Gold's Opening-Range Breakout: London Is a Trap, New York Isn't
Share of sessions (%)
The opening-range breakout (ORB) is a beginner favourite: mark the first hour's high and low, trade the break. On gold it feels especially natural because the session opens are so active. We measured it mechanically — the first hourly close beyond the opening range, tracked to session end — separately for the London and New York opens across 16 years. Nearly 7,900 sessions.
London: the breakout that eats you
| London ORB outcome | Count | Share |
|---|---|---|
| Clean success (extends) | 689 | 16.8% |
| Fakeout (closes back inside) | 1,446 | 35.3% |
| Roundtrip / "Judas" (both sides break) | 1,938 | 47.4% |
| No resolution | 19 | 0.5% |
Read that roundtrip number again: almost half of London sessions break both sides of the opening range before doing anything durable. This is the classic "Judas swing" — a false move one way to grab liquidity, then the real move the other way. If you take the first London break at face value, you are the liquidity. Clean success happens under 17% of the time.
New York: the same setup, a very different animal
| New York ORB outcome | Count | Share |
|---|---|---|
| Clean success (extends) | 876 | 22.7% |
| Fakeout (closes back inside) | 1,897 | 49.2% |
| Roundtrip / "Judas" (both sides break) | 860 | 22.3% |
| No resolution | 223 | 5.8% |
New York's clean-success rate is higher (22.7% vs 16.8%), but the real difference is the whipsaw rate: 22% versus 47%. New York breaks are far less likely to sweep both sides. Once New York picks a direction, it tends to keep it; London probes both pockets of liquidity first.
Both resolve near a third — the framing matters
Counting only sessions that resolved cleanly one way or the other, both sessions land around a 32% success rate. That is not an edge you trade blindly in either session — but it reframes the ORB correctly: it is not an entry signal, it is a map of where liquidity sits. In London especially, the opening range is better understood as the levels that will get swept than as the levels to trade the break of.
We re-ran it as the textbook trade — 143 recent sessions
The numbers above describe what price did after a confirmed hourly break, averaged over 16 years. A long-run base rate hides a lot: individual years vary widely around the aggregate, and a setup that trapped traders in 2015 could in principle behave differently in today's tape. So we re-tested the London ORB as the literal textbook trade on recent data — M5 candles, roughly seven months of London sessions, no confirmation filter: enter on the first break of the first-hour range, stop at the opposite edge, target one range-width away. Risk one to make one.
| 143 London sessions · M5 · first break · 1:1 | Share |
|---|---|
| Reached the 1× target first | 44% |
| Hit the stop first | 56% |
| Sessions that traded both edges of the range | 64% |
At one-to-one payoff, a 44% win rate loses money before costs. And the failure mode is the same one the 16-year data shows: on nearly two-thirds of recent breakout days, gold traded through both edges of the opening range before choosing a direction. The Judas swing is still the base case — slightly more often than the long-run average, not less. Two measurements, different candle sizes, different eras, same verdict: the first London break is not the move.
The video on this page walks through one of those 143 sessions candle by candle — the textbook short, the slow bleed toward the target that never arrives, the stop taken at the opposite edge, and the real move that starts only after it.
What this means
The precise stop and target geometry stays private, but the general lessons carry across instruments:
- The same setup has a different personality in different sessions. London's opening range is a two-sided liquidity trap; New York's is more directional. One number ("32% success") hides that completely.
- On gold, a first-hour break is usually not the move — especially in London, where the Judas swing is the base case, not the exception.
- Treat the opening range as a liquidity map, not a trigger. Where it breaks and holds is the signal; the first break rarely is.
The edge isn't the breakout. It's knowing which session turns that breakout into a trap.
Studies like this become the filters inside a documented playbook — the research → playbook → backtest → live loop, locked to one instrument at a time. That playbook runs live on Cortiq.