the stock market falls far enough from its peak, that year's spending is drawn from cash and bonds. The bear-market threshold (default 20%) is how far the market must fall before it engages. Equities are sold only as a last resort, once every other source runs out. Two settings shape it: which accounts it may draw from, and which classes it spends in what order. See below.
• Rebalance via withdrawal. When an asset class drifts above its target by more than the drift threshold (default 2 percentage points), that year's spending is sourced from it. Your spending does the rebalancing, so the year-end trade pass has less to do and realizes fewer capital gains.

A strategy never strands your spending. If a rule can only reach part of what you need, the rest is drawn from your default order, and the year detail says so. Open a year in Proof View and look for Which accounts, in what order. It names the rule that fired, the numbers it looked at, and the order it actually used, including the years where nothing fired.

Rebalancing is held back too. Deciding not to sell equities for spending is only half the job: year-end rebalancing would otherwise sell them anyway to bring you back to your target allocation. So while the bear rule is active, rebalancing is stopped from selling the protected classes, and that hold eases off as equities climb back toward their previous peak. Rebalancing is still free to buy a depressed class, which is the point. The year detail shows a "Year-end rebalancing" line whenever the hold applied.

Two settings, kept separate on purpose. "Sequence-risk defense" asks two questions, and they are different kinds of question.

• Accounts - all accounts, or only the ones I pick. This is a tax decision. Cash inside a Traditional IRA is not the same money as cash in checking, so narrowing the eligible accounts is how you keep a downturn out of the accounts you do not want touched.
• Spend in downturns - cash only, or cash, then bonds. This is a market decision: which classes get sold while equities are down, and in what order.

They stay separate because merging them into one selector would make you give up one to express the other. Inside each step your withdrawal order below still decides which account goes first, which is what keeps the tax ordering yours - cash in checking is spent before cash inside an IRA. This also differs from the Bucket Strategy: buckets say "put your money in these containers and the tool manages the containers", this says "leave your accounts where they are and tell the tool which classes to spend first when markets are down".

Where the other sleeves go. With "cash, then bonds", anything else those accounts hold - gold, a custom sleeve - is reached through your normal withdrawal order after the named classes, not proportionally alongside them. This is a change from the older behavior, which drew from everything a safe account held except equities, in proportion. Cash then bonds is more predictable and is what a plan written in tiers means; if you hold a sizable non-equity sleeve outside cash and bonds, expect it to be spent later than it used to be.

One number per class, and two things to do with it. Each class you spend takes a level in years of spending, and either mode reads that same number.

• Spend down to N stops the draw when the class is down to N years of spending and moves on to the next source.
• Maintain at N does that too, and also restores the asset back to N. The restore happens as part of your ordinary year-end rebalancing - it is a dollar floor on that class's target, not a transfer between accounts - and it never funds itself by selling a class the rule is protecting. In a bear market the floor may go partly unfunded and ease back in through the recovery.

Leave a level blank to spend that class all the way down; with no number, neither mode has anything to act on, so both are unavailable until you enter one. A level is not a hard stop on your plan: if honoring it would leave the year unfunded, the year is funded anyway, because a strategy never strands your spending. In the years a level holds dollars back, the year detail shows a "Cushion held" line.

Why "maintain at" is not just rebalancing. It nearly is, and that is worth being straight about: once you rebalance every year, which asset a withdrawal physically came from barely matters, because the year-end trade pass re-trues the mix anyway. The part a percentage target cannot express is the denominator. A 30% bond target shrinks in dollars exactly when the portfolio falls; eight years of spending held in bonds does not. That is why the level is measured in years of spending rather than in percent, and it is the one piece of this that a plain target allocation cannot say.

An honest note on all of this: the 2026 backtest measured the sequence-risk edge as real but modest, roughly one to four percentage points of success rate depending on how much you hold in safe assets, and stagflation windows were never rescued by it. The classes you spend and the levels you set buy fidelity to the plan you actually intend to run; they are not a success-rate lever.

"Rebalance via withdrawal" reads the drift your returns opened up over the past year, measured before rebalancing corrected it. At the default 2 percentage points it waits for roughly two years of divergence before acting; lower the threshold if you want it to engage every year. It needs a target allocation to compare against, so it does nothing on a plan with no rebalancing target.

How this works with the rest of your plan. The features above interact with your phases, your target allocation, and any bond ladder. All of it in one place:

• What each preset does to your phases. Sequence-risk defense: in a bear market year the order becomes class-first, and your phases decide which account goes first within each class. With Checking in Phase 1 and a Brokerage in Phase 2, a bear year spends Checking's cash, then the Brokerage's cash, then the Brokerage's bonds, then bonds anywhere else - before falling back to your phases for whatever is left. Rebalance via withdrawal: your phases run in most years; when a class drifts above target, that year jumps it ahead of them.
• Your phases outrank the cash sweep. Year-end rebalancing never refills an account you placed in an early phase - "drain this, then move on" means what it says. The cash your allocation calls for is held in your investment accounts' cash sleeves instead, so the allocation is still honored; only its location follows your phases. (Before this change, a Phase 1 checking account paying $100,000 of spending was refilled $75,000 by rebalancing the same year, every year, and a 2-year buffer took a decade to drain.) One consequence worth knowing: a "Maintain at" level on cash keeps the cash class at that level, wherever it can legally sit - if the cash account itself is in an early phase, the account still drains while the level is held in your investment accounts. If you want a specific account to stay topped up, put it in your last phase.
• A checking or savings account in an early phase is a spending buffer, and it sits outside your target allocation entirely. Your target percentages describe the money being managed. A cash account you placed early is fuel you have already decided to spend, so counting it would make the target defend a cash level you did not choose: the engine would rebuild that cash somewhere else as fast as you spent it - selling appreciated shares and paying capital gains tax to move cash between two of your own accounts. Instead the buffer is left alone, your percentages apply to everything else, and the buffer appears in your allocation chart as its own Spending Buffer slice that shrinks year over year as it drains. Two things follow from it: your automatically-generated glide path is derived from your other accounts only, so it no longer echoes the buffer's starting balance; and allocation-drift warnings for a phased plan go quiet, because drift is finally measured against the money the target actually governs.
• This applies to cash accounts only, not to every early phase. Ordering Brokerage before IRA before Roth is a tax decision about one managed portfolio, and those accounts stay in your allocation and keep being rebalanced for the years they take to drain - freezing them would let them drift with the market for a decade. Only Checking and Savings are treated as buffers, because placing one early has no tax purpose; the only reason to do it is that the money is earmarked for spending.
• "Maintain at" outranks your allocation percentage. Whenever the dollar level and the percentage disagree, the level wins - see "Why maintain at is not just rebalancing" above for why that is the point rather than a bug. The level is measured against the money being managed, so a spending buffer does not count toward it: two years of cash in a Phase 1 checking account plus a "Maintain at 2 years" cash level means two years of buffer to spend and two years held in your investment accounts.
• Bond ladders count toward a bond level. A "Maintain at" level on bonds is reduced by the value of your ladder rungs, so a ladder plus a level does not buy a second copy of the bonds your ladder already holds.

529 and HSA accounts can never be a sourcing target. Those balances only fund their qualified education and healthcare expenses.