“This fund grew 315% over the last ten years.” Sounds like 31.5% a year — except it isn’t. The same ten-year history can honestly be described two ways, and mixing them up is how people end up comparing apples to oranges. This guide shows what each number means, why they differ, and how to read both without fooling yourself.
Annualized return is the one steady yearly rate that would have produced the same ending value as the actual history — in other words, the compound growth rate. It is not the average of the yearly returns (that “mean” is a different and usually higher number), and it is not the total divided by the number of years.
If you looked up “annualized returns meaning” and want it in one line: it is the compound yearly rate that a headline total return works out to — the number behind “+14.7% a year over a decade”.
How it is calculated: annualized = (1 + total return)1 ÷ years − 1. SPY’s ten complete calendar years to the end of 2025 grew +294.9% in total; (1 + 2.949)1÷10 − 1 = +14.72% a year — the same figure shown as “Annualised (CAGR)” on the fund pages here.
| What it means | Example | |
|---|---|---|
| Total return | Everything your money grew over the whole period | +315% over 10 years |
| Annualized return | The steady yearly rate that produces the same result | ≈ +15% per year |
Both are true for the same investment. $10,000 growing 15% a year for 10 years compounds to about $40,450 — a gain of about 304%, close to that 315% example (exact figures depend on timing). The point: annualized is not “total divided by years.” 315% ÷ 10 = 31.5% would be wildly wrong, because it ignores compounding.
For a one-year window these two words describe the same number. For anything longer they do not, and that is the whole confusion.
An annual return is what one year did, on its own — our fund pages list every calendar year separately (SPY: +31.22% in 2019, −18.17% in 2022). An annualized return is the single steady rate for a multi-year window (+14.72% a year for SPY’s ten years to 2025). No individual year equals it, and it is not the average of them either.
| If you mean… | The number to use | SPY example |
|---|---|---|
| one year, as it happened | that calendar year’s return | 2019 +31.22% · 2022 −18.17% |
| a one-year window | annual = annualized, same figure | the year’s own return |
| several years, per year | annualized (compound) return | 10 years → +14.72% a year |
People also write this as “annualized vs annual”. If you keep one thing: over several years “annualized” is the compound rate while “annual” means the individual years — which is why one table can honestly show a +14.72% annualized figure and a −18.17% annual return side by side.
“Average return” is also called the mean return: add up the yearly returns and divide by the number of years. That number is not what you earned — it ignores compounding, and it treats a −50% year as if a +50% year cancelled it out (it does not). Here are real figures from this site’s own database, ten complete calendar years (2016–2025, dividends reinvested, net of fees):
| Fund | Average of the 10 yearly returns | Annualized (compound) | Gap |
|---|---|---|---|
| SPY | +15.81% | +14.72% | +1.09 pp |
| QQQ | +22.33% | +19.45% | +2.88 pp |
| TLT | +0.41% | −0.54% | +0.96 pp |
The gap is not a rounding artefact — it widens with volatility, which is why it is nearly 3 points for QQQ. And in TLT’s case the two numbers disagree about direction: the average says “+0.41% a year”, while the compounded result for the decade was a loss of about 5.3% in total. If someone quotes a “mean return” as a description of what an investment did, that is the trap.
Because losses hurt more than gains help, the order of good and bad years changes the ending total:
So a single headline number hides whether the ride was a steady climb or a rollercoaster that ended OK. That’s why looking at calendar years side by side — and at the worst drawdown — is worth more than any one average. A 15% annualized with a 35% drawdown and a 15% annualized with an 8% drawdown are not the same investment.
Exactly the same as the singular: the compound yearly rate. Fund tables often write “annualized returns” when they show several windows side by side (1y / 3y / 5y / 10y) — each of those figures is that window’s compound yearly rate, not an average of the years inside it.
The single steady yearly rate that would produce the same ending value as the actual history — the compound growth rate. $10,000 growing 15% a year for 10 years ends near $40,450, so a +315% ten-year total is “about 15% a year”, not 31.5%.
No, and this is the usual mix-up. A mean (average) return adds the yearly returns and divides evenly, without compounding, so it overstates what you actually earned. Over 2016–2025 our own data gives SPY a +15.81% mean against a +14.72% annualized figure, QQQ +22.33% against +19.45%, and TLT a +0.41% mean while the compounded result was negative.
Total return is the full gain over the period (+315% in 10 years); annualized is the steady yearly rate that produces it (≈15%/yr). Use annualized to compare periods of different lengths.
Because returns compound — 31.5% every year would end far above a 315% total. Annualized is the geometric average, not the arithmetic one.
Yes for a one-year window, no for anything longer. An “annual return” normally means what a single year did (SPY: +31.22% in 2019, −18.17% in 2022); an “annualized return” is the single steady compound rate for a multi-year window (SPY’s ten years to 2025: +14.72% a year). One year’s own return can sit far from the annualized figure without either number being wrong.
The return of one year — normally one calendar year, dividends reinvested and the fund’s fees already deducted. Our fund pages list these year by year under “Every calendar year”; the annualized figure shown above that list is the multi-year compound rate, not a yearly return.
A loss cuts deeper than a gain of the same size: −50% then +50% leaves you down 25%. Different orderings of the same yearly returns can end differently — another reason to look at calendar years.
Related: Maximum drawdown explained · Expense ratio explained