I lost money on the same candle twice before I understood what it was doing. It was a Tuesday in June, GBPUSD on the hourly chart: price pushed to a fresh high, took out the previous swing high by a few pips, and then collapsed. I shorted the collapse, got stopped out when price spiked back above that high, and then watched it fall two hundred pips without me. The market had not changed its mind twice. It had run the same trick twice: first it hunted the stops above the old high, then it hunted mine. The pattern behind both hunts has an ugly name and a clean structure, and once I learned to wait for its final step instead of jumping at its first one, it became the reversal setup I trust most on a funded account.
In this article
- What the Quasimodo pattern actually is
- The five rules that make a Quasimodo valid
- Entry, stop loss and targets
- Sizing a Quasimodo trade under a daily loss limit
- Worked example: a bearish Quasimodo on GBPUSD
- I tested it: ten Quasimodo trades, one log
- When the Quasimodo fails
- The session routine that keeps it honest
- Frequently asked questions
The pattern is called Quasimodo, after the hunched shape its swings make on a chart, and it is a reversal pattern with a built-in lie detector. Most reversal patterns ask you to believe that a trend has ended. Quasimodo asks you to wait until the market proves it, by sweeping a level, breaking structure, and then coming back to the scene of the crime. This article explains the pattern piece by piece, the five rules that keep it valid, the entry, stop and target rules, the sizing maths under a daily loss limit, and a ten-trade log from my own journal. If you keep getting stopped out of reversals that then work without you, this is probably the step you are skipping.
What the Quasimodo pattern actually is
Picture an uptrend running out of steam. Price makes a swing high, pulls back, and then pushes to a new high above it. That second high is the head of the pattern, and it is the trap: it takes out the stops of everyone who shorted the first high. Then price falls, and this time it does not stop at the old pullback low. It closes below it. That break of structure is the market admitting the higher high was a fake-out. Finally, price drifts back up to the level of the first swing high. That level is the Quasimodo line, and it is where the trade lives: sellers who missed the first fall get a second chance to sell, at a price the market has already rejected twice.
| Part of the pattern | What price does | What it means |
|---|---|---|
| Left shoulder | First swing high in the uptrend | The level breakout buyers will later defend |
| Head | A higher high that sweeps above the shoulder | Stops above the shoulder are hunted; late buyers trapped |
| The break | Price closes below the pullback low between shoulder and head | Structure flips; the higher high is confirmed as a fake-out |
| The Quasimodo line | The level of the left shoulder itself | The retest zone; the entry level for the reversal trade |
| Confirmation | Price rejects the Quasimodo line and turns down again | The reversal is real; the short is now justified |
The bullish version is the mirror image: a lower low that sweeps below a left shoulder low, a break above the pullback high between them, and a retest of the shoulder low from above. What matters in both versions is the order. A head and shoulders breaks a neckline; Quasimodo breaks the structure between shoulder and head and then retests the shoulder itself. That difference is the whole edge: the retest gives you an entry with a stop beyond a level the market has already failed to hold. My head and shoulders guide covers the neckline version; the two are not interchangeable. For a second walkthrough of the same structure, the Traders With Edge Quasimodo guide covers the pattern from a prop-firm-challenge angle.
The five rules that make a Quasimodo valid
Most Quasimodo losses come from trading shapes that look like the pattern but miss one condition. I keep five rules above my monitor, and a setup that fails any one of them goes in the journal as a skipped trade.
| Rule | The test | Why it exists |
|---|---|---|
| 1. An obvious sweep | The head must clearly exceed the shoulder, not just poke it by a pip or two | A shallow poke is noise; a real sweep takes out real stops |
| 2. A real break of structure | A full candle close beyond the pullback low (bearish) or high (bullish) | A wick through the level is not a break; a close is commitment |
| 3. A clean retest | Price returns to the Quasimodo line and rejects it, without closing back beyond it | The retest is the entry; chasing the break is how you buy the trap |
| 4. Room to the next level | At least twice the stop distance to the next support or resistance | A 50-pip stop needs 100 pips of open road, or the trade cannot pay |
| 5. No news in the window | No red-folder news on the pair within 30 minutes either side of entry | News turns a technical retest into a coin flip, and funded accounts punish coin flips |
Rule 3 is the one most traders break, because waiting feels like losing. In June I broke it twice on the same GBPUSD setup: once by shorting the break itself, once by shorting the first touch of the shoulder line before any rejection candle had closed. Both times price spiked through my stop and then reversed without me. The pattern did not fail. My patience did. The retest is not a suggestion; it is the trade. If price never comes back to the line, there is no trade, and a missed trade costs nothing. A forced trade can cost a fifth of your daily stop.
Entry, stop loss and targets
Once all five rules pass, the mechanics are simple and they do not change from trade to trade. I use two entries. The confirmation entry, my default, waits for a rejection candle to close away from the Quasimodo line, and enters on that close. The limit entry, used only when the retest is slow and orderly, places a limit order at the line itself with the stop already set. The limit entry gets a better price but risks entering a retest that keeps going; the confirmation entry pays a few pips for proof. On a funded account, proof is cheaper than a stop-out.
The stop sits beyond the head, plus a small buffer for spread and wick noise: above the head high for a bearish setup, below the head low for a bullish one. The buffer is 10 pips on majors and 15 on gold, and it is not negotiable: a stop placed exactly on the head gets hunted by the same players who made the head. Targets are structural. The first target is the broken structure level, the pullback low or high that price broke to confirm the pattern. The second target is the next clear swing level beyond it. If the first target is closer than the stop distance, the trade fails rule 4 and I skip it.
Sizing a Quasimodo trade under a daily loss limit
The pattern gives you the stop distance, and the daily limit gives you the size. My account is 100,000 dollars, the firm daily loss limit is 5,000 dollars, and my personal daily stop is 2,500 dollars, half the firm limit. My base risk is 500 dollars per trade: 0.5 percent of the account, one fifth of my personal daily stop. Five base-risk losses end my day. That is the point: the limit, not my mood, decides when I stop.
| Sizing step | Formula | Worked number |
|---|---|---|
| Stop distance | Head plus buffer minus entry | 50 pips |
| Dollars per pip per lot | Standard lot on GBPUSD | 10 dollars |
| Lots from base risk | 500 dollars / (50 pips x 10 dollars) | 1.00 lot |
| First target value | 50 pips x 10 dollars x 1.00 lot | +500 dollars (1R) |
| Second target value | 100 pips x 10 dollars x 1.00 lot | +1,000 dollars (2R) |
| Share of personal daily stop | 500 / 2,500 dollars | 20 percent |
Two sizing habits matter more than the arithmetic. First, the stop sets the size, never the other way round: if the head is 80 pips away, the same 500 dollars buys 0.63 lots, and I take 0.63 lots, not 1.00 lot with a tighter stop. Second, I run the numbers through the max daily drawdown calculator before the session, not after a loss, when the temptation is to size up. The liquidity sweep guide covers the same sizing logic from the sweep side; the maths is identical, only the entry trigger differs.
Worked example: a bearish Quasimodo on GBPUSD
Here is the June setup, rebuilt with the rules I now follow. GBPUSD on the hourly chart, in an uptrend that had already run for three days. The left shoulder printed at 1.2720, price pulled back to 1.2680, then pushed to a head at 1.2760, sweeping the stops above the shoulder by 40 pips. The next two candles closed below 1.2680, confirming the break of structure. Price then drifted back to 1.2720, the Quasimodo line, and printed a bearish rejection candle. That close was my confirmation entry, effectively filled at 1.2715.
| Level | Price | Role |
|---|---|---|
| Left shoulder (Quasimodo line) | 1.2720 | Retest and entry zone |
| Pullback low | 1.2680 | Break level; first target area |
| Head high | 1.2760 | Sweep extreme; stop reference |
| Entry (rejection close) | 1.2715 | Short on confirmation |
| Stop (head + 10-pip buffer) | 1.2770 | 55 pips of risk |
| First target | 1.2660 | 55 pips, 1R, back below the broken low |
| Second target | 1.2605 | 110 pips, 2R, next swing low |
Notice the stop maths changed slightly from the sizing table above, because the real fill was 5 pips below the line: 55 pips of risk instead of 50. At 500 dollars of risk that is 0.91 lots, not 1.00 lot, and the targets pay 500 dollars at 1R and 1,000 dollars at 2R, because the sizing keeps the risk fixed while the pip distance changes. I show the adjustment deliberately: funded traders lose more money to rounded-up size than to bad patterns. Price hit the first target the same evening, I took half off and moved the stop to breakeven, and the second target filled the next morning. The trade worked because every step had already happened before I entered: the sweep, the break, and the rejection.
I tested it: ten Quasimodo trades, one log
For three weeks I traded only Quasimodo setups that passed all five rules, on the hourly chart, in the London and New York sessions, with the same 500-dollar base risk. Everything else went in the journal as a skipped trade. The log below is the full set, wins and losses together, because a log that hides its losers is marketing, not testing. Two of the four losses were retests that closed back beyond the head, one was a news spike I should have filtered, and one was a shallow sweep I accepted against rule 1.
| # | Pair and direction | Outcome | Result |
|---|---|---|---|
| 1 | GBPUSD short | Hit second target | +1.6R (+800 dollars) |
| 2 | EURUSD long | Retest failed, stop hit | -1.0R (-500 dollars) |
| 3 | GBPUSD short | Hit second target | +2.0R (+1,000 dollars) |
| 4 | XAUUSD long | News spike through stop | -1.0R (-500 dollars) |
| 5 | EURUSD short | Hit first target only | +1.0R (+500 dollars) |
| 6 | GBPUSD long | Hit second target | +2.8R (+1,400 dollars) |
| 7 | EURUSD long | Retest failed, stop hit | -1.0R (-500 dollars) |
| 8 | GBPUSD short | Hit second target | +1.6R (+800 dollars) |
| 9 | XAUUSD short | Shallow sweep, stop hit | -1.0R (-500 dollars) |
| 10 | EURUSD short | Hit second target | +2.0R (+1,000 dollars) |
The totals: six wins, four losses, a 60 percent win rate, an average winner of 1.83R, and a net of +7.0R, or +3,500 dollars at 500 dollars of risk. That is the pattern working as designed: its winners run further than its losers because the stop sits beyond a level the market already failed to hold. Three of the four losses broke a rule I already had, and the fourth was a fair loss on a valid setup. On a funded account, that split matters more than the win rate: rule-breaking losses are avoidable, and avoidable losses breach daily limits.
When the Quasimodo fails
The pattern has five common failure modes, and recognising them before entry is worth more than any new indicator. The most expensive is the retest that never rejects: price drifts back to the line, hovers, and then closes beyond the head, turning the reversal back into a continuation. That is why the confirmation entry exists. The second is the shallow sweep, where the head barely exceeds the shoulder and no real stops were hunted. The third is the range-bound Quasimodo, which prints constantly inside sideways markets and means nothing, because there is no trend to reverse. The fourth is the news-driven fake break. The fifth is the higher-timeframe conflict, where a beautiful hourly Quasimodo points directly into a daily trend that has no intention of stopping.
| Failure mode | What you see | The filter |
|---|---|---|
| Retest never rejects | Price closes back beyond the head | Confirmation entry only; no entry on touch alone |
| Shallow sweep | Head exceeds shoulder by less than 10 pips | Require a clear sweep, or skip the setup |
| Range-bound pattern | Repeated QM shapes inside a sideways box | Trade only after a real trend leg, not inside a range |
| News fake break | Break candle on a data release | 30-minute news blackout either side of entry |
| Higher-timeframe conflict | Hourly QM against a strong daily trend | Skip, or demand a daily-level confluence first |
The trend pullback guide covers the other side of this coin: when the higher timeframe trend is intact, a pullback entry beats a reversal entry almost every time. Quasimodo is for the moments when the trend itself is exhausted, and telling those two situations apart is the skill.
The session routine that keeps it honest
Quasimodo setups print all day, but they do not pay all day. The sweep needs real liquidity to hunt, and the retest needs real participation to reject. I scan twice a day and ignore everything outside those windows.
| Session window | What to do | Why |
|---|---|---|
| Asian session (avoid) | No new entries; manage open trades only | Thin liquidity makes sweeps shallow and retests unreliable |
| London open, 08:00-10:00 GMT | Watch for the sweep and the break | Real volume hunts real stops; the pattern forms cleanly |
| London afternoon, 13:00-15:00 GMT | Take retest entries that confirm | Retests into the New York overlap reject hardest |
| New York, 15:00-17:00 GMT | Manage targets; no fresh scans | Late entries chase moves that already ran |
| News windows | Flat 30 minutes either side | Data breaks structure for reasons no pattern can price |
That routine is also a consistency guard. A funded account punishes erratic behaviour, overtrading, and revenge entries after a missed setup, not just big losses. Two scans a day, five rules, one base risk: that is what keeps a good pattern from becoming an overtrading habit. My give-back-profits article covers what happens when the routine slips.
Frequently asked questions
Is Quasimodo just a head and shoulders pattern?
No. A head and shoulders breaks a neckline drawn across the two pullback lows. A Quasimodo breaks the structure between the shoulder and the head, then retests the shoulder level itself. The Quasimodo stop sits beyond the head, and the trade is invalid the moment price closes back beyond it.
What if price never retests the Quasimodo line?
Then there is no trade. A break that runs without a retest is still a valid reversal, but it is not a Quasimodo entry, and chasing it means entering without the stop structure the pattern provides. Missed trades cost nothing.
Can I trade Quasimodo on the 15-minute chart?
You can, but the failure rate climbs fast. Lower timeframes print more sweeps, shallower breaks and weaker retests, and the spread eats a bigger share of a smaller stop. I trade it on H1 with H4 context, and not below M30 on a funded account.
How is Quasimodo different from a turtle soup or a liquidity sweep?
A turtle soup trades a single false breakout of a prior high or low. A liquidity sweep trades the hunt itself. Quasimodo trades the full sequence: sweep, break of structure, and retest of the shoulder. It is slower and rarer than either, which is why its winners run further.
What win rate should I expect?
My three-week log shows 60 percent, but ten trades is a tiny sample and you should not anchor on it. Judge the pattern on its expectancy, not its win rate: with a 1.83R average winner and 1R losers, it stays profitable down to roughly a 40 percent win rate, and that cushion is what matters.
Does Quasimodo work on gold and indices?
Yes, with wider buffers. Gold sweeps deeper and retests harder, so I use a 15-pip buffer instead of 10 and halve the size when the stop exceeds 80 pips. Indices behave similarly in liquid hours; avoid thin overnight sessions, where the pattern prints noise.
Risk disclaimer: this article describes a trading strategy for educational purposes only and is not financial advice. Trading leveraged products carries a high risk of loss, and past results, including the ten-trade log above, do not guarantee future performance. Always risk only money you can afford to lose and respect your firm’s rules.




