Volatility Drag: Why an Average Return Lies to You
A fund that gains 50% one year and loses 50% the next has an arithmetic average return of exactly zero, yet an investor who held it the whole time is down real money. Volatility drag is the mathematical gap between the average of a set of returns and the actual compounded result you experience, and it is the reason quoted "average" returns routinely overstate what an investor actually earned.
The core principle
The mathematics behind percentage gains and losses is asymmetric in a way that is easy to underestimate. A 50% loss requires a subsequent 100% gain just to return to breakeven, not a matching 50% gain, because the second calculation is applied to a smaller base. Volatility drag is the name for the systematic consequence of this asymmetry: when returns swing widely, the compounded, or geometric, result falls below the simple arithmetic average of those same returns, and the gap grows larger as volatility increases.
There is a widely used approximation that captures the relationship directly: geometric return ≈ arithmetic mean − (volatility2 / 2). The formula shows that the drag scales with the square of volatility, not volatility itself, which is why doubling an investment's volatility can roughly quadruple the resulting drag on compounded returns, holding the arithmetic average constant.
This is not an exotic or rare phenomenon confined to unusual products. It applies to every volatile investment to some degree, which is exactly why a fund's advertised "average annual return," if it is calculated as a simple arithmetic mean rather than a compounded (geometric) figure, can paint a more flattering picture than what an actual, buy-and-hold investor experienced.
The distinction between arithmetic and geometric averaging is not a minor technicality buried in a footnote; it is the difference between two genuinely different questions. The arithmetic mean answers "what was the average of each individual year's return," treating every year as an independent, equally weighted data point. The geometric mean answers "what single constant annual rate would have produced the same actual ending balance," which is almost always the number an investor actually cares about. Regulatory disclosure rules in most jurisdictions require funds to report standardized, compounded total return figures precisely because the arithmetic version is so easy to present in a misleading light.
How the math works
Example 1: the classic 50/50 case. An investment gains 50% in year one and loses 50% in year two. The arithmetic average is (50% + (−50%)) / 2 = 0%, suggesting a breakeven result. The actual compounded path tells a different story: $100 grows to $100 x 1.50 = $150 after year one, then falls to $150 x 0.50 = $75 after year two, a real loss of 25% on the original $100. The volatility drag here, the gap between the 0% arithmetic average and the actual −25% compounded result, is 25 percentage points, entirely attributable to the size of the swing rather than any change in the underlying business or asset.
Example 2: leveraged ETF decay over two volatile days. Suppose an index moves +10% on day one and −10% on day two. The index itself ends at 1.10 x 0.90 = 0.99, a modest 1% loss, close to what the roughly 0% arithmetic average would suggest. A 2x daily-leveraged fund tracking that same index moves +20% and then −20% on the same two days: 1.20 x 0.80 = 0.96, a 4% loss, not the 2% one might expect from simply doubling the index's 1% loss. The leveraged fund's loss is four times the underlying index's loss, not two times, because daily rebalancing compounds volatility drag on top of the leverage itself, a mechanical effect that has nothing to do with a wrong directional bet and everything to do with the structure of daily-reset leverage.
How it shows up in real portfolios
The most direct real-world exposure to volatility drag comes through leveraged and inverse exchange-traded funds, which are explicitly designed to deliver a multiple of a daily return, not a multiple of a long-term return. An investor who buys a 3x leveraged fund expecting three times the index's annual gain is very often surprised, because volatility drag compounds daily and can leave a leveraged fund flat or even negative over a period where the underlying index rose modestly, simply because the index chopped up and down along the way rather than rising in a straight line.
Volatility drag also has quieter, less dramatic implications for an ordinary diversified portfolio. Two portfolios can share an identical arithmetic average return over a decade while producing meaningfully different ending balances, purely because one experienced smoother, lower-volatility returns along the way. This is one of the standard arguments for diversification: combining assets that do not move in lockstep lowers overall portfolio volatility, which in turn narrows the gap between the average return quoted on a marketing sheet and the compounded return an investor actually receives.
For a professional building a long-term retirement portfolio, the practical implication is straightforward: a smoother path to a given average return compounds to a higher ending balance than a choppier path to the same average, which is one more reason a diversified, moderately volatile portfolio can outperform a concentrated, wildly swinging one, even when both are marketed with a similar historical average return.
Concentrated single-stock positions, whether from an employer stock plan or an outsized conviction bet, are where volatility drag tends to inflict the most real-world damage, precisely because individual stocks carry considerably higher volatility than a diversified index. Two employees can hold employer stock with an identical five-year arithmetic average return, and the one whose stock swung more wildly along the way, even without any difference in the company's actual business performance, ends up with a measurably smaller compounded gain purely as a function of that extra volatility, a cost that is easy to overlook when comparing headline average-return figures on paper.
Actionable breakdown
- Before trusting a quoted return, check:
- Whether it is labeled arithmetic or compounded (geometric).
- How volatile the underlying returns were year to year.
- Whether a leveraged or inverse product is involved.
- Whether the quoted period includes an unusually calm or wild stretch.
- Watch for these red flags:
- Marketing material that emphasizes "average" without specifying which kind.
- Leveraged ETFs marketed as long-term buy-and-hold vehicles.
- A fund with high year-to-year swings but an unremarkable long-term chart.
- Ignoring how much a large loss requires an outsized gain just to recover.
- Favor diversification to reduce swings and narrow the drag.
- Treat leveraged and inverse funds as short-term trading tools only.
- Judge any strategy by its compounded result, not its arithmetic average.
The same mathematics that makes volatility drag a headwind for a volatile investment makes rebalancing a genuine, if modest, source of extra return in a multi-asset portfolio, sometimes called a rebalancing bonus or diversification return. Periodically selling a portion of whatever asset class has recently risen and buying more of whatever has recently lagged, back to a target allocation, systematically captures some of the choppiness between asset classes rather than simply enduring it, converting a small piece of what would otherwise be pure volatility drag into realized, incremental return over time.
Common pitfalls
- Trusting a fund's advertised "average annual return" without checking whether it reflects a real, compounded, buy-and-hold result.
- Holding leveraged ETFs as a long-term position, where daily rebalancing and volatility drag can erode returns even when the underlying index is flat or modestly higher.
- Underestimating how much a single large loss requires a disproportionately larger gain just to fully recover.
- Assuming two investments with the same average return produced the same ending balance, when the smoother path almost always compounds to more.
Related concepts
For the underlying measure that drives this effect, see volatility and VIX. For the broader compounding mechanism that volatility interacts with, see compound interest. For how diversification reduces the swings that cause this drag in the first place, see the guide on understanding risk.
It is also worth noting that volatility drag is a mathematical certainty, not a probabilistic tendency that occasionally fails to show up. Any two return sequences with the same arithmetic mean but different volatility will always produce a lower geometric result for the more volatile sequence, without exception, which is precisely what makes the underlying formula a reliable planning tool rather than a rough historical pattern that might not repeat.
The bottom line
Because losses and gains are mathematically asymmetric, a smoother path to the same average return always compounds to a larger ending balance than a wildly volatile one.