A stop-loss (stop-market) order becomes a market order once the stop price is hit, so it gets you out but not at your price; a stop-limit order becomes a limit order, so it protects your price but may never fill. On a quiet day the difference is pennies. On a gap it is the whole trade. When $META opened at $526.00 on July 30, 2026, after a $585.61 close on July 29, a stop set 5% under that close was skipped entirely. One order type sold near the open. The other sat there and did nothing.
The definitions, straight from the SEC and FINRA
The SEC's Office of Investor Education and Assistance spells it out in its bulletin on stop, stop-limit and trailing stop orders. A stop order, "also referred to as a stop-loss order," triggers once the stock reaches the stop price, and at that point "a stop order becomes a market order." The bulletin adds the line every trader should tape to the monitor: "The stop price is not the guaranteed execution price for a stop order." The fill "can deviate significantly from the stop price."
A stop-limit order combines a stop and a limit. Once the stop price is reached it "becomes a limit order that will be executed at a specified price (or better)." The catch, again in the SEC's words: it "may not be executed if the stock's price moves away from the specified limit price." The stop and limit don't have to match. The SEC's own example pairs a $3.00 stop with a $2.50 limit.
FINRA's investor insight on stop orders makes the same point with a $50 floor: in a fast market "you could receive less—and potentially significantly less—than $50 per share." One more detail: the SEC says some firms trigger stops on last-sale prices and others on quotes. Ask yours which one it uses.
Regular hours, extended hours, and where a gap happens
A gap is the space between one close and the next open. The SEC's extended-hours bulletin defines regular trading hours as 9:30 a.m. to 4:00 p.m. Eastern Time. It warns that companies may release news outside those hours, and that extended-hours prices may not match the next regular open.
Three official points decide whether a stop can react before the bell:
- Day orders expire. The SEC's order-types bulletin says orders are "Day" orders unless you specify otherwise. They "generally cancel at the end of regular trading hours" and "will not carry over to the extended hours session or the next trading day."
- Extended hours often means limits only. The extended-hours bulletin says "many brokerage firms currently accept only limit orders" in those sessions.
- Carry-over is a firm policy. The SEC tells investors to check whether orders placed during regular hours carry over to extended hours. Good-'til-canceled (GTC) time limits also "may vary based on the brokerage firm."
The mechanical upshot: if your firm does not work stops in extended hours, the regular open is the first print your stop can see. A gap means that print may already be far below your stop.
Every gap below came the session after the company furnished quarterly results under 8-K Item 2.02. Reddit's July 30 8-K scheduled its call for 4:30 p.m. ET on July 30, 2026, after the regular session had closed. Meta's (July 29), Cerebras' (August 12) and Ulta's (March 12) 8-Ks are listed in the sources.
Four real 2026 gaps, run through the math
The setup is the same for every name. A hypothetical sell stop sits 5% below the prior close, and a hypothetical limit sits 1% below that stop. Prices are FMP daily data in US$. Daily bars don't show the order of trades, so the "plausible fill" for a stop-market order is the opening print. Real fills depend on the broker and the liquidity at that moment.
| Stock | Prior close | Hypothetical stop / limit | Gap-day open | Gap-day high / low | Stop-market (plausible) | Stop-limit that day |
|---|---|---|---|---|---|---|
| $META | $585.61 (Jul 29) | $556.33 / $550.77 | $526.00 (Jul 30) | $539.88 / $524.49 | ~$526.00, −10.2% | No fill: high $10.89 short |
| $RDDT | $178.04 (Jul 30) | $169.14 / $167.45 | $147.70 (Jul 31) | $151.38 / $135.22 | ~$147.70, −17.0% | No fill: high $16.07 short |
| $CBRS | $262.06 (Aug 12) | $248.96 / $246.47 | $220.00 (Aug 13) | $239.99 / $219.10 | ~$220.00, −16.0% | No fill: high $6.48 short |
| $ULTA | $624.70 (Mar 12) | $593.47 / $587.54 | $566.28 (Mar 13) | $579.45 / $534.68 | ~$566.28, −9.4% | No fill: high $8.09 short |
$META, July 30, 2026. The plan was to risk $29.28 a share, which is 5% of $585.61. Selling at the $526.00 open loses $59.61 a share, about twice the plan. That open sat only $1.51 above the day's $524.49 low and $13.03 under the $539.03 close. The stop-limit never had a chance: the day's high of $539.88 never got back to the $550.77 limit.
$RDDT, July 31, 2026. The planned risk was $8.90 a share, and the open-print exit cost $30.34. Here the open was not the worst price of the day. $RDDT closed at $140.67 and traded as low as $135.22. The stop-limit's $167.45 limit stood $16.07 above the day's $151.38 high.
$CBRS, August 13, 2026. The newly listed Cerebras opened at $220.00, just $0.90 above its $219.10 low, then closed at $231.01. The stop-market seller ate roughly $42.06 a share against a planned $13.10. The $239.99 high fell short of the $246.47 limit.
$ULTA, March 13, 2026. This was the "14% plunge" IMA covered in The Lipstick Effect Is Cracking. The close-to-close drop was 14.24%, from $624.70 to $535.72. Only 9.35% of that was the gap, though. The rest came during the session, so an exit at the $566.28 open beat the close by $30.56. The $579.45 high never reached the $587.54 limit.
The stop-limit's hidden bill: the ride after the miss
An unfilled stop-limit is not a loss avoided. It is a position still open. A Day order would have expired on the gap day. The table below assumes a GTC order your firm kept alive, and it marks the first session when the high reached the limit again.
| Stock | Lowest low while unfilled | First session the high reached the limit | Sessions after gap day |
|---|---|---|---|
| $META | $524.49 (Jul 30) | Jul 31, 2026 (high $558.33) | 1 |
| $CBRS | $215.77 (Aug 14) | Aug 17, 2026 (high $261.08) | 2 |
| $RDDT | $135.22 (Jul 31) | Aug 14, 2026 (high $184.28) | 10 |
| $ULTA | $443.60 (Jun 30) | Oct 14, 2026 (high $592.63) | 148 |
Read it both ways. In three cases a patient GTC limit got a second chance at a better price than the gap-open exit. $RDDT even opened at $175.78 on August 14, which IMA covered that day. But $ULTA holders waited 148 sessions and sat through a $443.60 low before the limit came back into range. A daily bar shows the price reached the limit, not that any given order filled there. Hindsight also cuts the other way: $CBRS closed at $120.66 on October 19, 2026, far below its $220.00 gap-day open.
Which order fits which goal, mechanically
| Goal | Order type | What the mechanics guarantee | What they don't |
|---|---|---|---|
| Be out once a level breaks | Stop (stop-loss) | Becomes a market order on trigger | Any price; a gap fills at or near the open |
| Never sell below a floor | Stop-limit | No fill worse than the limit | Any fill at all; a gap can leave you holding |
| Follow a rising stock | Trailing stop | Stop level ratchets with favorable moves | Gap protection; it still triggers like a stop |
| Sell at a target above the market | Plain limit | Limit price or better | That the target ever trades |
| Trade pre- or post-market | Limit (often required) | Price control in thin markets | Liquidity; quotes may be wide or absent |
These rows describe mechanics, not advice. On trailing stops, the SEC defines one as "a stop or stop limit order" whose stop price is a set percentage or dollar amount from the market. It "trails" favorable moves and stays fixed on unfavorable ones. So a trailing stop inherits every gap problem above. It just moves the trigger.
Three mistakes a gap punishes
- Stops parked on round numbers. A stop at $150 on $RDDT isn't analysis. It's a tidy number. On July 31, 2026, the stock opened at $147.70, below that level, so the stop would have triggered on the first print.
- Stop-limits with a too-tight limit. A 1% band missed on all four gaps. To have any chance of filling on the gap day, the limit had to be at or below the day's high. That was 7.81% under the prior close for $META, 8.42% for $CBRS, 7.24% for $ULTA and 14.97% for $RDDT. The SEC's example band, $2.50 under a $3.00 stop, is 16.7%.
- Forgetting the calendar. All four gaps followed an Item 2.02 results release. A stop held through an earnings date is effectively a bet on the gap, not just on the stop level. Our guide to reading an 8-K covers the filing itself.
More gap-day breakdowns sit under Markets and on the I'M A STOCK TRADER homepage.
Bull/Bear Verdict
Bull Case: Knowing the mechanics turns a vague "safety net" into a choice you can measure. A stop-market order prioritizes getting out, and a stop-limit order prioritizes the price. Either can be sized around a realistic gap instead of the stop distance alone.
Bear Case: Neither order caps a gap loss. In all four 2026 examples, the stop-market exit cost roughly 1.9 to 3.4 times the planned risk. The stop-limit filled none of the four on the gap day. Treat the stop price as a promise and the gap will correct you.
FAQ
Is a stop-loss order the same as a stop-market order?
Yes. The SEC describes a stop order as "also referred to as a stop-loss order." Once the stop price is reached, it becomes a market order.
Does a stop-limit order protect against an overnight gap?
It protects your price, not your exit. If the stock opens below your limit and never trades back up to it, the order stays unfilled. That happened in all four examples above.
Do stop orders trigger in pre-market or after-hours trading?
That depends on your firm. The SEC says Day orders generally don't carry over to extended hours, and many firms accept only limit orders in those sessions. Ask your broker how it handles stops outside 9:30 a.m. to 4:00 p.m. ET.
How far below the stop should the limit be?
There is no official number. The SEC's example uses a $3.00 stop with a $2.50 limit. A wider band raises the odds of a fill and lowers the worst acceptable price.