Over the three months to October 23, 2026, 502 of 658 leveraged and inverse ETFs finished short of their own stated multiple. The median fund landed 4.2 percentage points below what its name promises, against a median distance travelled by the underlying of 5.3% and median underlying volatility of 47%. These funds are built to deliver their multiple over a single day, then reset and do it again the next day. Because daily results compound rather than add, a run of days does not equal the multiple times the period's overall move. The gap between stated and delivered widens with how far the underlying travels during the period, not with where that underlying ends up.
The pattern holds across holding periods. At one month, 309 of 694 funds were short, with a median difference of just −0.2% against 1.6% of underlying movement and 43% volatility. By six months the median difference had widened to −7.3%, and by three years, across a smaller surviving set of 95 funds, the median fund stood 26.2 percentage points short of its stated multiple, with underlying moves averaging 35.6% and volatility down to 18%. The direction is consistent: longer windows and bigger underlying moves bring wider gaps, regardless of whether volatility itself is rising or falling.
A separate cut sorts the three-month results by the underlying's volatility rather than by holding period. Funds tracking underlyings with volatility of 50% to 80% had a median difference of −10.1%, the widest of the four bands, on median volatility of 68%. The quietest band, 0% to 30% volatility, showed a median difference of only −1.8%. The highest-volatility band, 80% and above, came in at −9.6%, slightly better than the 50-80% band despite higher volatility, a reminder that the relationship is not a straight line once moves get large enough.
Where more than one issuer offers the same stated multiple on the same underlying, the outcomes can differ by several percentage points even though the mandate, the days held, and the underlying's move are identical. Among 129 such underlyings, BABA at +2x shows the widest recorded spread: with the underlying down 11.1%, KBAB finished at +11.4% and BABU at −3.8%, a spread of 15.2 percentage points between four issuers offering the same trade. Smaller but still measurable spreads appear elsewhere, including PDD +2x at 6.2 points between two issuers, and BMNR +2x at 5.4 points between three issuers tracking a 41.9% move in the underlying. These comparisons hold the mandate and the period fixed, isolating the effect of issuer and structure alone.
Readers can look up any individual fund on ETFIQ's fund pages to see its stated multiple, its underlying, and its measured difference over the same one-month, three-month, six-month, one-year or three-year windows shown here.