Why do stocks gap up or down between one day’s close and the next day’s open? Because nothing traded in between. Earnings land at 4:15pm, an analyst cuts a rating before the bell, a competitor’s warning drags the whole sector overnight — and none of it has a price attached until the exchange opens again. The last trade on Tuesday and the first trade on Wednesday can sit an hour apart, a day apart, or a full weekend apart, with a pile of new information wedged in the gap between them. The stock doesn’t walk to its new price. It jumps.
That’s the mechanical answer, and it’s the whole answer. There’s no hidden hand moving the stock while the market’s closed. There’s just a market that stops accepting trades at 4pm and starts again at 9:30am, and every piece of news that shows up in between gets priced in one shot instead of gradually. A gap isn’t the market being irrational. It’s the market catching up.
Why do stocks gap up or down overnight in the first place
Regular trading hours are the only hours where a trade actually clears at a public price on the tape most people watch. Outside that window, information doesn’t stop arriving — earnings releases, guidance cuts, upgrades, downgrades, a Fed comment, a whole index moving on macro data — it just has nowhere to land. Pre-market volume exists, and some of that news does get traded on thin overnight liquidity, but the number that matters to most of the market is the next regular-session open. When the bell rings, every account that wanted to react to the news reacts at once, and the opening print reflects all of it in a single trade instead of the usual grind of one order following another.
Earnings are the cleanest example because the timing is scheduled. A company reports after the close, beats on revenue, guides higher for the next quarter — and the stock that closed at 41 opens the next morning at 46. Nobody sold it up through 42, 43, 44, 45 the way it would happen mid-day. The five dollars just isn’t there on the chart. That’s the gap, and it’s the direct, mechanical result of the news hitting while the tape was shut.
The same thing happens with analyst moves. A downgrade posted at 6am doesn’t wait for the stock to open before it starts influencing pre-market orders. By 9:30, the order book already reflects traders who read the note and decided what they wanted to pay. Market-wide gaps work the same way at a bigger scale — a weak overnight session in Asia or Europe, a surprise data print before the US open, and stocks across the board open away from where they closed, independent of anything specific to each individual name.
Treat the gap as a brand-new level, not a promise
Here’s where I stop agreeing with how most people talk about gaps. There’s a persistent idea that a gap is either going to hold or going to fill, as if the pre-market price carries some built-in conviction about which one it’ll be. It doesn’t. The pre-market price is built on a fraction of the volume a stock does during regular hours — sometimes a few thousand shares deciding where a stock “should” open, on a name that trades tens of millions of shares once the real session starts. That thin volume can produce a gap that has almost nothing to do with where real buyers and sellers will actually transact once everyone’s back at the desk.
So I mark the gap edges the way I’d mark any other level — the prior close, the pre-market print — and then I do nothing with them until the regular session tells me something. Not because the gap is meaningless. Because it’s unverified. A level built on light volume before 9:30 hasn’t been tested by the crowd that actually moves the stock, and treating it as settled fact before that test happens is how a clean piece of news turns into a bad entry.
What the first part of the session actually tells you
Once the bell rings, the gap starts writing its own story, and there are three versions of it. The gap can hold — price opens at the gap level and just keeps going in the same direction, which tells you the crowd showed up and agreed with the pre-market number. The gap can fill partially — price gives back some of the move but stalls before getting back to yesterday’s close, which tells you there’s real disagreement, some conviction on both sides. Or the gap can fill completely — price grinds all the way back to where it closed the day before, which tells you the pre-market move had no support once volume actually showed up.
Each of those is a different piece of information, and none of them is knowable from the pre-market print alone. This is the same discipline I use on an opening range breakout — the first candles after the bell tell you more than the setup itself — and it rhymes with how I treat a fair value gap, where the untraded space is a place to watch, not a certainty about what happens next. A gap is the same category of thing. It’s a mark left by an imbalance. What matters is what price does once real volume gets a vote.
The trade that taught me to wait
This was Snowflake, a Thursday, after a mixed earnings report the night before — revenue beat, guidance came in soft. The stock closed at 178 and was trading around 171 in the pre-market on maybe 40,000 shares changing hands over three hours. Down about 4%, on thin volume, on a report that read to me like the market was overreacting to the guidance line and ignoring the beat.
I had an opinion, and I acted on it before the opinion was tested. At 9:41, eleven minutes into the session, I bought at 172.40, expecting the gap to fill back toward 178 once cooler heads looked past the guidance number. The regular session opened at 170.85 — below my pre-market read already — and kept sliding. By 9:58 it was 167.20. I was down almost $5 a share on a trade I’d built entirely on a pre-market price nobody outside 40,000 shares had actually agreed to. I sat in it another twenty minutes hoping for the fill I’d predicted, then took the stop at 166.90, a loss of 5.50 a share on a full position. The stock kept falling into the afternoon and closed near 163. There was no fill that day. There wasn’t one the next day either.
What I got wrong wasn’t the read on the earnings report — plenty of names do bounce back after an overreaction like that. What I got wrong was treating a pre-market price built on 40,000 shares as if it were a confirmed level, instead of waiting to see what 9:30 actually did with it. If I’d waited even the first fifteen minutes of real volume, I’d have watched the gap get sold harder, not bought — the opposite of what I needed to see before betting on a fill. The information was sitting right there in the first candles. I just traded before I let the chart give it to me.
What I do differently now
These days a pre-market gap gets the same treatment as any other unconfirmed level on my chart. I mark the prior close and the pre-market print, and I watch. I’m not looking for a signal in the first tick — I’m looking for the shape of the first ten or fifteen minutes of real volume. Does price open near the pre-market print and hold there. Does it immediately start eating back into the gap. Does it blow through the pre-market level entirely, in either direction, showing conviction nobody could see in the thin overnight tape.
None of that tells me where the stock ends the day. It tells me whether the gap is worth respecting yet. A gap that holds through the first real volume of the session has earned some trust — real size showed up and agreed with the pre-market read. A gap that gets sold or bought through immediately hasn’t earned anything, no matter how clean the news looked at 6am. The report is the same either way. The chart’s answer is what changes, and the chart’s answer only exists once the exchange is actually open.
The habit underneath all of this
I don’t trade news. I trade what the chart does after the news lands, and a gap is just the clearest version of that gap between the two. The mechanics explain why the price jumped. They don’t tell me whether the jump holds. That second question only gets answered by the same thing every other level on my chart gets answered by — volume showing up and either defending a price or abandoning it. Most mornings, the right move on a gap isn’t a trade at all. It’s a mark on the chart and twenty minutes of watching before deciding there’s nothing to do.
I trade with a blank chart, no news feed running, and levels built from wherever price already left something behind — gaps included — in Static, the free daily chart room run by Draw Lines Make Money. If waiting for the open to confirm the gap instead of guessing at it makes sense to you, you can sit in and watch how it’s done live:
Join the free Static chart room →
Disclosure: that’s an affiliate link — I may earn a commission if you join a paid tier later, at no extra cost to you. The free room is free.