Gap Trading — Gap-and-Go vs Gap Fill: What Backtests Really Say

A price gap is the most dramatic thing you'll see on a chart: the market opens a few percent away from yesterday's close, leaving behind a void where not a single trade took place. Two opposing schools have grown up around that void. Gap-and-go says: the gap is fuel, play the continuation. Gap fill says: the gap is an overreaction, play the close. Both have their believers, and both can show pretty screenshots of profits.

Now for the part you won't find in course ads: backtests of simple, mechanical gap rules — tested on large samples by QuantifiedStrategies — averaged around 0.06% per trade. Six hundredths of a percent. Before commissions and slippage. This article is about where that gap between the legend and the numbers comes from — and what can honestly be done with gaps.

Educational disclaimer: this material is for educational purposes only and is not investment advice. Trading at the session open is the most chaotic moment of the day — slippage and widened spreads are the norm, and most retail day traders lose money. The backtest results cited are historical and don't guarantee future results.

Where Gaps Come From and Why Not Every Gap Is the Same

A gap forms when the price valuation changes between a close and an open with no trading in between to carry it there smoothly: earnings after hours, macro data before the open, weekend geopolitical events. On markets with breaks (stocks, indices, commodities), gaps are an everyday occurrence. Crypto spot has none — the market runs 24/7 — but there's a well-known exception: CME futures contracts don't trade on weekends, so Monday's open can leave a "CME gap" on Bitcoin.

Critical to the whole strategy: not all gaps are equal. The classic taxonomy — covered in more depth in our article on gap types — identifies four:

• Common gap — a small gap with no catalyst, mid-range, on average volume. Statistically the most likely to fill, often the same day.
• Breakaway gap — a gap breaking out of a consolidation, backed by news and volume. This is the one most likely to continue the move.
• Runaway (continuation) gap — a gap mid-way through an established trend, a sign of strength.
• Exhaustion gap — a trend's last gasp, followed by a reversal. The problem: you can't tell it apart from a runaway gap until after the fact.

And here's the first honest answer to the two schools' dispute: both are right, just about different gaps. QuantifiedStrategies' test results form a coherent picture: breakaway gaps backed by news statistically continue more often, while plain common gaps — without a catalyst — tend to fill during the same session. A "play every gap the same way" strategy mixes these two populations together and gets an average close to zero as its reward.

Gap-and-Go — Playing the Continuation

The variant popularized by American day traders (Warrior Trading among others): you look for a stock that's up several percent before the open on a specific catalyst — earnings, an upgrade, industry news — with above-average premarket volume. Entry comes after the open, in the direction of the gap, most often on a breakout of the high of the first few minutes of trading (mechanics close to the ORB strategy). Stop below the opening low, target based on a multiple of risk.

The logic: a large gap with news is information the market is still digesting. Funds won't buy their whole position at the open — they'll keep buying, and latecomers will pile in as they see strength. That's a real mechanism, but it has boundary conditions: catalyst, volume and selection. Gap-and-go played on every gap, without a news filter, degenerates into a coin flip with costs.

Gap Fill — Playing the Close

The opposing school: a gap unsupported by fundamentals is emotional overreaction from overnight orders. Once the initial demand is exhausted, price returns to yesterday's close — it "fills the gap." We break down the detailed statistics of this phenomenon in the article on gap fills; here's the short version: small gaps without a catalyst really do fill surprisingly often within the same-session horizon.

But that sentence is exactly what scammers feed on. "The gap always fills" is a myth built on the absence of a time horizon: given infinite time, almost every level gets touched, which makes the claim untestable and useless. The tradeable version reads: "gap type X fills Y% of the time by the end of the session" — and you need to know that statistic for your own market before betting money on it. On top of that comes a risk asymmetry: playing for a fill means, by definition, standing against a fresh move. Hit a breakaway gap, and the market can leave you behind without a second glance.

What the Numbers Say — 0.06% of Truth

The most important number in this article: simple, mechanical gap rules (buy a gap down / play a continuation of a gap up, no filters) averaged around 0.06% profit per trade in QuantifiedStrategies' tests. To be clear: that's before costs. On a retail account with commission and slippage at the open — the worst moment of the day for execution — that edge disappears or flips sign.

What follows from this? Three things:

1. The gap alone is not enough. The fact that the market opened with a gap is too widely known and too easy to spot to be an edge by itself.
2. If an edge exists, it lives in selection. Gap size, presence of a catalyst, relative volume, trend context — the filters that separate a breakaway from a common gap are the whole game. That's the exact same conclusion drawn from ORB research: a strategy without a day-quality filter dies, with one it can get interesting.
3. Costs decide everything. With an edge measured in hundredths of a percent, the difference between a 0.02% commission and a 0.1% one is the difference between a strategy and a machine for grinding down your account.

An honesty note: the QuantifiedStrategies numbers come from their publications — verify them at the source before using them in decisions, and more importantly, reproduce them on your own data with your own costs.

How to Approach Gaps Step by Step — If at All

1. Gather statistics for your own market. Before playing anything: pull the data, bucket gaps by size (e.g. in ATR units) and the presence of a catalyst, and measure continuation and same-session fill rates. It's one evening of work that saves you months of losses.
2. Pick one school for one gap type. Continuation → only large gaps with news and volume (the macro calendar is required reading before a session). Fill → only small gaps with no catalyst, mid-range.
3. Entry and stop. Gap-and-go: entry on a breakout of the first few minutes' high, stop below the opening low. Gap fill: entry on the first signs of exhaustion, stop beyond the day's extreme, target at yesterday's close.
4. Fixed-percentage position sizing based on the stop — and awareness that opening-bell slippage tends to run wider than usual.
5. Close by end of session. Fill and continuation statistics apply to an intraday horizon; carrying a position overnight is a different strategy with a different (unresearched-by-you) distribution.

Numerical example (illustrative): a stock closed yesterday at $100, opens today at $96 (−4%) with no news whatsoever — a typical gap-fill candidate. After 20 minutes price stops falling at a low of $95.20. Entry long at $95.80, stop at $94.90 (risking $0.90 per share), target $100 (potential of $4.20, roughly 4.7R). $10,000 account, 1% risk = $100 → position size 100/0.90 ≈ 111 shares. If the gap hasn't filled by the end of the session, you close at market — no negotiating.

When It Doesn't Work and Common Pitfalls

• Playing every gap. That's the surest road to the average 0.06% minus costs. Most gaps are noise — no trade is itself a position.
• Gap fill against the news. Playing for a close after strong earnings is catching a falling knife in reverse: breakaway gaps, by definition, don't fill (within a tradeable horizon).
• The "CME gap" myth on Bitcoin. Yes, CME futures gaps have often gotten filled. No, nobody credible has shown an edge after costs on a large sample — the sample of Monday gaps is small, and a few extreme cases drive the result. Test it before you believe it.
• Execution at the open. A backtest assumes the chart price; a real open means wide spreads and slippage. With a thin-edge strategy, that's not a footnote — it's the whole ballgame.
• Exhaustion disguised as runaway. A gap in the direction of a strong trend looks like confirmation and can be its ending. You only tell them apart after the fact — which is why the stop loss isn't optional.
• Small samples. Large catalyst-driven gaps happen a dozen or so times a year on a single instrument. Conclusions from 15 observations are an anecdote, not a statistic.

The lesson from gaps is the same lesson as with most "simple" strategies: the phenomenon is real, but the edge doesn't live in the phenomenon — it lives in the selection. A gap isn't a signal — it's a question, and the answer is context: catalyst, volume, gap type, and your calculated costs. Whoever trades gaps without their own statistics is providing liquidity to those who have some.

FAQ

Rafał — Krypto Bez Ściemy

Trader and founder of Strefa Tradingu. He’s been breaking crypto down on YouTube for years — no hype, no signals, with a focus on market structure and risk management.

kryptobezsciemy.plInstagramYouTubeKlub Zero