BITX: twice the daily move, and the arithmetic nobody puts in the brochure

What BITX is

BITX is a registered investment company under the Investment Company Act of 1940 that seeks 200% of the daily return of bitcoin. It holds CME bitcoin futures through a wholly owned Cayman Islands subsidiary, with Treasury bills as collateral. It listed on Cboe BZX on 27 June 2023 and held $1,210,141,600 on 31 August 2026, making it the largest leveraged crypto ETF in the United States by a comfortable margin.

That is the description. The rest of this page is the arithmetic, because with a leveraged fund the maths is not a footnote to the product — the maths is the product. Almost every complaint ever filed about a fund like this comes down to an investor who read "2x bitcoin" and assumed it meant twice bitcoin's return over whatever period they happened to hold it. It does not, and the gap is not a defect. It is the design working exactly as documented.

The daily reset

BITX targets twice bitcoin's move for one trading day. At the close, the exposure is rebalanced so that the following day starts fresh at 2x of the new, larger or smaller asset base. That reset is what makes the objective achievable day after day, and it is also what severs the link between the daily multiple and any longer-horizon multiple.

Over any stretch longer than a session, your return is the product of the daily levered returns, not twice the cumulative return. Compounding is not a rounding error here; it is the dominant force. Two paths that end at the same bitcoin price but travel differently will hand you materially different results in BITX, and there is no way to arbitrage that away because it is baked into the rebalancing.

Volatility drag, with numbers

Work through the simplest possible example. Bitcoin starts at 100. Day one it rises 10%, to 110. Day two it falls 10%, to 99. Over the two days bitcoin is down 1%.

BITX starts at 100 as well. Day one it gains 20%, to 120. Day two it loses 20%, and 20% of 120 is 24, so it finishes at 96. Bitcoin is down 1%; the fund is down 4%. Twice the unlevered loss would have been 2%. The fund lost double that again. Nothing went wrong, no fee was charged in the example, and the tracking was perfect on both days. The loss came purely from multiplying 1.2 by 0.8 instead of adding the moves together.

Scale that up. The standard approximation for the drag is one half times L times (L minus 1) times variance per unit of time. At L equal to 2 that expression collapses to roughly the variance itself — so the annual drag is approximately bitcoin's annualised volatility squared. Bitcoin's realised volatility routinely sits between 50% and 70%. Square those and you get roughly 25% to 50% a year of structural erosion in a market that ends the year where it started. Then add the 2.75% expense ratio on top.

Path matters

What happens in a trend

Compounding is not uniformly hostile, and it would be dishonest to present it that way. In a sustained one-way move, the daily reset works for you. Each up day increases the base on which the next day's 2x is applied, so a steady climb delivers more than twice the cumulative gain. This is why BITX occasionally posts numbers that look impossible against a 2x label, and why it attracts the assets it does.

The same mechanism runs in reverse on the way down, and this part is genuinely useful to understand. In a steady decline each loss is applied to a progressively smaller base, so the fund loses less than twice the cumulative fall. That is also why compounding alone cannot take the fund to zero — it can only take it asymptotically towards zero. A single-day bitcoin fall of 50% would end the fund outright; short of that, the arithmetic grinds rather than executes.

Grinding is not survivable in practice. Defiance's ETHI, a leveraged long-and-income ether product, closed after last trading on 26 January 2026, having delivered roughly minus 66%. The product did what it said it would do. The holders still lost two thirds of their money, and then lost the fund itself. Leveraged crypto products close, and that risk belongs in your assessment alongside the price risk — see our guide to Bitcoin ETF risks.

The 2.75% question

BITX charges a total expense ratio of 2.75%, of which 1.85% is the management fee. ProShares' BITU targets economically similar 2x daily bitcoin exposure — using swaps and futures referenced to a spot bitcoin index rather than futures alone — and charges 0.98%. The gap is roughly 180 basis points a year for a comparable objective.

And yet BITX is about 2.4 times the size of BITU. We will say plainly what that looks like from the outside: distribution and first-mover advantage, not product merit. BITX launched in June 2023, nine months before BITU, it owned the shelf space, and flows in leveraged products follow familiarity and screen presence far more reliably than they follow cost. A reader choosing between the two should be able to say what the extra 1.77 percentage points buys. If the answer is "nothing I can articulate", that is the answer.

Be careful about the comparison in one respect: the two funds are not identical under the hood. BITX runs futures through a Cayman subsidiary; BITU uses swaps and futures referenced to a spot index. Those differences can produce different tracking against the daily target, particularly during stressed markets, and neither approach is automatically superior. But the difference would have to be worth 180 basis points a year to justify the price, and over the multi-day holding periods leveraged funds are designed for, it is not obvious that it is. Both funds sit side by side on our leveraged Bitcoin ETF comparison, and the wider fee picture across the category is on the fee analysis page.

Why there is no 3x fund

No 3x bitcoin ETF trades in the US as of 2 September 2026. Cboe BZX filed SR-CboeBZX-2026-065 on 10 August 2026 seeking to list Volatility Shares 3x Bitcoin and 3x Ether ETFs, structured as commodity pools under the Securities Act of 1933 rather than as 1940 Act funds. An initial SEC decision is expected around 3 October 2026, and even a favourable one would not put the funds on screen — trading would still depend on the registration statement going effective.

Worth knowing: leveraged and inverse structures are ineligible for the generic listing standards route that opened the door to the wave of single-asset crypto ETFs. They go through the older, slower rule-change process every time. If a 3x product does arrive, apply the drag formula again — the coefficient triples at L equal to 3, which turns an already punishing sideways-market cost into something close to unmanageable. The mirror image of all this, on the short side, is covered on our inverse Bitcoin ETF page.

Who this is actually for

Short-horizon traders with a specific directional view, a defined exit, and the discipline to take the position off. That is the whole audience. BITX is a well-constructed instrument for that job and it is genuinely liquid, which matters when you need out quickly.

It is the wrong instrument for building a bitcoin position, for expressing a long-term view, for anything inside a retirement account you will not monitor, and for any investor who cannot restate the two-day example above from memory. If the goal is simply owning bitcoin exposure, the unlevered funds on the Bitcoin ETF list and the broader crypto ETF directory do that job without the reset, and our picks across the asset classes narrow the field. If you would rather hold the asset than a claim on futures, see fund versus coin, and if you have settled on a fund, how to buy a Bitcoin ETF covers the mechanics.