Leveraged ETF Decay Calculator (Volatility Drag)
How volatility erodes a 2× or 3× leveraged (or inverse) ETF over time — expected return versus the naive multiple, break-even index return and the daily-rebalancing path for a chosen index move and volatility.
How it works
Leveraged return ≈ (1 + R)^L × e^(−(L² − L)/2 × σ² × T) × (1 − fees)^T − 1; volatility drag per year ≈ (L² − L)/2 × σ²
A leveraged ETF delivers its multiple every day, then resets, so over longer periods it compounds — and compounding a volatile series costs money. If the index rises 10% then falls 9.1% it is back where it started, but a 3× fund is up 30% then down 27.3%, ending 5% lower. The lognormal approximation shows the damage grows with the square of volatility and with (L² − L): a 3× fund on a 20% volatility index gives up roughly 12% a year to drag before fees, and financing the borrowed exposure costs several percent more at today’s rates. In a strong steady trend the same maths works in your favour, which is why these products can outperform their multiple in a smooth bull run and then be crushed in a sideways market. They are trading tools, not buy-and-hold investments.
Worked example
Index +10% over one year with 20% volatility: a 3× fund is expected to return about 16% after 1.5% of costs — versus the 30% a naive investor expects — with 11% lost to volatility drag; it needs the index to gain roughly 4.6% just to break even.
Frequently asked questions
Why did my 3× ETF lose money when the index was flat?
Volatility drag: daily rebalancing sells after down days and buys after up days, so a choppy flat market bleeds value. At 30% volatility a 3× fund loses about 27% a year just from that.
Are inverse ETFs worse?
Yes — an inverse fund has (L² − L)/2 = 1 for −1× and 3 for −2×, the same drag as a 2× and 3× long fund, plus it fights the market’s long-run upward drift.
Related calculators
Educational estimate, not financial or tax advice. Rules and thresholds change — confirm current figures with the IRS/HMRC or a qualified adviser before acting.