It was a Monday in early October, and EURUSD had been sliding for two days. A big red candle printed into 1.0800, then a small green candle appeared — tucked completely inside the red candle’s body. A textbook bullish harami. I bought the small candle’s close at 1.0812, feeling clever. Forty minutes later the next candle swept straight through the big red candle’s low, my stop went at 1.0790, and I was $380 poorer with a lesson I now repeat to every trader I mentor: a harami is a stall, not a trigger.
In this article
The next week the same pattern appeared on GBPUSD. This time I waited. The small green candle’s high broke, my buy stop filled at 1.2644, my stop sat below the big red candle’s low — and price climbed for +1.9R without ever looking back. Same pattern, same pair family, same me. The only difference was confirmation. That is what this article is about: the harami is one of the oldest two-candle reversals in the book, but on a funded account it is only tradeable through strict rules. Here is my full 6-rule plan — the bullish and bearish setups, the confirmation entry, the stop, the target, the session filter — plus the 0.5%-risk sizing maths that keep every trade inside a prop firm daily loss limit.
What the harami pattern actually is
Harami is Japanese for “pregnant,” and the picture says it all: a large mother candle followed by a small candle tucked entirely inside the mother’s body. It is the conceptual mirror image of the engulfing pattern — where the second body swallows the first, the harami’s second body hides inside it. Where engulfing screams reversal, harami whispers hesitation: the dominant side has stopped pushing, and the other side is probing. That hesitation is real information, but it is not a buy signal on its own. (Babypips’ harami definition page lays out the same textbook criteria I cross-checked mine against: prior trend, large first body, small second body fully contained.)
There are two variants, and you must trade them as mirror images — not as one vague “reversal candle”:
| Variant | What I need to see | What it means |
|---|---|---|
| Bullish harami | Clear downtrend, then a long red mother candle; the next candle is small and green, its body fully inside the red body’s range | Sellers exhausted themselves on the mother candle; buyers are probing — a possible bottom |
| Bearish harami | Clear uptrend, then a long green mother candle; the next candle is small and red, its body fully inside the green body’s range | Buyers ran out of steam; sellers are probing — a possible top |
| Harami cross (either direction) | Same setup, but the small candle is a doji — open roughly equals close | Maximum indecision: the probe found zero conviction either way, so the reversal odds improve |
Two details most guides gloss over. First, containment is judged on bodies, not wicks — the small candle’s wicks may poke outside the mother’s range, but its open and close must both sit inside the mother’s body. Second, the mother candle must be genuinely large relative to recent candles. A “harami” where the first candle is average-sized is just two candles in a range — it carries no exhaustion signal and fails constantly. My rule of thumb: the mother’s body should be at least 1.5 times the average body of the previous ten candles.
The 6-rule harami plan
These are the exact rules I trade today. Rule 4 — the confirmation entry — is the one that separates my current results from that -$380 Monday.
| Rule | The rule | My notes |
|---|---|---|
| 1. Trend context | A clear trend first: 8–12 candles of lower highs and lower lows (bullish harami) or higher highs and higher lows (bearish harami) | No trend, no harami. In chop the pattern is noise — I skip it even when it looks perfect |
| 2. Location | The pattern must sit at support or resistance, a supply or demand zone, or a round number | Confluence is what turns a 50/50 stall into a tradeable reversal |
| 3. Real exhaustion | Mother-candle body at least 1.5× the 10-candle average body; the small candle’s body fully inside the mother’s body | This one filter removes most of the fake haramis I used to lose money on |
| 4. Confirmation entry | Buy stop 2–3 pips above the small candle’s high (bullish) or sell stop 2–3 pips below the small candle’s low (bearish) | The stop order is everything: no follow-through, no fill, no loss — the cheapest outcome in trading |
| 5. Stop loss | Beyond the mother candle’s extreme — below the mother low for bullish, above the mother high for bearish — plus a 5–10 pip buffer | Objective invalidation: if price takes out the mother’s extreme, the exhaustion thesis is dead |
| 6. Session filter | London and New York sessions only; nothing within 30 minutes of red-folder news | Asian-session haramis stall into nothing; news spikes invalidate every candle signal (see the news trading guide) |
The 0.5%-risk sizing maths
Harami stops are honest but not tiny — the mother candle’s extreme plus buffer usually means 25–40 pips on H1 majors. That is fine, because the rule that matters is not stop size but risk size: 0.5% of the account per harami trade, fixed, no exceptions. Two haramis may be open at once (1% total heat), but the third setup of the day gets watched, not traded.
Position-size formula
Lots = Risk ($) ÷ (Stop distance in pips × Pip value per standard lot)
| Step | My worked EURUSD example |
|---|---|
| Account and risk | $100,000 account × 0.5% = $500 maximum loss on the trade |
| Entry and stop | Buy stop at 1.0845 (above the small candle’s high); stop at 1.0813 (below the mother low + 8-pip buffer) = 32 pips |
| Pip value | $10 per pip per standard lot on EURUSD |
| Position size | $500 ÷ (32 × $10) = 1.56 — I trade 1.55 standard lots |
| Target | Minimum 1:1.5R = 48 pips to 1.0893; I bank half at 1.5R and trail the rest below rising swing lows |
| Outcomes | Win: +$750 (+0.75% on the account). Loss: −$500 (−0.5%, just 10% of a 5% daily limit) |
I check every setup against the max daily drawdown calculator before placing the order — the planned $500 loss has to fit inside the day’s remaining allowance, especially if an earlier trade already scratched the account. And I keep a personal daily stop at half the firm’s limit (the personal daily stop guide explains why), which means four consecutive harami losers end my day at −2% instead of letting frustration end my account.
When I trade it: sessions, pairs and the news blackout
A harami is a handoff pattern — one side stalls, the other probes. That handoff needs genuine order flow behind it, which is why rule 6 exists. Here is my routine:
| Window | How I treat harami patterns |
|---|---|
| London open, 07:00–10:00 GMT | Prime time — deep liquidity, and mother candles print with real exhaustion behind them |
| New York morning, 12:00–15:00 GMT | Second best; US data can supercharge the reversal, which is exactly why the news blackout applies |
| Asian session | Skipped — thin ranges print harami-looking candles that mean nothing |
| 30 minutes around red news | Never — a CPI print will slice through a mother candle’s extreme like it is not there |
Pairs and timeframes: EURUSD and GBPUSD on H1 and H4, where mother candles have real size and stops sit at honest levels. On XAUUSD the pattern works but the contract maths change — I size gold with the XAUUSD lot size calculator instead of improvising. M15 and below print haramis constantly, but spread and wick noise eat the edge; those belong to someone else’s account.
I tested it: 10 haramis on a $100k demo
Over three weeks I logged every harami that passed all six rules on a $100,000 demo, risking $500 (0.5%) per trade. Two more patterns looked tempting but failed rule 3 (undersized mother candles); I logged them as no-trades, and both failed — which is exactly why the exhaustion filter exists. Here is the trade-by-trade record:
| # | Pair and variant | Entry | Result | Notes |
|---|---|---|---|---|
| 1 | EURUSD bullish | 1.0842 | +1.8R | Textbook: support + round number, half at 1.5R, runner trailed |
| 2 | GBPUSD bearish | 1.2690 | +2.2R | Clean resistance rejection, full target hit |
| 3 | EURUSD bullish | 1.0788 | −1.0R | London-open wick swept the mother low by 4 pips |
| 4 | GBPUSD bullish | 1.2615 | +1.5R | Half banked at 1.5R, runner stopped at breakeven |
| 5 | EURUSD bearish | 1.0922 | −1.0R | My error: entered 22 minutes after CPI — rule 6 now has a name and a scar |
| 6 | GBPUSD bullish | 1.2580 | +2.6R | Best of the set: harami cross at demand, ran to resistance |
| 7 | EURUSD bearish | 1.0955 | +1.2R | Exited early at the first sign of trouble — left money on the table, kept the discipline |
| 8 | GBPUSD bearish | 1.2745 | −1.0R | Range day: mother extreme taken out by 3 pips, then reversed without me |
| 9 | EURUSD bullish | 1.0820 | +1.9R | Support confluence, smooth 1.9R ride |
| 10 | GBPUSD bullish | 1.2640 | +1.4R | Closed into the weekend per my no-weekend-hold rule |
The totals: 7 wins, 3 losses — a 70% win rate — with an average winner of +1.80R against a fixed −1.0R loss. Net: +9.6R, roughly +$4,800 at $500 risk per trade. Before you get excited, run the same numbers through the risk of ruin calculator: 70% wins at a 1.8:1 payoff is a genuinely fundable edge, but three straight −1R losers still happen, and the 0.5% risk cap is what keeps them boring instead of fatal.
When the harami fails: five ways it lies
Every pattern has failure modes; the harami’s are specific enough to name. Knowing them is half the strategy.
1. No trend, no signal. A harami in the middle of a range is two random candles. I only trade it after a genuine impulse leg — 8 to 12 trending candles minimum.
2. A weak mother. If the first candle is average-sized, there was no exhaustion and no handoff. The 1.5× body filter exists because of this exact failure.
3. A barely-inside second body. When the small candle’s body nearly fills the mother’s, it is not a probe — it is continuation. I want the small body clearly contained, ideally under 40% of the mother’s body size.
4. Entering before confirmation. The number-one killer, and my original sin. Buying the small candle’s close feels proactive; in practice it converts a 70%-win-rate plan into a coin flip with worse entries.
5. News and dead sessions. A harami that forms ten minutes before NFP is not a setup, it is a trap. Rule 6 is a rule, not a suggestion.
How the harami fits prop firm rules
A strategy is only as good as its fit with the account’s rulebook. Here is how the 6-rule plan maps onto the three rules that end most funded accounts:
| Firm rule | How the harami plan respects it |
|---|---|
| Daily loss limit (typically 5%) | Fixed 0.5% risk per trade, max two concurrent positions — the worst normal day is −1%, and my personal daily stop ends the session at −2.5% before the firm’s line is ever in sight |
| Trailing drawdown | Stops sit at the mother extreme from entry — objective and unmoved. I only trail after 1R of profit, and never inside the mother candle’s range, so the trailing line can never catch a stop I did not choose |
| Consistency rule | Steady 1.2–2.6R winners with no monster days keep the best-day percentage healthy — the consistency rule guide shows why one 6% day can fail a challenge that ten 1.5% days would pass |
The pin bar reversal remains my highest-conviction reversal trigger, but the harami earns its place beside it: it fires earlier, at the first sign of exhaustion, which means better entries when the confirmation follows through — and a free no-fill when it does not.
Harami pattern questions I get asked
What is a harami pattern in simple terms?
A two-candle pattern where a large “mother” candle is followed by a small candle whose body sits entirely inside the mother’s body. It signals that the dominant side has stalled — buyers or sellers are probing for a reversal.
What is the difference between a harami and an engulfing pattern?
They are mirror images. In an engulfing pattern the second candle’s body swallows the first — aggressive reversal. In a harami the second body hides inside the first — hesitation. Engulfing needs no confirmation in most plans; a harami always does, which is why my entry waits for the break of the small candle’s extreme.
What is a harami cross?
A harami where the second candle is a doji (open roughly equals close). It shows even deeper indecision than a regular harami, and in my log the cross variants have slightly better follow-through — trade 6 above was one.
How reliable is the harami pattern?
On its own, mediocre — textbooks quote roughly 50–55% reversal follow-through. Through my 6-rule filter (trend, location, exhaustion, confirmation, mother-extreme stop, session), my 10-trade sample ran at 70% with a 1.8:1 average payoff. Small sample, honest caveat — but the structure of the edge (confirmation plus fixed fractional risk) is what makes it repeatable, not the win rate.
Which timeframe is best for trading the harami pattern?
H1 and H4. The pattern needs a mother candle with genuine size, which lower timeframes rarely provide cleanly. On M15 the spread-to-stop ratio alone destroys the maths.
Can I trade harami patterns on a funded account?
Yes — it is a standard price-action pattern, not a prohibited strategy. The funded-account question is never “is the pattern allowed” but “does the plan respect the drawdown rules,” and the 0.5%-risk cap plus mother-extreme stops are built for exactly that.
Risk disclaimer: trading involves substantial risk of loss and is not suitable for every investor. Past performance — mine or anyone’s — does not guarantee future results. Never trade a funded account with money you cannot afford to lose, and always follow your firm’s rulebook over any article, including this one.





5 comments
[…] candle’s body to swallow the first; it is a stronger momentum signal but a wider stop. A harami needs the second candle to sit inside the first; it is a hesitation signal that usually needs more […]
[…] because my site already covers the neighbours of this pattern, and confusing them costs money. My harami guide trades the two-candle signal itself, which is a faster but far noisier entry — the Thursday loss […]
[…] because my site already covers the neighbours of this pattern, and confusing them costs money. My harami guide trades the two-candle signal itself, which is a faster but far noisier entry — the Tuesday loss […]
[…] candle, while the homing pigeon’s second candle is black. I cover the harami separately in my harami pattern strategy guide, and the difference is worth memorising: white inside black is buyers stepping in, black inside […]
[…] and three-candle reversals, and confusing them is the easiest way to trade the wrong rules. The Harami needs the second candle’s whole body inside the first, which a counterattack line does not […]