Simplifications

Every forecasting tool simplifies. Most do not tell you where, which makes it hard to know how much weight a number deserves.

This page lists the shortcuts that can move your result, sorted by which direction they bend it. Nothing here is hidden elsewhere in the product; several are already on the list to fix. What follows is not an apology, it is the information you need to read your own forecast properly.

Shortcuts that flatter your plan

These make a projection look better than reality probably is.

A Roth’s starting balance is treated as entirely contributions. There is no way to know how much of an existing Roth balance is contributions and how much is growth, so the model assumes the whole opening balance is contributions. Since Roth contributions come out tax free and penalty free at any age, this makes early Roth withdrawals cleaner than they might really be. It matters most for early retirees drawing on a Roth before 59½.

The first year’s inflation does not compound into the “today’s dollars” figures. The result is that today’s-dollar amounts across the whole projection run roughly 1% to 3% generous. Nominal figures are unaffected.

All healthcare spending is assumed HSA-qualified. The categories very nearly coincide, but not entirely. Medigap premiums are never qualified, and health insurance premiums before 65 are not either, apart from COBRA and coverage held while on unemployment. If your healthcare line is mostly pre-65 insurance premiums, the model spends your HSA a little too freely.

All education spending is assumed 529-qualified. Tuition, fees, books, and room and board (including off-campus housing and meals, up to the school’s published allowance) are qualified. Travel, a car, or ordinary living should be their own expense rows. If those sit on the education line, the model spends the 529 a little too freely.

Investment income is credited before withdrawals are taken. A dollar you withdraw in March still earns a full year of dividends in the model. The effect is small, but it points the same way in every drawdown year.

No fees. No expense ratios and no advisory fees. If you pay 1% to an advisor, that is roughly 1% a year the projection never charges you.

Healthcare costs rise with general inflation. Medical costs have historically risen faster than CPI. A long retirement with significant healthcare spending is likely to be more expensive than shown.

Shortcuts that penalize your plan

These make a projection look worse than reality probably is.

No employer 401(k) match. For anyone still working with a matching employer, this is the largest single omission on the page. A 4% match on a $150,000 salary is $6,000 a year of contributions the model never makes.

No Roth conversions or conversion ladders. An early retiree who plans to convert traditional balances to Roth during low-income years cannot express that here, and the model will show them paying more tax and hitting penalties they would have avoided.

No 72(t) / SEPP payments and none of the early withdrawal exceptions. Another way early retirement looks harder here than it can be.

No itemizing, and almost no tax credits. The standard deduction is always applied. The personal exemption credit is in the tables for Arkansas, California, Nebraska, and Oregon. Other credits are not. If you itemize substantially, your real tax bill is lower than shown.

No HSA receipt reimbursement. Paying medical bills in cash, saving the receipts, and reimbursing yourself years later is what lets a disciplined saver treat an HSA as a stealth Roth. Not modeled, so that saver does better in life than here.

No contributions in a year the plan is drawing down. A household selling assets to cover spending makes no retirement contributions in the model, even though deferring while drawing down is a real strategy.

State pension and retirement income exclusions outside the eight modeled states. Illinois, Iowa, Mississippi, Pennsylvania, Alabama, Hawaii, Georgia, and New York are in the tax tables. A Kentucky, South Carolina, or New Jersey retiree is still shown too much state tax. New York’s $20,000 cap is applied to every pension, including public pensions that are fully exempt.

Shortcuts that are simply approximations

These are not biased in an obvious direction, but they are worth knowing.

The mix can sit a long way from its target for years. Under the default strategy a taxable account with a large unrealized gain is not sold just to rebalance, so a gap that could only be closed by realizing a gain is carried forward instead. That is the tax budget working rather than a missed trade. A 529 or an HSA is never steered toward the retirement mix at all, because it is earmarked for something else. Which way this bends your number depends on which way the drift ran. How rebalancing works has the comparison.

Stocks and bonds are assumed to have gained at the same rate before you entered them. You give a taxable account one cost basis, and it is split across that account’s holdings in proportion to what each is worth today. In reality the stocks you have held for years usually carry the larger share of that gain, so selling them realizes slightly more tax than the model charges. It only affects the gain built up before your plan starts; everything after is tracked per holding.

Each year is one lump. Nothing compounds inside a year, no event knows which month it falls in, and tax is a single annual payment rather than withholding. Over a thirty-year horizon this is a good trade; for a single specific year it is coarse.

The first year is partial. A plan starts from today’s balances and only counts the remaining whole months. Tax for that stretch uses full-year brackets (remaining income as this year’s pace) but only takes the remaining share of the bill from cash, because withholding on earlier paychecks is already out of the balance you entered. See How taxes are estimated.

Losses are never harvested. A crash year produces no offsetting tax benefit, so a bad market costs you a little more here than it would in life.

Retirement account balances are assumed reachable. No plan loans, no vesting schedules, no employer restrictions.

How to use this page

Two practical suggestions.

Find your own exposure. Most of these will not apply to you. A retiree with no employer match, no cash-heavy accounts, and no 529 can ignore most of the page. Someone five years from retiring with a matching employer and a large savings balance is affected by two entries pulling in opposite directions.

Trust comparisons more than absolutes. Almost every shortcut here applies equally to two plans you are comparing, so the difference between them is far more reliable than either number on its own. That is the single most useful habit when reading anything this tool produces.

If you think something on this page is wrong, or you have found a shortcut we have not listed, please write to feedback@ForecastYourMoney.com. It all gets read.