Two investors. A earns a steady +8% every year. B alternates +40% and −20% — an average of +10% a year, better than A on paper. After ten years, A has more money. Roughly 20% more.
The trap: arithmetic vs geometric
The "average return" you see in marketing is usually the arithmetic mean: add the yearly returns, divide by the number of years. But money doesn't add — it compounds. One +40% year followed by one −20% year is 1.40 × 0.80 = 1.12, not 1.20. The 20% loss eats part of the 40% gain, because after the loss there's less capital left to grow.
Volatility drag
This gap between the arithmetic average and what you actually end up with is called volatility drag. The wilder the swings, the bigger the drag — a mediocre-but-steady strategy can beat a spectacular-but-volatile one on real, compounded money. It's one more reason why the maximum drawdown matters so much.
CAGR: the number that counts
CAGR (compound annual growth rate) answers the only question that counts: at what constant yearly rate would my capital have actually grown from start to finish? It bakes the losses in. That's why every track record on this site reports CAGR — never a flattering arithmetic average — alongside volatility, Sharpe and drawdown, on ~5 years of point-in-time data. The methodology is public: how we compute the numbers.
How to spot the trick in the wild
- "Average yearly gain" without the word compound → ask for CAGR.
- Returns quoted without drawdowns or costs → half the story.
- Short windows starting right after a crash → cherry-picked base.
Educational material — not investment advice.