Double top and double bottom: the pattern and its neckline

In 20 seconds

Shape

Two peaks of similar height, separated by a clear trough. The double bottom: two similar troughs, separated by a peak.

Where

After an uptrend for the double top, after a downtrend for the double bottom.

What it tells you

The price hits the same ceiling twice, then the floor held between the two peaks finally gives way.

Confirmation

A close below the neckline. Before that, there are only two peaks.

Invalidation

The price moves back above both peaks.

The double top is one of the simplest patterns in chart analysis. Its shape earns it the nickname "M", and the double bottom is called a "W". It tells the story of an uptrend that hits the same ceiling twice.

Like the head and shoulders, it jumps out at you in hindsight. On the right edge of the chart, when you have to decide, nothing yet tells you whether the second peak really is one. Here you will see how to recognize it, where it is confirmed, and why I publish no figures on it.

This guide describes a training method, not investment advice. Trading carries a high risk of capital loss.

Recognizing it

1. The first peak, at the end of an uptrend.
2. The trough between the two peaks.
3. The second peak, at a height close to the first.
4. The neckline: the horizontal line drawn through the trough.
5. The break: the first close below the neckline.

The two peaks don't need to be identical. Compare the gap between them to the height of the pattern, from the trough to the peaks: a small gap relative to that height keeps the pattern intact, while a second peak that closes clearly above the first one cancels it. The uptrend has then resumed.

The trough matters as much as the peaks. Two peaks separated by three hesitant candles are a resistance tested twice, not a pattern. You need a clear pullback between them.

The double bottom is the mirror image: two troughs at a similar height, at the end of a downtrend, separated by a peak. Its neckline runs through that peak, and it is confirmed by a close above it. Three peaks at the same height make a triple top, which you read the same way.

What it tells you

At the first peak, the uptrend stalls and the price pulls back to the trough. Buyers come back and push up to the old peak, where sellers are waiting: the price hits the ceiling a second time. The peak has become a resistance, and the trough a support.

As long as that support holds, the price moves back and forth between two boundaries, as in a range. It can bounce off the trough a second time. A break of the neckline only tells you the floor has given way.

What the textbooks take from it

• Entry: on the close below the neckline, or when the price comes back up to it from below, if it does.
• Stop: above the higher of the two peaks. That is where the idea of an exhausted uptrend becomes wrong.
• Theoretical target: the height of the pattern, from the peaks to the neckline, projected below the breakout point. It is a calculation convention, not a forecast, used to judge whether the trade is worth its risk.

Example: peaks at 110 and 109, neckline at 100. The theoretical target is 90. Selling on the break at 100, with a stop at 111, you risk 11 to aim for 10: less than 1 to 1. Waiting for a retest of the neckline and selling at 103, same stop, you risk 8 to aim for 13, or 1 to 1.6. The retest doesn't always come, and the trade can leave without you. The risk/reward ratio tells you which of the two plans holds up.

Reading it on the right edge of the chart

1. Check the setting. A double top reverses an uptrend. Without an uptrend before it, two peaks at the same height are just the top of a range.
2. Draw the neckline horizontally through the trough as soon as the second peak starts to pull back.
3. Wait for the close below the neckline. A wick that dips below and comes back confirms nothing.
4. Place the stop above the peaks, with a margin for wicks, then calculate the size with the 1% rule.
5. Write the plan in your trading journal before entering: entry, stop, target, ratio.

What 5 years of Bitcoin say

I tried to count the pattern on the same data as the other guides: BTC/USDT on 4-hour candles, from 2021 to 2025. A program detects peaks and troughs, then looks for two peaks separated by a trough, with a limited gap relative to the height of the pattern. It then waits for a close below the neckline within the 30 candles that follow the second peak. The same in reverse for the double bottom.

Two settings weigh on the result: how large a pullback has to be for a peak to count (3, 5 or 8%), and how much gap is tolerated between the two peaks (15 or 30% of the height). Depending on these choices, I found between 8 and 162 patterns per direction over five years. The margin of chance ranged from 7.6 to more than 30 points. And the result flipped direction: depending on the setting, the price fell after the break less often, or more often, than after any random candle.

A result that depends on the setting proves nothing: you could keep tuning until you got the number you wanted. So I publish no success rate for the double top or the double bottom. The doji and the RSI, whose definitions fit in one line, have their figures on this site.

Hold the rates you read elsewhere to the same standard. Ask which market, which period, which definition and how many cases.

Common mistakes

• Seeing the pattern before it exists. As long as the neckline holds, you have two peaks, not a double top.
• Entering on a wick below the neckline, without waiting for the close.
• Taking the theoretical target as a promise. The price is under no obligation to get there.
• Ignoring context. In the middle of a range, two peaks at the same height are the norm.
• Moving the stop below the peaks to improve the ratio. The stop goes where the idea becomes wrong, and the size adjusts afterward.

Practice deciding before you know

TradaZen doesn't label double tops. Training works on what the pattern demands: deciding on the right edge of the chart, while what comes next is hidden. During the decision, you can draw a horizontal line through the trough to set your neckline, or across the two peaks.

The "Range" unit teaches you to read a market moving back and forth between two boundaries, and "The breakout" unit to tell a real exit from a range from a false start. Those are the two stages of a double top. You choose Long, Short, or "I stay out", the app sets the Stop and the Target, then the market reveals what it did, on a simulated balance. To learn more, read the page on the trading simulator.

Frequently asked questions

Mehdi Azizi created TradaZen and published it on the App Store through the Devoria studio. The app's scenarios are real past charts, never made-up data.

TradaZen is a training tool: you play on past market data with a simulated balance. It is not a broker and not investment advice. Real trading carries a risk of capital loss.