Over the three months to October 20, 2026, 504 of 656 leveraged and inverse ETFs finished short of their own stated multiple. The median fund finished 4.5 percentage points below what its name promises. This is the gap between what a fund says it will deliver and what a holder who stayed in the fund actually got over that stretch.
The reason sits in how these funds work. They deliver their stated multiple over a single day, then reset. String days together and the result is not the multiple applied to the period's overall move, because daily gains and losses compound on top of each other. ETFIQ's data across windows shows this gap widening the longer a fund is held: a median difference of -1.0% at one month, -4.5% at three months, -7.5% at six months, -11.6% at one year, and -25.8% at three years. The median distance the underlying travelled over each window also grows alongside it, from 1.5% at one month to 37.1% at three years, which points to the real driver: the gap tracks how far the underlying moves around during the period, not where it ends up.
Volatility matters more than direction. Grouping the three-month numbers by the underlying's volatility shows the median difference from stated growing from -1.8% in the 0% to 30% volatility band to -10.6% in the 80%-and-above band, with the share of funds falling short also rising, from 120 of 176 funds to 82 of 125. The median underlying volatility for the three-month window overall was 48%, and 504 of 656 funds landed short of their stated multiple at that level of churn.
Even funds built on the same underlying with the same stated multiple from different issuers do not land in the same place. The widest spread in the data came from NBIS +2x: four issuers tracked the same underlying move of -21.0%, yet the closest fund, NBIC, finished at -8.7% against stated while the furthest, NEBX, finished at -15.8%, a spread of 7.1 percentage points. Similar gaps appear in BMNR +2x (6.5 points between BNMC and BMNG), NXPI +2x (5.7 points between NXPG and NXPX), and GME +2x (5.7 points between GMEC and GMEU). These are funds with identical mandates, held over identical days, against the identical underlying move, and the outcomes still diverge by several points depending on the issuer running the fund.
None of this is a defect limited to a handful of products. Across 94 funds held for three years, the median difference from stated reached -25.8%, against a median underlying distance travelled of 37.1%. The pattern is consistent: the stated multiple describes a single trading day, and the longer the holding period or the more volatile the underlying, the further a fund's actual result tends to sit from that number.
Readers can look up any individual leveraged or inverse fund on ETFIQ's fund pages to see its own stated multiple, its underlying, and the measured difference over the window that matches their own holding period.