Double Bottom Pattern: Definition, Structure and Trading Strategies

The double bottom pattern is one of the best-known chart patterns in technical analysis: two lows at about the same price, separated by a peak, that form after a downtrend. It resembles the letter W. Technical analysis reads it as a potential bullish reversal, confirmed only when price breaks above the peak between the two troughs, a level called the neckline.

This guide covers the definition, the structure, how to identify a double bottom on a price chart and how traders use it, with entry, stop-loss and target. It also answers the question that decides whether to trade it: what happened after the pattern. Our test bench detected 6,422 double bottoms on daily charts across stocks, crypto, forex, ETFs, indices and commodities, and compared each trade with the same trade on random dates.

What is a double bottom pattern?

A double bottom pattern is a bullish reversal chart pattern. Price falls to a low, rebounds to a peak, falls back to about the same low, then rallies and breaks above the peak. The two troughs show that sellers failed twice at the same level; the breakout shows that buyers took control.

Like other reversal chart patterns, it is read as a sign that a downtrend may be ending and an upward trend may begin. In the classic reading, it shows a shift in market sentiment: selling momentum is fading at a price where buyers stepped in twice. Traders look for it on stock, forex and crypto charts, and in financial markets in general, from day trading to long-term investing.

Robert Edwards and John Magee described double tops and double bottoms in Technical Analysis of Stock Trends (first edition 1948), and their warning still applies: many W shapes appear on a chart, and only a confirmed breakout turns one into a double bottom. Until then, it is just a market testing a support level.

Our bench uses a written rule so that a computer can apply it without judgment: a prior decline (on a textbook pattern, the close one pattern duration earlier sits at least 60% of the height above the peak), a first trough, a peak, then a second trough within 10% of the pattern height of the first (15% on the looser standard grade), confirmed by the first close above the peak. The first trough must be a true low: over one first-leg duration before the pattern starts, price never went more than 10% of the pattern height below it. Swing points are only counted once price has moved away from them by at least 2 times the average true range (ATR), and the same rule is run at 4 and 8 times the ATR to catch larger formations.

The chart above is not drawn by hand. It is the median shape of the double bottoms our bench detected on daily charts across all markets, with the second trough slightly above the first, and the breakout above the neckline on the right.

Anatomy of a double bottom pattern

A double bottom has three points and one line. Each one has a role in the reading of the pattern.

The first low

The first trough ends a decline. At that time nothing distinguishes it from any other low in a downtrend: sellers push price down, buyers absorb the selling and price bounces. Bearish market sentiment is still dominant.

The peak or neckline

The bounce stops at a peak, where sellers return. A horizontal line drawn through that peak is the neckline. It acts as resistance: price must break it to confirm the pattern. The distance from the neckline down to the troughs is the height of the pattern, used later to set a price target.

The second low

Price falls again and stops near the first trough. This second test is the core of the double bottom: sellers had a second chance to break the support and failed. The second trough does not need to match the first to the cent. On our median shape it sits slightly above it, which the classic reading takes as a sign that selling pressure is weaker.

How to identify a double bottom pattern on your charts

To identify a double bottom pattern, check four points in order: the decline before it, two distinct troughs at a similar level, a clear peak between them, and a close above that peak.

Visualizing the "W" shape

Look for a W on the price chart: down, up, down, up. The two troughs should be distinct, separated by a real rally, not two candles next to each other. A rule of thumb that matches our detection: the gap between the two troughs should be small compared with the height of the W (within 10% of it on a textbook pattern, 15% on the looser standard grade).

The importance of time frames

The pattern appears on every timeframe, from hourly charts to weekly charts. On our statistics, one double bottom shows up every 1,320 daily bars and every 1,528 hourly bars on average, so it is far less frequent than a doji, which shows up once every 7 daily bars. The median double bottom our bench found spanned 57 weeks on weekly charts and 76 hours on hourly charts.

Distinguishing between support levels and double bottoms

A second touch of a support level is not yet a double bottom. Price can bounce from the second trough, stall under the neckline and break down through the troughs instead. Our bench measures how often a W shape is completed: on daily charts, only 42.2% of the shapes found closed above the neckline within the deadline of the rule, half the duration of the pattern on a textbook shape and 0.75 of it on the standard grade; on hourly crypto charts, 33.1%. More than half of the candidates never become a confirmed double bottom pattern.

The role of volume in pattern confirmation

Classic technical analysis wants rising volume on the rally from the second trough and on the breakout, as evidence of buying pressure. We tested the second part. On daily charts, double bottoms whose breakout bar had volume above 1.5 times the 20-bar average reached the 2:1 target 20.9% of the time, against 21.2% for breakouts on normal volume, with expectancies of +0.17 R and +0.22 R. High volume did not help. This breakdown is descriptive (not corrected for multiple testing), but it is a reason not to treat volume as a required confirmation.

The Real examples section shows double bottoms from our bench with their actual outcome. Trade results are in R, the initial risk of the trade.

How to trade the double bottom pattern

The classic double bottom pattern trading strategy buys the breakout above the neckline, with a stop-loss below the troughs and a target based on the height of the pattern. Our bench simulates a version of this trade on every occurrence: entry at the close of the breakout bar, stop below the lower of the two troughs, target at twice the risk, exit after 100 bars if neither level is hit.

Entry strategies and signal confirmation

There are three ways to enter:

Setting your stop loss levels

Calculating price targets and profit taking

The measured-move target adds the height of the pattern to the neckline. With an entry near the neckline and a stop at the troughs, that target is close to 1 times the risk. We tested a farther target at twice the risk. On daily charts, trades that reached it took a median of 51 bars; stopped trades took a median of 34 bars. The median trade moved 0.89 R in its favor at best, which is about the distance to the measured-move target. Of the trades that hit their stop, 13.7% had first reached +1 R. Taking partial profit at the measured move, then moving the stop, is one way to handle that; we have not tested it.

Double bottom vs. double top: understanding the difference

The double top pattern is the mirror image: two peaks at the same level after a rise, an M shape, confirmed by a close below the low between them. It is a bearish reversal pattern, traded with a short position. Our bench applies the same rule to both, on inverted prices.

On daily charts, both patterns hit their target about as often as the same trade on random dates, and neither average result per trade beat its random baseline by a margin that passes our statistical test (+0.20 R against +0.19 R for the double bottom, −0.22 R against −0.25 R for the double top). The gap between them comes from the market, not from the shapes: the same trades placed on random dates show it too (22.2% of targets hit for the long trade, 12.3% for the short one), because the assets we test rose more often than they fell over their history.

Related bullish formations follow the same logic. The triple bottom adds a third test of the troughs (18.6% of 2:1 targets hit on daily charts, against 20.4% on random dates), and the inverse head and shoulders has a lower middle low (28.1% against 29.2%).

Key advantages of using the double bottom pattern

• It is easy to spot: a W with a clear neckline.
• It defines the trade: entry above the neckline, stop below the troughs, target from the height.
• It is drawn the same way on stocks, crypto, forex and indices, on every timeframe.
• The breakout filter removes more than half of the W shapes before any trade is taken.

Common pitfalls and disadvantages to watch for

Identifying false signals

False reversals are the main risk of any reversal pattern. A false signal is a breakout that does not hold: price closes above the neckline, then falls back into the pattern and through the troughs. In our statistics this is common. On daily charts the 2:1 target was hit 21.0% of the time; most other trades ended at the stop or at the 100-bar exit. In the four crypto cases described under Example 1 and Example 2 of this guide, two breakouts ended at the stop.

The risk of pattern failure

The larger the pattern, the farther the target. On daily charts, double bottoms built from the smallest swing points (2 ATR) reached the 2:1 target 29.6% of the time; those built from the largest swing points (8 ATR) only 2.5%, because the target was rarely reached within 100 bars. Taller patterns need more patience and a smaller position.

Common mistakes traders make with double bottoms

Trading before confirmation

Ignoring the broader market trend

A trend filter would favor a double bottom that forms above the 200-day moving average, inside a larger bullish trend. We measured it: on daily charts, the 2:1 target was hit 19.8% of the time above the 200-day average and 23.1% below it, with almost the same expectancy (+0.22 R against +0.17 R). The larger trend changed the result only a little.

Other mistakes: drawing the neckline after the fact to fit a breakout, accepting troughs that are far apart in price, and using a stop so tight that normal noise inside the pattern triggers it.

Advanced tips

Strong signals and weak signals

Not every W carries the same information. One way to rank potential setups is by their structure:

• Strong: a clear downtrend before the pattern, two distinct troughs, and a strong close well above resistance.
• Strong: the second trough forms on weaker bearish momentum, for example a higher RSI reading than at the first.
• Weak: troughs only a few bars apart, a neckline barely above the troughs, or a breakout bar that closes back near resistance.
• Weak: a pattern that forms against strong bearish market conditions, such as a falling market index.

These rules come from reading price action, not from our statistics: our bench counts every confirmed pattern the same way. Use them to filter potential trades, and record the result to see whether they help in your own trading.

Combining with technical indicators

You can add indicators like moving averages, the RSI or the MACD. A bullish divergence (the second trough lower or equal in price, but higher on the RSI) is the classic addition. We have not tested indicator filters: if you use one, test it on your own market. The moving average filter is the one we did measure, with the small difference reported above.

Analyzing market sentiment

The double bottom is a picture of crowd psychology: fear at the first trough, doubt during the bounce, and a refusal to sell at the same price on the second trough. News, positioning and the behavior of the broader market can confirm or contradict that picture. Treat sentiment as context; it is not part of the pattern.

Reliability and success rates

The double bottom pattern is not reliable on its own: on no market or timeframe we tested did it beat random dates of the same assets in a way that survives our correction for multiple testing. The trade on daily crypto charts comes closest (39.6% of 2:1 targets hit against 27.0% on random dates) and reads promising: it passes a simple test, not the correction. The highest raw row with enough cases, daily indices (73.0% of 126 double bottoms higher 10 bars later, against 57.5% of random bars), still reads No clear edge: its excess return over the rest of the market is +0.15%, with an interval that includes zero. On 4-hour and hourly crypto charts, the row reads Worse than chance: a rise 10 bars later was rarer than after random bars, and the excess return over the rest of the market stays below zero after our correction. The looser ugly double bottom, whose second trough sits 15% to 50% of the pattern height above the first, does worse: on daily charts across all markets it reads worse than chance, with an excess return over the rest of the market of −0.30%. Here are the numbers from our test bench, which applies the same rule to full price history and compares what it finds with random bars.

Read the table in pairs of columns. On daily charts, 54.7% of double bottoms were followed by a higher price 10 bars after the breakout. That sounds bullish until you see that 53.4% of random days on the same assets were too. The excess return over the rest of the market after 10 days is −0.15%, with a 95% confidence interval from −0.40% to +0.09%: zero is inside the interval, so there is no clear edge.

The trade columns tell the same story. The expectancy of +0.20 R per trade before costs looks attractive, but the same entry, stop and target on random dates averaged +0.19 R. The profit comes from the upward drift of the assets over the holding period, not from the pattern. On hourly crypto, the pattern hit its target 18.8% of the time against 19.6% on random dates; its average result per trade does not pass our statistical test: the row reads No clear edge, and we do not count it as an edge. On direction, hourly crypto double bottoms were higher 10 bars later 46.6% of the time against 48.2% for random bars, and their excess return over the rest of the market stays below zero after our correction for the thousands of tests we run.

What the breakdowns show

• Pattern height: on daily charts, double bottoms shorter than the median reached the 2:1 target 26.9% of the time, taller ones 15.0%. A taller pattern means a farther target.
• Volume on the breakout bar: 20.9% with high volume, 21.2% without.
• 200-day moving average: 19.8% above, 23.1% below.
• Stability: the excess return was −0.15% in the first half of the history and −0.15% in the second half, both close to zero.

These breakdowns are descriptive and not corrected for multiple testing. None of them turned the pattern into an edge over random dates. The full table, with every market, timeframe and verdict, is in the statistics section of this page. The method is published in our methodology.

Double bottom case studies

Here are real double bottom patterns our bench found on daily crypto charts, with their actual outcome. They are separate from the gallery cards of the Real examples section. Quotes are in USDT; every date is a completed daily bar, the most recent in December 2023, before the statistics were generated.

Example 1: a breakout that reached the target

ETH/USDT made a first trough at 1,531.01 on September 11, 2023, a peak at 1,751 on October 1 and a second trough at 1,521 on October 12. On October 23 it closed at 1,765.46, above the neckline. With a stop at 1,521, the 2:1 target was 2,254.38; price reached it on December 4, 2023. ETH was 2.0% higher 10 days after the breakout. The measured-move target, at 1,981, was reached on the way, on November 9.

ETC/USDT did the same in 2020, more slowly: troughs at 4.5031 (September 3, 2020) and 4.6885 (November 4), neckline at 5.777, breakout close at 6.0431 on November 18, and the 2:1 target at 9.1231 reached on January 10, 2021.

Example 2: breakouts that failed

LINK/USDT made troughs at 6.2 (August 29, 2022) and 6.306 (October 13), with a neckline at 8.54. It broke out at 8.715 on November 4 and fell back within days: it hit the stop at 6.2 on November 9, 2022, a full 1 R loss, and was 27.8% lower 10 days after the breakout. ETH/USDT broke out at 498.76 on July 17, 2018, above a neckline at 495.98, was 5.7% lower 10 days later, and hit its stop at 404.48 on August 3, 2018. More real cases, winners and losers, are shown in the Real examples section.

Tools for spotting the pattern

Charting platforms let you draw the neckline and mark the two troughs, and some scanners flag W shapes automatically. Check what a scanner calls "equal" troughs and whether it requires a close above the neckline: definitions differ, and so do the results. Before trading the pattern with real money, mark the last 30 double bottoms on your market, write down the entry, stop and target, and compare the result with the same trade on random dates. That exercise is the test we run at scale.

Frequently asked questions

Summary of key points

• A double bottom pattern is two troughs at about the same level with a peak between them, confirmed by a close above that peak (the neckline).
• More than half of the W shapes on daily charts never closed above the neckline.
• On 6,303 daily trades, the 2:1 trade reached its target 21.0% of the time, against 22.2% for the same trade on random dates.
• The expectancy of +0.20 R before costs matched the +0.19 R of random dates: the market drift, not the pattern, explains it.
• Volume and the 200-day moving average changed the results only slightly.

Final thoughts on risk management

If you trade the double bottom, treat it as a place to define risk, not as a reason to buy. Wait for the close above the neckline, place the stop below the troughs, size the position on that distance, and keep the risk per trade small. The pattern gives you a clear structure; any edge has to come from what you combine it with.

About the author and the data. Written by Amaury, founder of PatternsWizard, who is passionate about finance, quantitative trading and data analysis. He built the PatternsWizard test bench, which applies the same published rules to full market history: crypto, forex, US stocks, ETFs, indices and commodities. Trades and win rates are compared with random bars on the same assets, excess returns with the rest of the market, with error margins and a correction for multiple testing, and the whole pipeline is checked on random-walk prices, where nothing can be predicted. Statistics on this page were generated on November 5, 2026. The full methodology explains every step.