Why Fidelity and this planner may produce different projections
Two retirement planners can start with the same balances, spending, Social Security, and allocation yet produce different future balances or success rates. The visible inputs may match while the calculations underneath them do not.
Written by James Wilson · Reviewed August 3, 2026 · Educational information, not individualized advice
First make sure the labels mean the same thing
Fidelity materials describe confidence-level results in which “Significantly Below Average” is a result that 90% of modeled scenarios met or exceeded, “Below Average” corresponds to 75%, and “Average” corresponds to the middle or 50% result. This planner uses plain-language outcome labels:
| Fidelity-style confidence label | This planner's outcome label | Plain-language meaning |
|---|---|---|
| Significantly Below Average | Very Cautious | About 9 out of 10 modeled futures finished above this result |
| Below Average | Cautious | About 3 out of 4 modeled futures finished above this result |
| Average | Middle | Half finished above and half below |
These are conceptual percentile mappings—not equivalent calculations. The simulations, capital-market assumptions, time periods, asset proxies, cash flows, and tax treatment can still differ.
A common source of extra caution
In this planner, Market Assumption Long-Term Returns sets the return level around which the complete group of simulated futures is centered. Very Cautious, Cautious, and Middle then identify outcomes within that group.
Selecting Lower and then reading Very Cautious applies two layers of caution: a lower long-term return target followed by a low-end result from those already-lower simulations. A Fidelity 90%-confidence result may instead be drawn from simulations centered on Fidelity's standard assumptions. The percentile can match while the result does not.
Why balances and success rates diverge
Dollar basis and reference year
Fidelity commonly offers today's-dollar and future-dollar views. This planner's Inflation-Adjusted Dollars use first-projection-year purchasing power. If the projection starts in a later year, those are not automatically today's dollars. Compare the same view and reference year.
Market data and simulation construction
Fidelity uses its own published methodology, benchmark histories, asset-mix assumptions, and simulation process. This planner's default risk model samples documented multi-year historical sequences, recenters them to the selected long-term return and inflation targets, and runs 1,000 seeded paths. Neither method predicts the future.
Cash-flow timing
This planner models complete calendar years, January 1 starting balances, split half-year growth, and midyear income, spending, taxes, withdrawals, RMDs, and Roth conversions. Different timing conventions can materially change compounding over several decades.
Taxes, Medicare, and withdrawals
This planner explicitly estimates federal and state income taxes, taxable Social Security, RMDs, Medicare, IRMAA, and a defined withdrawal order. Another result may use an estimated tax rate, different account liquidation rules, or different healthcare treatment.
Investment classification
Stock, bond, cash, and “Other” holdings may map to different benchmarks. For risk analysis, this planner proportionally redistributes Other across U.S. stocks, foreign stocks, bonds, and cash. Fidelity's published materials also describe reclassification for some unclassified holdings, but the exact holdings and proxies may still differ.
Fees and expenses
Confirm whether advisory fees, fund expenses, transaction costs, or other investment expenses are included. Small annual differences can compound into large ending-balance differences.
Success definition and age
This planner reports spending coverage through both the Primary Horizon Age and Ending Age. Fidelity generally evaluates the selected planning age in the applicable experience. A 95% result through age 80 is not comparable with a 95% result through age 95.
A comparison checklist
- Use the same starting date, retirement age, horizon age, and ending age.
- Match every account balance, future contribution or rollover, and account type.
- Match first-year spending and whether taxes and healthcare are inside or added to it.
- Use the same Social Security claim age, estimate year, and current- versus claim-year dollar basis.
- Match pension start dates, COLAs, survivor assumptions, and other income.
- Match domestic stocks, foreign stocks, bonds, cash, Other, and investment fees.
- Use the same inflation rate and the same future- or inflation-adjusted display.
- Compare corresponding confidence levels: Very Cautious with 90%, Cautious with 75%, and Middle with 50%.
- Compare success through the same age and read each tool's definition of success.
- Document the date and version of both results because assumptions and tax rules change.
How to use disagreement productively
If one planner reports a much stronger result, test the assumptions most likely to cause it: dollar basis, market percentile, inflation, planning age, fees, taxes, spending timing, and healthcare. Then use this planner's year-by-year table to locate the first year balances begin to separate.
A strategy that works only under one tool's favorable assumptions deserves more stress testing. A strategy that remains workable across different reasonable models may provide a more useful planning margin, but neither result is a guarantee or individualized recommendation.
Published Fidelity references
Related guides
Compare assumptions before comparing answers
Use the planner's Methodology, year-by-year details, and risk outcomes to understand what drives a difference instead of relying on one ending balance.
Open Retirement Income Planner