Hammer candlestick pattern

What a hammer is, down to the threshold

A hammer is a single candle that appears after a decline, and whose shape says one thing only: sellers pushed price far down during the period, and kept none of that ground by the close.

Three conditions, and that is all. One: the lower shadow is at least twice the height of the body. Two: the upper shadow is absent or negligible. Three: the candle arrives in a downward price trend, not anywhere.

The number in the first point is the only one that is debated, and that is exactly why you must pick one and stick to it. Serious sources require a shadow of two to three times the body. Take two, write it down, and apply it to every count. A pattern whose threshold moves from one reading to the next cannot enter any statistic, and you will never know whether it works for you.

Body colour is not part of the definition. A hammer with a bullish body is regarded slightly more highly, and we will see below that the measurements agree, but a hammer with a bearish body is still a hammer.

What the candle tells you, and what it hides

The mechanism is simple and worth more than the name of the pattern. During the period, selling wins: price falls, sometimes violently. Then something absorbs that selling and sends it back. By the close, price has returned near its starting point. The lower shadow is the trace of what was attempted and refused.

It is a rejection of a level, and that is the only information the candle contains. It does not say who bought, how much, or why. It also does not say in what order things happened inside the period: a hammer may have plunged early then climbed back slowly, or stayed high then plunged and bounced in the final seconds. Those two stories produce the same candle and are not worth the same.

Hence the rule the rest of this lesson applies: context carries the signal, the pattern only timestamps it. A hammer sitting on a level the market has already defended is an event. The same hammer in the middle of nowhere is a quoting accident.

How to identify it, in four checks

One. The trend before. Look at the previous five to ten candles. If they are not falling, this is not a hammer, it is a candle with a long lower shadow. The name changes because the position changes.

Two. The shadow to body ratio. Measure it, do not eyeball it. Lower shadow divided by body height, at least two. Most platforms show the four prices on hover: the subtraction takes three seconds and removes half the false positives.

Three. The upper shadow. It must be close to nothing. As soon as it reaches the height of the body, you no longer have a clean rejection, you have an indecision candle, and its message is not the same.

Four. The close. A live candle is not a pattern. Until the period closes, the body keeps moving and the extremes can only widen. A good number of hammers disappear in their final two minutes.

What it is actually worth, with the numbers

Here is the part almost nobody publishes, because it is disappointing. The most complete publicly available measurements on candlestick patterns are those of Thomas Bulkowski, who ranked more than a hundred candle types across thousands of cases. His results for the hammer, quoted as they stand.

Theoretical performance: bullish reversal. Tested performance: it acts as a bullish reversal 60% of the time. In other words, four times out of ten the decline simply continues.

Overall performance rank: 65 out of 103 candle types ranked. The hammer is therefore not in the top third, it is in the lower half. The best known pattern in the catalogue is an average pattern.

And one figure worth pausing on: its best average move over ten days is recorded at -4.12%, in a bear market with a downward breakout. Its best ten-day performance is therefore obtained when price keeps falling, the exact opposite of what the pattern is supposed to announce. That is not a contradiction, it is what happens when you measure instead of assert: a hammer that fails, fails properly.

Three observations from the same body of work, useful for filtering. Hammers near the yearly low perform best. Those near the yearly high often do act as reversals, but downward. And hammers with a white body, meaning bullish, give the best results.

Source: Thomas Bulkowski's research pages on thepatternsite.com, consulted in September 2026. These figures cover US stocks on daily data. Nothing guarantees they carry over to forex on a five-minute chart, and that is precisely the subject of the last section.

The three patterns mistaken for it

The hanging man. Exactly the same candle, but after a rise rather than after a decline. The shape is identical, the reading inverts: it is no longer a rejection of the low, it is a warning at the high. If you only look at the candle, you cannot tell them apart. It is the best demonstration that position matters more than drawing.

The inverted hammer. Body at the bottom, long upper shadow, after a decline. It also reads as a bullish reversal signal, but it tells the opposite story: buyers pushed and were pushed back. It demands clearer confirmation.

The shooting star. Same drawing as the inverted hammer, but after a rise. Bearish signal.

The fourth case, often forgotten: the dragonfly doji. Long lower shadow, but almost no body at all, open and close together. Many hammer definitions exclude it because the shadow to body ratio becomes infinite. Decide once and for all whether you count it, and write your decision down.

How to trade it, if you trade it

The most defensible entry is not on the hammer, it is after it. You wait for the next candle to trade above the high of the hammer. It costs a few points of entry and it removes part of the four failures out of ten.

The stop goes below the tip of the shadow, not below the body. That is the level the market tested and refused: if it goes back there, the reading is denied and there is no reason left to stay. Mind the mechanical consequence: the longer the shadow, the further the stop, so the smaller the position at constant risk. A spectacular hammer is an expensive hammer.

The target does not come from the pattern. The hammer carries no price objective, unlike continuation patterns which have one by construction. Aim at the first level the market has genuinely defended above, not at an arbitrary multiple.

And the filter that makes the most difference, quoted above: a hammer inside a zone that has already been defended, at the bottom of a move, is worth several times an isolated hammer. If you were to keep one sorting criterion, that would be the one.

What it gives in your own journal, and that is the only number that matters

The 60% quoted above comes from US stocks on daily data. You may be trading gold on five minutes, or an index at the US open. Nothing says the figure follows you, and nobody can tell you in your place.

The good news is that it is measurable in one evening. Tag the entries you take on a hammer with a dedicated label, then compare them to the rest of your journal on three metrics, and three only.

The win rate, obviously, but it is not enough and it often misleads: a high rate on tiny gains pays nothing. Maximum adverse excursion, meaning how far price went against you before coming back: it tells you whether your stop under the shadow sits in the right place or whether you are being taken out just before the reading proves itself. And maximum favourable excursion, how far price went in your favour: it tells you whether you exit your winning hammers too early.

Thirty tagged entries are enough to see a trend, a hundred to be sure. It is long, and it is the only way to know. Tradoshi computes these three measurements automatically on every imported trade and lets you filter the journal on a label: the reading then takes seconds instead of an evening in a spreadsheet.

The outcome of that exercise is often the same, and it is useful: the pattern carries almost nothing on its own, and a great deal once crossed with where it appears. That is also what the numbers in the previous section say.

Frequently asked questions

Is a hammer bullish or bearish? Bullish by definition, which is why it is often called a reversal hammer, but measured at 60% only, and its best ten-day move is recorded to the downside. Treat it as an indication, never as a signal.

Does body colour change anything? It does not change the definition, but the measurements favour bullish bodies. If you have to filter, filter there.

Which timeframe? The longer the period, the more participants the candle aggregates and the less noisy it is. A daily hammer carries more weight than a one-minute hammer, which is often just an order going through. And remember that a daily hammer is a whole session seen on a five-minute chart.

Should you wait for confirmation? Yes, if you care about the win rate. No, if you accept a lower rate in exchange for a better entry price. Both are defensible, provided you measure the one you chose.

Are a hammer and a dragonfly doji the same thing? No, and the difference is the body. The dragonfly doji has almost none. Many definitions exclude it from the hammer. Pick your convention and keep it.