The golden pocket is the band between the 0.618 and 0.65 Fibonacci retracement of an impulse move — the inverse golden ratio with a conventional upper bound wrapped around it. The playbook: an impulse leg establishes direction, price pulls back into the pocket, and you join the trend there with a stop tucked beyond the 0.786 level. Crypto trading made it arguably the most-quoted retracement entry in the market, and on any trending chart you can find beautiful examples within minutes.
What you cannot find is an honest mechanical test, because the setup hides three free choices: which swing anchors the leg, whether you measure wicks or closes, and what counts as the entry trigger. We publish only numbers our own deterministic engine produced on real data, and the engine does not yet ship a fib-retracement-zone primitive (automatic leg anchoring plus band math). So this page does the next-best thing — it pins the exact rules in advance, like a pre-registered study. The 12-month BTC and ETH results appear here the moment that primitive ships; until then, the replay terminal and its Fibonacci tool are how you build your own sample.
The exact rules we’ll test
- Impulse leg: the most recent confirmed swing low to confirmed swing high (fractal lookback 5, wick to wick) on 15-minute candles; inverse for shorts.
- Golden pocket = the 0.618–0.65 retracement band of that leg.
- Setup is live only while the leg’s origin holds; a close beyond the 0.786 level before entry cancels it.
- Entry: the first 15-minute candle that trades into the band and closes back above it (below for shorts) — a confirmation close, not a blind limit.
- Stop beyond the 0.786 level of the leg; target 2R; risk 1% per trade; 0.05% commission per side.
- Missing primitive: a fib-retracement-zone detector that anchors the leg algorithmically and computes the band — the engine doesn’t ship it yet, so results publish when it lands.
Backtest data: in the queue
We publish only numbers our own deterministic engine produced on real data — no borrowed or estimated stats. This setup needs a detection primitive the engine doesn’t ship yet, so its 12-month results will appear here the moment that lands. The strategies with completed research are on the strategies hub, and you can already practice this setup bar-by-bar in the free replay demo.
Why no two traders draw the same pocket
Hand the same chart to five fib traders and you get five drawings. The leg is a judgment call — the last swing pair, or the dominant move on a higher timeframe? Anchored to wicks or to closes? Re-drawn when price makes a marginal new high, or kept? Each choice moves the pocket by enough to turn a fill into a miss, which is why "price respected the golden pocket" is so often visible only in hindsight, after the winning anchor is obvious.
The band itself is also a convention, not a law. The golden pocket is 0.618–0.65; ICT’s related OTE concept is usually drawn 0.62–0.79 with 0.705 as the sweet spot; older fib traders simply watched 0.5 and 0.618 as lines. And the trigger varies again — resting limit at first touch versus confirmation close versus a lower-timeframe structure shift. None of these is "wrong", but each produces a different trade list, a different stop distance, and a different sample. A backtest means something only after every choice is pinned — which is what the rule list above does, and what the engine’s future fib-zone primitive has to encode.
How to backtest the golden pocket on Secuora
Until the fib-zone primitive ships, the replay terminal plus the Fibonacci tool is the honest way to build a sample of your own.
- Open the free replay demo at /backtest/demo — no sign-up, and the drawing tools include a Fibonacci retracement.
- Replay to a clean impulse leg and draw the fib from the swing low to the swing high (top to bottom for shorts), wick to wick.
- Step forward candle by candle and act only when price trades into the 0.618–0.65 band — no repositioning the anchors after the fact.
- Sign up free and enter in the replay terminal with a simulated order on your chosen trigger — first touch or confirmation close — stop beyond the 0.786, take-profit at 2R.
- Log every attempt in the journal with screenshots and confluences; thirty replayed trades will tell you more than any thread, and the AI backtester takes over the moment the fib-zone primitive lands.
Frequently asked questions
What is the golden pocket?
The price band between the 0.618 and 0.65 Fibonacci retracement of a move. After an impulse leg, fib traders treat a pullback into that pocket as the highest-probability place for the trend to resume, with invalidation usually placed beyond the 0.786 level or the leg’s origin.
Why 0.618 and 0.65 specifically?
0.618 is the inverse golden ratio and the most-watched Fibonacci level; 0.65 is the conventional upper bound that turns the line into a tradeable band. There is no physical law behind the numbers — the honest case is reflexivity (enough traders watch the same levels) plus the fact that a roughly 62–65% pullback is a deep retracement that still leaves the leg intact.
Is the golden pocket the same as ICT’s OTE (optimal trade entry)?
They overlap but aren’t identical. The golden pocket is the 0.618–0.65 band, while OTE is typically drawn as the 0.62–0.79 region with 0.705 as the sweet spot. Both are fib-retracement entry concepts — which is exactly why a backtest is meaningless until the band, the anchors and the trigger are pinned exactly.
When will golden pocket backtest results be published?
When our engine ships a fib-retracement-zone primitive — automatic leg anchoring plus the band math. We publish only numbers our own deterministic engine produced on real data, so this page carries the pre-registered rules now and gets its 12-month BTC/ETH results, win or lose, the moment the primitive lands.