Double Top and Double Bottom Patterns: How to Trade Both (2026 Guide)

Double top and double bottom patterns are the two cleanest reversal shapes on a chart: price tests one level twice, fails twice, and then breaks the swing between the two tests. The double top is an M that ends an uptrend. The double bottom is a W that ends a downtrend. This guide covers the structure of both, two worked trades with entries, stops and targets, what volume should be doing at each stage, where the stop actually belongs, and the specific conditions that turn these patterns into losing trades.

What double top and double bottom patterns are

A double top forms when price reaches the same resistance level twice, fails to break above it on both attempts, and then breaks below the trough that formed between the two peaks. Draw a horizontal line through that trough and you have the neckline. The double bottom is the mirror image: two tests of the same support level, a peak between them, and a break above that peak.

The mechanism is the part worth holding onto, because it tells you when the shape means something. At the first peak, buyers run into a supply of sellers and back off. Price pulls away, buyers regroup, and they try the level again. When the second attempt stalls at the same price, you have learned something concrete: the sellers sitting at that level were not cleared out by the first attempt, and the second wave of buying was not enough to clear them either. The break below the trough is the moment the buyers who bought the dip in between give up too.

That is also why the shape alone is worthless. Two bounces off a level in the middle of a choppy range is just a range. The same two bounces at the end of a long trend is a reversal signal, because there is a trend to reverse. One widely shared trading clip put it bluntly this month: a classic double bottom "is not enough to justify a trade," and its position within the overall market structure matters more than the shape.

A double top pattern traced on a price chart: an uptrend from 38 to a first peak at 52, a pullback to a trough at 46, a second peak at 52 that fails at the same level, then a break below the 46 neckline down to a measured target of 40. The stop sits at 49, halfway back into the pattern.

How to trade a double top, step by step

Take the setup in the chart above. A stock rallies from $38 to $52, sellers show up, and it pulls back to $46. Buyers try again two weeks later and stall at $52 to the cent. Price rolls over toward $46 a second time.

1. Check there was a trend. A double top reverses an uptrend. If price has been chopping between $46 and $52 for four months, those two peaks are range boundaries, not a topping pattern.
2. Mark the neckline. The trough between the peaks sits at $46. That single level is the trigger, the invalidation reference, and the measuring point for the target, so draw it off the actual swing low. Where you would prefer the level to sit is not a level.
3. Wait for a close below $46. This is the step most traders skip. Until price closes through the neckline, an M shape is a forecast. Shorting the second peak because it looks like a top is a guess dressed up as analysis, and the guess is wrong often enough to matter.
4. Enter on the close, or on the retest. The conservative entry is a decisive daily close below $46 on volume above the recent average. The second entry is the retest, covered below.
5. Measure the target. The pattern height is the peaks minus the neckline: $52 minus $46, so $6. Project that down from the break: $46 minus $6 gives $40. That is the measured move, and it is an estimate rather than a promise.
6. Place the stop at $49. Halfway back into the pattern, between the neckline and the peaks. The reasoning behind that specific level is worth its own section.

Entry at $46, stop at $49, target at $40. Three dollars of risk against six of reward, defined before you click anything. If the numbers on your chart do not produce a ratio you would accept, the correct response is to skip the trade, not to move the target.

Two peaks are a shape. A close below the trough between them is a trade.

How to trade a double bottom, step by step

Flip every element. A stock falls from $32 into a low at $20, bounces to $24, sells off to $20 a second time, and holds. The $24 peak between the two lows is the neckline.

• Entry: a close above $24, ideally with volume clearly above the recent average.
• Stop: $22, halfway back into the pattern between the neckline and the lows.
• Target: pattern height is $24 minus $20, so $4, projected up from the break: $28.

The double bottom tends to behave slightly differently from its mirror image, and the difference is about how tops and bottoms actually form. Tops are often quick and emotional, with two sharp pushes into the same supply. Bottoms are frequently slower, because the second low is made by sellers running out, and exhaustion takes longer than enthusiasm. It is common for the second low to undercut the first by a small amount, sweeping the obvious stops below the first low before reversing hard. That undercut does not invalidate the pattern. It is close enough to the same level that the story still holds, and traders who insist on two exactly equal lows will miss most of the real ones. If you want the mechanics of that stop sweep in detail, our guide to liquidity and stop hunts covers what is happening underneath it.

The same tolerance applies to the double top: a second peak a fraction above the first, followed by an immediate rejection, is still a double top. What invalidates it is a close well above the peaks, because at that point the resistance you were trading has been cleared.

A double bottom pattern traced on a price chart: a downtrend from 32 to a first low at 20, a bounce to 24, a second low at 20, then a break above the 24 neckline up to a measured target of 28. The stop sits at 22, halfway back into the pattern.

What volume should be doing

Volume is the cheapest filter you have for separating a real double top from two peaks that happen to line up. The textbook signature has three parts: solid volume on the first peak, noticeably lighter volume on the second peak, and a neckline break that occurs on volume above the recent average.

Each part means something specific. Heavy volume into the first peak says a lot of people participated in that push, which is what makes the level significant in the first place. Lighter volume on the second peak says fewer participants showed up to retest it, which is the early warning that the move is running out of fuel. Then the break on expanding volume says the sellers are committed rather than opportunistic. A neckline break on a quiet day is the one that snaps back and stops you out.

Bar chart of the volume signature across a textbook double top: the first peak trades at about 130 percent of the 50-day average volume, the trough between the peaks at 70 percent, the second peak at 85 percent, and the neckline break at 145 percent. The pattern is heavy volume first, lighter on the retest, then expansion on the break.

Two caveats keep this honest. First, the signature is a tendency, not a rule, and plenty of patterns that work look messier than the chart above. Weak volume on the break is a reason to size smaller or wait for the retest. Plenty of thin breaks still reach target. Second, volume is far more readable on stocks than on crypto and forex, where 24-hour sessions and fragmented venues distort the bars. On those markets, lean harder on the price structure and on whether the level lines up with anything else.

Where the stop actually goes

Most guides say to put the stop above the peaks on a double top. That placement is defensible, and it is often too wide to trade.

There is a better answer, and it comes from thinking about what each level does to your trade. A stop sitting right on the neckline gets taken out by the ordinary retest that follows most breaks, which means you lose on patterns that went on to work. A stop above the peaks survives almost everything, but on the example above it would sit near $52.50 against an entry at $46, turning a six dollar target into a six dollar risk and destroying the reason to take the trade. Halfway back into the pattern, at $49, is the compromise: far enough above the neckline to survive a normal retest, close enough that the reward-to-risk stays worth taking, and high enough that if price gets there the breakout has clearly failed.

You can hold the peaks as a hard invalidation level in your head while placing the working stop at the midpoint. The two are not in conflict. One is where the pattern is objectively dead; the other is where your specific trade no longer makes sense. For the general framework behind sizing that risk, see where to place a stop loss.

The retest, and the second entry it gives you

After the neckline breaks, price often climbs back to it from below, taps it, and gets rejected. The level that was support has flipped to resistance. That retest is one of the better entries available on these patterns, because you are entering at the same price as the breakout traders but with a much tighter stop and with one extra piece of evidence in hand.

It also will not always happen. Sharp breaks sometimes run straight to target without looking back, and waiting for a retest that never comes is how traders miss the move they correctly identified. Treat it as a bonus. Take the close-through entry with the midpoint stop, and if a retest materializes, use it to add.

What the retest must not do is close back inside the pattern. A wick into the old neckline is normal. A daily close back above it on a double top says the break failed, and the trade is over regardless of where your stop sits.

How far apart the peaks should be

The two tests need separation, in both price and time. Without it you are looking at consolidation.

• Time between the peaks. On a daily chart, a few weeks between the two tests is typical, and patterns that develop over months carry more weight than ones that form in three sessions. Two peaks two bars apart is a double-topped candle, not a double top.
• Depth of the trough. The pullback between the peaks needs to be meaningful. A trough that retraces a small fraction of the run up gives you a shallow pattern, a small measured move, and a neckline that price chops through constantly. Deeper troughs give a cleaner trigger and a bigger target.
• How close the two peaks are. Within a percent or two of each other is the usual guidance on daily charts, wider on volatile names. The point is not precision, it is that both attempts clearly failed at the same supply.

Timeframe matters as much as any of it. These patterns form and fail all day on the 1- and 5-minute chart, where a "double top" is frequently just two ticks against a round number. The daily and 4-hour versions are worth trading; the ones you find by scrolling a 1-minute chart at 10am usually are not.

Why these patterns fail

They fail constantly, and the failures cluster into recognizable causes.

The pattern was inside a range. This is the big one. Two touches of the top of an established range are exactly what a range does. Without a prior trend, a double top has nothing to reverse, and the break through the neckline is just as likely to be the range's normal rotation to the other side.

The trader entered before the break. Shorting the second peak feels smart and is the single most expensive habit around these patterns. The second peak is a hypothesis until the neckline goes, and price makes new highs from apparent double tops often enough that the early entry has to be paid for out of the trades that work.

The break had no participation. A neckline break on volume below average is the profile of a break that gets reclaimed. It can still work. It fails more often.

The level was not real. Necklines drawn through minor intraday wiggles produce triggers that mean nothing. If the level does not show up as support and resistance that other traders would also mark, the break through it is not information.

The market structure contradicted the pattern. A double bottom inside a downtrend that is still making lower highs and lower lows is a counter-trend trade, whatever the shape says. Those can work, but they are a different trade with a different success rate, and they deserve smaller size.

The deeper problem with all reversal patterns is discretion. Two traders looking at the same chart will draw different necklines, accept or reject the same second peak, and end up in opposite positions. That subjectivity is why individual results with these patterns vary so much, and why a rule you cannot write down is a rule you cannot evaluate. Thomas Bulkowski's work on chart pattern performance remains the standard reference for anyone who wants measured base rates.

Double top vs triple top vs head and shoulders

These three are the same family and traders mix them up constantly. The difference is only in the peaks:

• Double top: two peaks at roughly the same level, an M.
• Triple top: three peaks at roughly the same level. Rarer, and a failed third attempt at a level that has already rejected twice is a stronger statement than a failed second.
• Head and shoulders: three peaks where the middle one is clearly higher. Covered in full in our guide to trading the head and shoulders pattern.

Once any of them breaks, the mechanics are identical: draw the support through the lows, wait for a close below it, project the height of the pattern down from the break for the target. Telling them apart helps you spot the setup. It changes nothing about how you manage the trade. All three sit in the wider family covered in stock chart patterns, which is the place to start if you are still learning which shapes are worth your attention.

Common mistakes

• Trading the shape before the neckline breaks. The M is a forecast until price closes through the trough. Shorting the second peak is guessing with a story attached.
• Putting the stop on the neckline. The retest will take it out, and you will lose on the patterns that worked.
• Demanding two identical peaks. A second test a little above or below the first is normal, and often the better signal, because it cleared the obvious stops before reversing.
• Skipping the trend check. In a range, these are not reversal patterns. They are the range.
• Treating the measured move as a promise. It is a reasonable estimate derived from the pattern's own height. Take partial profits into it.
• Trading them on the 1-minute chart. Mark them on the daily or 4-hour, then drop down for the entry if you want a tighter stop.
• Forcing the pattern. If you have to argue a chart into an M, there is no M. This is where a scanner genuinely helps, because it has no opinion about the trade you already want to take; the apps that detect stock patterns are compared here.

Frequently asked questions

Is a double top bullish or bearish?

A double top is bearish. It forms after an uptrend, shows price failing twice at the same resistance, and signals a reversal down once the neckline breaks. A double bottom is bullish: it forms after a downtrend and signals a move up once price closes above the peak between the two lows.

Where should the stop loss go on a double top trade?

Halfway back into the pattern, between the neckline and the peaks. A stop right on the neckline gets hit by the ordinary retest that follows most breaks, and a stop above the peaks is so far away that the reward-to-risk collapses. The midpoint is the level where the breakout has clearly failed but the trade is still worth taking.

How do you set a target on a double bottom?

Measure the height of the pattern, from the lows to the neckline, then project that distance up from the break point. If the lows are $4 below the neckline, the first target is roughly $4 above it. This is the measured move, and price frequently stops short of it.

How reliable is the double bottom pattern?

It improves the odds when the conditions line up, and it is not close to certain. Reliability rises when the pattern follows a genuine downtrend, sits at a level that already mattered, shows lighter volume on the second low, and breaks out on volume above average. It falls sharply when the pattern forms inside a range, on a very short timeframe, or against the higher-timeframe structure. Always trade it with a stop.

Do the two peaks have to be exactly the same price?

No, and insisting on it will cost you most of the real patterns. Within a percent or two on a daily chart is the usual guidance, wider on volatile names. A second peak that pokes slightly above the first and is rejected immediately is still a double top. What kills the pattern is a close well above the peaks.

Can you trade a double top on an intraday chart?

You can, with two adjustments. Use it on the higher intraday timeframes. On the 1-minute these shapes appear and fail all session. Then check that the pattern lines up with a level that matters on the daily chart. An intraday double top at a daily resistance level is a real setup. The same shape in the middle of the day's range is noise.

What happens if price closes back above the neckline after breaking down?

The pattern has failed. A wick back into the old neckline is a normal retest, but a close back inside the pattern means the sellers who caused the break did not follow through, and the trade is over whether or not your stop has been hit. Failed reversal patterns frequently run hard in the opposite direction, so this is not a situation to sit in and hope.

Reading the pattern without drawing it yourself

Once you know the structure, spotting a double top is mostly discipline: find the level, check that a trend preceded it, wait for the close through the neckline, measure the target, and place the stop at the midpoint, where the trade stays worth taking. The hard part is not the geometry. It is being honest about whether the shape is really there when you already want the trade.

Quant AI reads a chart screenshot and marks the levels and patterns it finds, including the M and W shapes covered here, with the neckline and the measured move drawn in. It does not know whether the setup fits your account, your timeframe, or your risk, and it does not predict what price will do next. What it removes is the part where you squint at a chart at 9:45am and talk yourself into a pattern that is not there.