Fading the Overnight Gap: How Gap Fill Trading Works on Micro Futures
"Gaps always fill" might be the most repeated line in futures trading. Like most trading folklore, it's built on something real and then stretched past the truth. I trade the gap fade every week with an automated strategy, so let me give you the version I'd want if I were starting out: what a gap actually is, why fading it works often enough to build a strategy on, and the filters that keep the bad days survivable.
What is an overnight gap in futures?
Futures trade nearly around the clock, so the "gap" isn't an empty space on the 24-hour chart — it's the difference between yesterday's regular-session close and today's regular-session open. Overnight, the Globex session drifts on thinner participation: Asia reacts, Europe repositions, a headline lands at 3 a.m. When the U.S. regular session opens somewhere other than where it closed, that distance is the gap.
The key insight: the overnight move happened on thin participation. The regular session brings the full crowd back, and the full crowd gets to vote on whether the overnight price deserves to stand.
Why do gaps fill?
Because often the overnight move was an overreaction that daytime liquidity walks back. Traders caught the wrong way unwind; value-oriented traders see the stretched price and lean against it; and the prior close acts like a magnet because it's the last price the full market agreed on. When that repricing happens, price "fills the gap" — trades back to yesterday's close.
Fading the gap means positioning for exactly that: the market gaps up, I enter short and target the fill; gaps down, I enter long. It's a mean-reversion trade with an unusually clean, pre-defined target.
When does gap fading go wrong?
Three ways, and every one of them shapes how my strategy is built:
The news gap. A gap driven by a real overnight catalyst — a Fed surprise, an earnings shock in the index heavyweights, a supply headline in crude — isn't an overreaction to fade; it's new information being priced. These are the gaps that don't just refuse to fill, they extend. My strategy pairs with my News Calendar indicator so a major release inside the gap window makes it stand down for the day automatically.
The gap that's too big. Past a certain size, the fill target is simply too far away to reach within the session at acceptable risk. A configurable size filter skips those days.
The gap that's too small. A few ticks of gap isn't worth the spread, slippage, and risk to capture. Same filter, other side.
Between the two size cutoffs sits the tradeable middle: gaps big enough to pay, small enough to plausibly close.
How do you time the entry?
The rookie version fades the open immediately — and gets steamrolled on the days the market opens and keeps running. My version waits for evidence: the entry trigger is the first bar that closes back inside the gap. That close says the fade has actually started; I'm joining a reversion in progress instead of guessing its start. There's deliberately no multi-bar confirmation on this strategy — gap fades tend to move fast or not at all, and waiting for extra bars eats the exact profit room the trade depends on.
One more filter I added after watching it matter: a profit-room check. If price has already drifted most of the way to the fill before a valid entry appears, the day passes. Late entries on a mostly-filled gap are exactly the low-reward, full-risk trades a rules-based system exists to refuse.
Do all gaps fill?
No — and be suspicious of anyone quoting a precise "percentage of gaps that fill" without telling you the market, the gap size, and the window they measured. The honest summary: intraday fills happen often enough to build an edge on when you filter by size and skip news days, and rarely enough that unfiltered "fade everything" is a fast way to donate. That gap between folklore and edge is the whole reason this is a filtered, automated strategy and not a slogan.
Which micro futures suit the gap fade?
I run it as TurnKeyGapFade on the CME micros, and my tuned templates cover it on gold (MGC) and silver (SIL) among others — metals are prone to overnight headline drift that daytime liquidity reassesses, which is the exact behavior the fade wants. The Markets page covers how each of the six micros behaves. Micros keep the dollar risk per trade small, which is what makes the losing days data instead of drama.
If you want the other half of the picture, the opening range breakout post covers the trend side of the same morning — the two setups want opposite market conditions, which is exactly why I run both.
FAQ
What's the difference between a gap fade and a gap-and-go?
Opposite bets on the same open. The fade positions for the gap to close back to yesterday's close; gap-and-go positions for the gap to extend in its direction. Filters decide who's on the right side: catalyst-driven gaps favor continuation, quiet stretched opens favor the fade.
When do you consider a gap "filled"?
When price trades back to the prior regular session's closing price. Partial fills happen; my strategy's stop and target sizing accounts for the fact that not every fade reaches the full close.
Do all gaps fill?
No. Fills happen often enough to build a filtered edge on, and rarely enough that fading every gap without size and news filters is a reliable way to lose money. Anyone quoting you a precise fill percentage without naming the market, the gap size, and the measurement window is selling folklore.
Want the implementation instead of the homework? TurnKeyGapFade runs this setup in NinjaTrader 8 with the size, news, and profit-room filters already built in — Starter is $99/mo for any one strategy.
Educational content, not financial advice. Futures trading involves substantial risk of loss and is not suitable for all investors. Any performance data referenced on linked pages is hypothetical backtested data — see the disclosure on those pages.