We use privacy-friendly analytics to learn which calculators help, and nothing loads until you agree. Read our privacy policy.
Investment
The five ways published fund returns mislead, and the comparisons that survive scrutiny.
By FreeCalculators Editorial · Published 2026-09-01 · Updated 2026-09-04 · 4 min read · 917 words
Published fund returns are accurate and routinely misleading, because the choices behind them are not visible on the chart. Which period was chosen, which benchmark it is shown against, whether the funds that closed are included, and whether costs and taxes are netted off all change the picture. Reading a comparison properly means asking what was excluded, not just what the number says.
Survivorship bias is the largest. Funds that performed badly get closed or merged, and they vanish from the category average, which makes the surviving group look better than the original group ever was. Any comparison drawn from funds available today has this built in.
Then there is start-date sensitivity: shifting the beginning of a ten-year window by six months can reverse a ranking. Benchmark mismatch is next, where a fund holding mid-cap stocks is shown against a large-cap index. Fourth, returns are usually quoted before an adviser fee and always before your tax. Fifth, a track record often belongs to a manager who has since left.
| Trap | How it distorts | The check |
|---|---|---|
| Survivorship bias | Failed funds leave the average | Prefer indexes over category averages |
| Start-date sensitivity | Ranking flips with the window | Compare rolling periods, not one window |
| Benchmark mismatch | Style difference reads as skill | Verify the index matches the holdings |
| Costs excluded | Gross returns flatter the fund | Use net-of-fee figures, add your platform cost |
| Manager change | Record belongs to someone else | Check tenure against the period shown |
Compare total return, which includes reinvested dividends, rather than price change. For an income-paying fund the difference is large, and price-only charts systematically understate what a holder actually earned. Then compare against the correct benchmark over multiple rolling periods, so a single lucky start date cannot carry the conclusion.
Adjust for risk where the funds differ in volatility. A fund that beat its benchmark by taking substantially more risk has not demonstrated skill; it has demonstrated leverage to the same market move, and the same exposure would have been available more cheaply.
How the same fund tells two stories (2026)
Fund X, US mid-cap equity
Version shown in the marketing material
10-year annualised, price only, gross 9.8%
Benchmark shown: large-cap index 8.9%
Apparent outperformance +0.9%
Corrected comparison
Total return with dividends, net of 0.95% fee
9.1%
Correct mid-cap benchmark total return 10.4%
Actual underperformance -1.3%
Add your platform fee of 0.25% 8.85%
Underperformance you experienced -1.55%
Nothing in the marketing figure was false.
It used price returns, gross of fees, against
a benchmark the fund does not track.A 2.4-point swing came entirely from correcting the basis of comparison. This is the single most common way a well-intentioned investor buys underperformance.
Use a corrected comparison to eliminate rather than to select. It reliably identifies funds charging active fees for index-like holdings, funds whose record predates the current manager, and funds whose apparent edge disappears once costs are netted off. That elimination is where the value is.
For selection, weight cost and structure. Regulatory disclosure helps here: the Securities and Exchange Commission requires funds to publish a fee table and to show the effect on a hypothetical investment, which makes cost the one comparison that is standardised across every fund you look at.
Comprehensive Guide
Read our investing guide for stocks, bonds, ETFs, and portfolio strategy.
Try the calculatorWas this page helpful?
How this guide was created
This guide was written and reviewed by FreeCalculators Editorial, drawing on published formulas, official government sources, and real calculator outputs from our 4 calculators in this category. Every claim is sourced; every formula is auditable. Read our review policy.