Quasimodo Pattern (QML) — The Reversal Smart Money Loves

The Quasimodo pattern (QML, from "Quasimodo Level") is a favorite reversal setup among smart-money-style traders: price in an uptrend makes one more, higher high — and then, instead of correcting, breaks the previous higher low. What's left on the chart is a lopsided, hunchbacked structure, which is how the pattern got its name from the bell-ringer of Notre-Dame. In a downtrend, everything works as a mirror image.

Why does this structure generate so much excitement? Because it describes something ICT terminology names outright: the new high is a liquidity sweep above the old high, and the break of the low is a market structure shift (MSS/CHoCH). QML is the same mechanism we cover in the article on liquidity sweep vs liquidity run — just packaged as a pattern with a defined entry zone and a defined place for a stop.

One caveat up front, in the spirit of "no sugarcoating": QML has no Bulkowski statistics and no published, rigorous measurement of its performance. Every "70–80% win rate" you find online for it is unverified. Here we're trading structure and risk-reward, not a measured edge — and we'll say that honestly.

How to Identify a Quasimodo Pattern

Bearish variant (end of an uptrend):

1. Uptrend: price makes higher highs (HH) and higher lows (HL).
2. Higher-timeframe context is mandatory: price reaches a significant higher-timeframe level — resistance, a supply zone on the daily or 4-hour chart. A QML drawn in the middle of nowhere isn't a signal, it's decoration.
3. The head: price makes one more, higher high (HH) — this is what sweeps stops and breakout orders above the previous high.
4. The break: instead of holding structure, price falls and breaks the previous higher low (a new LL). At this point, the uptrend sequence is violated: you have an HH followed by an LH-and-LL forming.
5. The QML zone: the area between the second-to-last high and the head-high. This is where you wait for price to return and enter short.

The bullish variant is a mirror: a downtrend (LL/LH) reaches higher-timeframe support, price makes one more, lower low (the head sweeps liquidity below the previous lows), then breaks the previous lower high. The QML zone spans between the second-to-last low and the last low.

The best environment for the pattern is the 4-hour and daily timeframe — that's where the levels QML forms at carry real weight and the structure isn't noise. On the 1- to 5-minute chart, the identical shape forms a dozen times a day and means as much as any other squiggle of microstructure.

What the Numbers Say — Honestly: There Aren't Any

QML doesn't appear in the Encyclopedia of Chart Patterns. Bulkowski never catalogued this structure, and nobody else has published a test on thousands of cases with a clear failure definition. This needs to be said plainly, because online, QML has grown a body of numbers pulled out of thin air.

What do we know indirectly? QML is a structural cousin of the head and shoulders pattern — three extremes, the middle one highest, a signal on the break of the neckline. The head and shoulders has real measurement: an average decline of about 22% on US stocks on the daily chart, and 29% in a bear market with a mere 1% failure rate — one of the best bearish patterns in the catalog. The difference: QML doesn't wait for a symmetrical second shoulder — it enters earlier, on price's return to the zone between the two highs. Entering earlier means a better price and a tighter stop, but also less confirmation. Whether that trade-off comes out ahead has not been measured.

A second indirect clue: the sweep-plus-structure-shift mechanism is the core of both ICT and Wyckoff thinking (UTAD in distribution is structurally the same move). Three independent schools converging on the same mechanism is circumstantial evidence it describes something real in market microstructure — circumstantial, not proof.

The practical conclusion: since you don't have a measured edge, discipline around risk-reward has to substitute for it. Trade QML only where the stop beyond the head and the target at the opposite extreme give at least a 1:2 ratio — at that ratio, even a 40% win rate keeps the account in the black.

How to Trade QML (Bearish Variant)

1. Higher-timeframe context. On the daily/4-hour chart, find resistance or a supply zone that the uptrend is reaching. Without this level, there's no setup — it's the zero filter.
2. Wait for the full structure. A new HH (the head) and a break of the previous HL. A new HH alone isn't QML; a drop with no prior sweep isn't either. The structure needs both legs.
3. Mark the QML zone between the second-to-last high and the head. Inside the zone, look for confluence: an order block, an FVG, the level the break originated from — this refines the entry.
4. Wait for price to return to the zone. Don't chase price. If the market leaves without a retest — fine, wait for the next one.
5. Confirmation on a lower timeframe. Drop to the 5- or 15-minute chart and wait for a downside structure shift after price enters the zone. A blind entry "because it touched the zone" is asking for a second sweep.
6. Stop loss: beyond the head, with a buffer. Not right at the extreme — the head is a magnet for a repeat test.
7. Take profit: the QML pattern's low (the break level / the last LL). Minimum RR 1:2 — if the geometry doesn't offer it, the setup doesn't qualify. Extended target: the next higher-timeframe level.

Bullish variant: everything mirrored — the zone sits between the last two lows, the stop goes below the head, the target sits at the pattern's high.

QML's relationship to CHoCH — because this is where confusion reigns. Change of Character is a break of the last structural low/high — confirmation of a trend change. QML can form earlier, on the break of a swing that wasn't structural — giving an earlier, but less confirmed, signal. The practical hierarchy: every CHoCH contains a QML structure inside it, but not every QML gets a CHoCH. The QML zone serves as the entry location, CHoCH as directional-bias confirmation — mix up these roles and you enter either too early or too late.

The most common mistakes from practice: drawing QML in the middle of consolidation (without a trend there's nothing to reverse), treating every higher high as "the head" (the structure requires both legs: the sweep and the break), placing the stop exactly at the head's extreme (a second test of that level is routine — a buffer is mandatory), and trading a bullish QML against a bearish daily bias — a lower-timeframe structure doesn't beat higher-timeframe context.

Myth vs. Measurement — What's True Here and What's Marketing

Structural truth: QML does a solid job describing the trap mechanism — a breakout that sweeps liquidity followed by a fast structure shift. Three schools describe the same phenomenon in three languages (sweep+MSS, UTAD, bull trap). The pattern also offers clear, falsifiable rules: a specific zone, a specific stop, a specific target.

Marketing: "an institutional pattern with 80% accuracy," "a level the banks always defend." Nobody has measured this. QML without a higher-timeframe level, without confirmation, and without a 1:2 RR is just catching a falling knife with a nicer name. A separate trap: signal overproduction — on low timeframes, HH-plus-LL structures form by the dozen every day, and most mean nothing. QML is rare by definition, because it requires simultaneity: a trend, a higher-timeframe level, a sweep, a break.

How do you verify for yourself whether QML works on your market? Keep a journal: log every setup that meets the full definition (trend, HTF level, sweep, break, return to the zone) before it resolves, with a stop and target set in advance. After 50–100 cases you'll know more than any smart-money course could teach you. That's an inconvenient piece of advice — it takes months instead of a weekend — but it's the only one that turns an unmeasured pattern into your own, measured statistic. Unmeasured patterns aren't forbidden; pretending the measurement exists is.

Standard caveat: the reversal-pattern numbers cited above (head and shoulders: ~22% average decline) come from Bulkowski's measurements on US stocks, daily timeframe. QML itself remains unmeasured — treat it as a structural framework with forced risk-reward discipline, not as a statistical edge.

FAQ

What is the Quasimodo pattern (QML)? A reversal pattern: in an uptrend, a new higher high (the head, a liquidity sweep), after which price breaks the previous higher low. The zone between the last two highs is the short entry location. In a downtrend — mirrored. It requires a higher-timeframe level, otherwise it isn't a signal.

Where do you place the stop loss and take profit on a QML setup? SL beyond the head with a buffer, TP at the opposite extreme of the pattern, minimum RR 1:2. Geometry that doesn't offer 1:2 disqualifies the setup — it's the only protection you have on a pattern with no measured edge.

Does the QML pattern have any confirmed statistics? No. It doesn't appear in Bulkowski's catalog and nobody has published a rigorous measurement of it. Its closest measured relative — head and shoulders — averages about a 22% decline (US stocks, daily). Win rates for QML circulating online are unverified.

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