The Gap Map: When US Equity Gaps Continue, Fade, and Mean Nothing
- 4 days ago
- 3 min read
TL;DR — We ran 6,552 overnight-gap events across SPY, QQQ, IWM, NVDA, TSLA and AAPL from 2022 to 2026. Large gaps (>2%) do not extend — the day is already mostly done. Mid-size up-gaps (1–2%) are the only bucket where we see a real edge (53.7% continuation). Everything in between is structural drift. The trade isn't the gap, it's how we read the first 30 minutes after it.
Every retail course teaches the same two slogans: gap and go or gaps fill. Both are right somewhere and wrong somewhere else. The desk question we actually care about is sharper: at what gap size does same-day drift beat zero, where does continuation cross 50%, and where is the open just noise around yesterday's close?
What 6,552 gap events tell us
We pulled daily OHLCV from yfinance for SPY, QQQ, IWM, NVDA, TSLA and AAPL between January 3, 2022 and May 13, 2026. For every session we computed the overnight gap (Open / PriorClose − 1) and the intraday move (Close / Open − 1), then bucketed the gaps into nine bands. Three findings stand out, and none of them is "buy the gap up."
The boring middle drifts up. In gap buckets between −0.5% and +0.5%, median intraday return is positive — roughly +0.04% to +0.14%. Small, but reliably one-sided. That's the structural bid in liquid US equities on calm tape, nothing more.
Large up-gaps don't extend. Gap-ups above 2% (n=327) had a median intraday return of only +0.03% and a same-sign continuation rate of 50.5%. By the time price has already moved 2%+ pre-market, the day is mostly done. Paying for the breakout adds risk without adding edge.
The trade-able regime is the mid-size up-gap. The 1–2% bucket (n=628) is the only band where continuation beats coin-flip (53.7%) with a positive median drift (+0.11%). It's a real edge — and the most crowded one. Execution quality is what keeps it positive after costs.
Why "gap and go" stopped working
The retail playbook was built on a microstructure that's gone. Pre-2020, big overnight moves were rare, options gamma was a smaller share of opening flow, and a 1.5% gap usually reflected new fundamental information the regular session hadn't finished pricing. Today, large gaps are routinely driven by overnight options expiry, ETF rebalances and concentrated single-name flow in mega-cap tech. Most of the information content is already exhausted before the bell.
The first 30 minutes tell us which regime we're in
An open doesn't need to be traded at 09:30. The first hour is where the day's regime declares itself. Looking at the median minute-by-minute drift from the 09:30 open across SPY's last 60 trading days, gap-up sessions drift up modestly in the first hour and then stall. Gap-down sessions grind — no fast snap-back. Flat sessions oscillate until late-day flow takes over.
Our practical read: the highest-quality continuation evidence shows up between roughly 09:45 and 10:30, not at the bell. By 12:30 the curve flattens regardless of regime — that's when new positions pay for liquidity without much expected drift in return.
How we trade the gap map
Bucket the gap before the bell, not after. The size of the overnight move tells us what kind of day to expect — and large gaps aren't the high-edge bucket. Knowing that a 2.4% pre-market move has a 50/50 continuation rate is more useful than knowing the stock is "up big."
Let the regime declare itself. The first 15–30 minutes carry more information than the gap. Our Vortex Flow order-flow stack — CVD, VWAP location, heatmap structure — separates a gap being absorbed from one being defended. A range break with confirming CVD is a different trade than the same break with flat CVD and a widening spread.
Treat routing as part of the trade. Mid-size gaps are statistically the highest-edge bucket and the most crowded. The difference between a profitable and break-even quarter is the cost of liquidity. Our DMA through Sterling Trader Pro — with precise limit offsets and disciplined cancel logic — isn't a back-office detail; it's the strategy.
The takeaway
The gap is a regime label, not a signal. The biggest gaps mean the day is mostly done. The mid-size gaps are where statistical edge lives, and only for traders who can route into them without paying the spread twice. Everything else is the structural drift of liquid US equities — real but small, and best harvested with size and discipline rather than chased candle-by-candle.
At Vortex Capital Group, we give qualified traders the infrastructure to operate at the level the data demands: sub-millisecond DMA, real-time order-flow tooling, and capital sized to the strategy rather than to the trader's account balance.
Comments