Gap risk in leveraged ETFs: the loss your stop never sees

Most traders treat a stop loss as a hard floor. Set it 8% below entry and 8% is the worst case. That model works while the market is open. It fails between the close and the next open, and a leveraged ETF multiplies the failure before any rule, human or automated, gets a chance to act.

The market is closed more than it is open

The regular US session runs six and a half hours. For the other seventeen and a half, plus weekends and holidays, price keeps forming anyway. Futures trade, overseas markets trade, companies report after the close, macro numbers print before the open. Your stop order participates in none of it.

A gap is the distance between yesterday's close and today's open. When the open lands below your stop level, the order does not fill at your level. It triggers at the open and becomes a market order into whatever liquidity shows up. You asked for 8%. The market hands you the opening price instead, and the open does not negotiate.

What 3x does to a gap

Leveraged ETFs reset daily. Each fund aims to deliver a multiple of its index's return for that one day, and the overnight gap counts toward that day. So if the index gaps down 4% at the open, a 3x fund opens somewhere around 12% down, give or take financing costs and tracking noise. An 8% stop triggers immediately and fills near minus 12, possibly worse if the open is thin. The extra four points were never a decision. Nothing in your system, and nothing in you, could have acted, because nothing can act while the market is closed.

This is a different animal from the slow bleed people usually mean by leveraged ETF decay. Decay is a path cost that accumulates across choppy weeks. A gap is a single print. Both fall out of the same daily-reset design, and the SOXL vs SOXS pair is the cleanest case study in how that design behaves in both directions.

Calendars are risk inputs, not trivia

Gaps are not evenly distributed through the year. They cluster around dates anyone can look up in advance. Earnings dates are published weeks ahead. The macro calendar is public for months ahead: CPI prints, Fed decisions, payrolls. A large share of the truly violent overnight moves land on exactly these squares of the calendar.

That makes the calendar a risk input rather than trivia. Holding a leveraged position into the underlying's earnings is not a stop-loss problem, it is a selection problem: you have chosen to accept an event-sized move at 3x with no exit available until it is over. Single-stock leveraged ETFs concentrate this to its sharpest form, since one company's earnings gap is routinely larger than any index gap. The NVDL vs NVDA comparison walks through what that leash length means in practice.

Score the name before entry, don't trust the stop after

Once you accept that a stop cannot protect you from a gap, the logic inverts. The protection has to happen before entry, at selection and sizing time, when you still hold every option. Some systems formalize this by scoring gap exposure per name before a position exists, from inputs like:

• Proximity to the name's next earnings date
• Scheduled macro events during the expected hold
• How the name has historically behaved overnight
• How extended the trend already is
• Whether the vehicle itself is leveraged, and by how much

If the score is bad, the answer is not a tighter stop. The answer is a smaller position, an unleveraged vehicle, or no position at all. This is how Coil (coil.trade) is built: it is long-only, it will only touch a leveraged ETF on the long side, at reduced size, on a high-conviction name, and when nothing qualifies it holds cash rather than forcing a trade. The research behind the ranking is laid out at /how-it-works. None of that makes leverage safe. Leveraged funds can still lose money quickly, including badly. It just moves the risk decision to the one moment you actually control.

Where stops still earn their keep

None of this means stops are useless. During the session, a stop does what it says, and a stop placed under a real support level, rather than at an arbitrary percentage, tends to be a more honest line in the sand. That approach is covered in structural stops explained. The point is ordering: selection first, sizing second, the stop last. A stop is the final backstop for trades you chose to take, not a license to take trades you should have skipped.

Leverage does not change that ordering, it just raises the price of getting it backwards. A 3x fund turns a survivable index gap into a serious one, and no order type sold by any broker changes what the opening print will be.

FAQ

Do stop losses work on leveraged ETFs?

During market hours, yes, they work the same as on any listed security. Across an overnight gap, no. If the open lands beyond your stop level, the order triggers at the open and fills at or near the opening price, which on a 3x fund can be roughly three times the index gap away from your intended exit.

How can I reduce overnight gap risk in leveraged ETFs?

Reduce position size, avoid holding leveraged products through earnings and major macro dates, prefer unleveraged vehicles on names with elevated gap risk, and keep holding periods short. The common thread is making the decision before entry, since no exit rule can act while the market is closed.

Is gap risk the same as leveraged ETF decay?

No. Decay is a slow path-dependent cost that builds up across many choppy sessions. Gap risk is a one-shot loss delivered at a single open. Both come from the daily-reset design of leveraged ETFs, but they hurt on different timescales and call for different defenses.