Open almost any fund factsheet and you’ll find a tidy headline: “average annual return, 9%.” It sounds like a promise — leave $10,000 alone for five years and compounding will do the rest. But run the numbers on what that fund actually did, and the money that comes out the other end is reliably less than 9%-a-year would suggest. The gap isn’t hidden fees or bad timing. It’s arithmetic, and once you see it you can’t unsee it.
The culprit is that there are two different “averages,” and the financial world quietly uses the flattering one. Understanding the difference is one of the highest-leverage things a long-term investor can learn, because it changes how you read every return number you’ll ever be shown.
Two averages, two very different stories
The arithmetic average is the one you learned in school: add up each year’s return and divide by the number of years. The geometric average is the single steady rate that would have grown your money from its real starting value to its real ending value. The first answers “what was a typical year?” The second answers the only question that pays your bills: “how much money do I actually have?”
They diverge the moment returns stop being identical — and real returns are never identical. Consider a fund with five perfectly ordinary years:
| Year | Return | Balance (from $10,000) |
|---|---|---|
| 1 | +20% | $12,000 |
| 2 | −10% | $10,800 |
| 3 | +15% | $12,420 |
| 4 | −5% | $11,799 |
| 5 | +25% | $14,749 |
A five-year sequence with an arithmetic average of exactly 9% per year.
The arithmetic average is clean: (20 − 10 + 15 − 5 + 25) ÷ 5 = 9% a year. So a naïve projection says $10,000 should become $10,000 × 1.09⁵ = $15,386. But the actual ending balance is $14,749 — about $637 short. The rate that genuinely turned $10,000 into $14,749 over five years is roughly 8.1%, not 9%. That 8.1% is the geometric return, and it’s the number you really earned.
The one-line version
Whenever returns vary, the geometric (compounded) return is lower than the arithmetic (simple) average. The simple average describes a typical year; only the geometric average describes your bank balance.
Why losses cost more than they look
The reason sits in a quirk of percentages everyone knows but few connect to their portfolio: a loss needs a bigger gain to undo it. Lose 10% and you need +11.1% to get back to even, because you’re now growing a smaller base. The deeper the hole, the more punishing the climb:
| If you lose… | …you need this gain just to break even |
|---|---|
| −10% | +11.1% |
| −20% | +25.0% |
| −30% | +42.9% |
| −50% | +100.0% |
| −90% | +900.0% |
Recovering a loss takes a disproportionately larger gain.
This asymmetry is why volatility quietly erodes growth. Take the most vivid case: a year of +50% followed by a year of −50%. The arithmetic average is a cheerful 0% — “you broke even on average.” But $10,000 grows to $15,000, then falls by half to $7,500. You didn’t break even; you lost a quarter of your money. The geometric return is about −13.4% a year. Same two numbers, wildly different verdicts, depending on which average you trust.
Volatility drag: putting a number on the gap
The shortfall between the two averages has a name — volatility drag — and a remarkably simple approximation:
Geometric return ≈ Arithmetic return − (Variance ÷ 2)
Variance is just the square of the standard deviation, the standard measure of how spread out returns are. In our five-year example the standard deviation of the annual returns is about 14%, so the drag is roughly 0.14² ÷ 2 ≈ 1.0 percentage point. Sure enough, the arithmetic 9% minus about 1 point lands right on the ~8.1% we computed the hard way. The formula isn’t a coincidence; it falls straight out of how compounding works, and it carries a profound implication:
For a given average return, more volatility means less money. Two funds can advertise the same average annual return, yet the calmer one leaves you measurably richer — purely because its drag is smaller.
That single sentence reframes a lot of investing. It’s the mathematical backbone behind why diversification helps, why “risk-adjusted return” is a real concept rather than jargon, and why chasing the highest-average, wildest-riding fund often disappoints. A useful way to compare investments on exactly this basis is the Sharpe ratio, which scores return per unit of volatility.
Where this quietly costs people money
1. Reading fund marketing
“Average annual return” on a factsheet is often the arithmetic figure, while the “annualized” or “compound” return is the geometric one. They can differ by a percentage point or more for a volatile fund — and over decades, a point of annual return is enormous. If a marketing number looks better than the fund’s own annualized figure, you’ve probably found the arithmetic average dressed up for the occasion.
2. Projecting your own future
If you take a historical average return and plug it into a compound-growth projection, you’ll overstate your future wealth. For any multi-year forecast, use the geometric return — the CAGR calculator converts a start value, end value, and number of years into the honest compounded rate, and the investment return calculator works it out from a full series of cash flows. When you then model growth with the compound interest calculator, feed it the geometric rate, not the average of the good years.
3. Leveraged and “daily” products
Volatility drag turns vicious with leverage. A fund that delivers 3× the market’s daily move does not deliver 3× the market’s return over a year — in choppy markets it can badly trail, or even lose money while the index is flat, because the drag is multiplied along with the returns. Many investors who bought “3× long” products in a sideways market learned this the expensive way.
4. Fees stack on top
Drag is the cost of volatility; fees are a separate, additive leak — and they compound just as relentlessly. A 1% annual fee doesn’t cost you 1% once; it quietly removes a slice of every future year’s compounding. The investment fee impact calculator shows how a seemingly small percentage becomes a six-figure hole over a working lifetime.
How to read returns honestly
- Ask which average it is. If a number isn’t labelled “annualized” or “compound,” assume it’s the flattering arithmetic one and treat it with suspicion.
- Project with the geometric rate. Past average returns belong in a single-year estimate; past compounded returns belong in a multi-year plan.
- Respect volatility. Between two investments with similar average returns, the steadier one usually compounds to more. That’s not timidity — it’s the maths.
- Mind the drag on anything leveraged. The more amplified the ride, the more the gap between “average” and “actual” works against you.
The takeaway
None of this means returns are fake or that markets don’t build wealth — they do, spectacularly, over long horizons. It means the average return is a storyteller, and the geometric return is the accountant. When you want to feel good, look at the average. When you want to know what you’ll actually have, compound the geometric rate — and give volatility the respect it’s quietly earning at your expense.
Want to see it on your own numbers? Drop a start value, an end value, and a time span into the CAGR calculator to find the rate you truly earned, then compare it to the simple average of your yearly returns. The gap you find is volatility drag, in dollars.