Double Top and Double Bottom Patterns: How to Read a Failed Retest

Price walks up to a level, gets rejected, pulls back, then comes back and gets rejected again at roughly the same spot. That double rejection is the heart of a double top. Flip it upside down and you get a double bottom: two failures to push lower off the same floor. These are two of the most talked-about reversal shapes in trading, and for once the reputation is mostly earned. They are simple, they map to something real about buyers and sellers, and they give you a clean place to be wrong.

The catch is that most people trade them backwards. They see two peaks, decide the top is in, and short into strength before anything has actually confirmed. This piece is about doing it the other way around: understanding what the shape is telling you, waiting for the neckline to do its job, and setting invalidation before you have any opinion about where price goes next.

What a double top and double bottom actually are

A double top looks like the letter M. Price rallies to a high, sells off to an interim low, rallies again to about the same high, and fails. The line drawn across that interim low is the neckline. The pattern is not complete, and is not really a pattern yet, until price closes below that neckline.

A double bottom is the mirror image, a W shape. Price drops to a low, bounces to an interim high, drops again to about the same low, and holds. The neckline runs across the interim high, and the pattern confirms on a close above it.

Why two touches and not one? Because a single rejection is just support and resistance doing its normal thing. The second failure is the tell. It says the side that was in control (buyers into a top, sellers into a bottom) showed up again and could not make new progress. That loss of momentum at a known level is the whole idea. The market tried the same thing twice and got the same answer.

The story behind the shape

It helps to read the pattern as a small drama rather than a picture. On a double top, the first peak is an uptrend running into supply. The pullback is normal profit-taking. The second push is the trend trying to prove itself. When that push stalls at the same price, the buyers who were driving things have run out of new fuel, and the traders who bought the first pullback start to reconsider. The break of the neckline is the moment enough of them give up that price falls away.

None of this guarantees a reversal. It describes a shift in the balance of pressure, and a shift in pressure is a probability, not a prophecy. Keep that framing and you will hold these patterns far more sanely than someone who thinks two peaks mean the top is guaranteed.

The neckline is the whole trigger

If you take one thing from this article, take this: the pattern does not exist until the neckline breaks on a closing basis. Two equal highs with no neckline break is just a range. Plenty of ranges have two, three, five touches at the top and never reverse at all.

Here is why the closing break matters so much. Intraday, price will poke through necklines constantly and snap back. A wick below the neckline of a double top is not confirmation, it is bait. Waiting for a candle to close beyond the level filters out a large chunk of the noise. It costs you a little entry price, and it saves you from a pile of fakeouts. That trade is almost always worth it.

Some traders add a second filter: they wait for the neckline break, then wait for price to come back and retest the broken neckline from the other side. On a confirmed double top, that means price breaks below the neckline, rallies back up to it, fails to reclaim it, and rolls over. The retest gives you a tighter place to set risk and confirms the level flipped from support to resistance. It also means you sometimes miss the move entirely when price never retests. There is no free lunch here; you are trading a better entry for a lower hit rate on getting filled.

Volume, if you have it

Classic technical analysis says volume should fade on the second peak of a double top and pick up on the neckline break. The thinking is reasonable: a weaker second push and a decisive break is a cleaner story. In practice, volume is more useful on some markets than others. On 24-hour markets like crypto and spot gold, session volume can be messy, so treat it as a supporting detail rather than a requirement. If volume confirms, good. If it is ambiguous, do not let that alone keep you out of an otherwise clean setup.

Measuring a target without predicting one

Double tops and bottoms come with a built-in measured move, and it is worth knowing even though you should not marry it.

The method: measure the height of the pattern, from the peaks (or troughs) to the neckline. Then project that same distance from the neckline in the direction of the break. On a double top, subtract the height from the neckline for a downside target. On a double bottom, add it above the neckline.

The honest caveat: this is a guideline, not a magnet. Price is under no obligation to travel exactly one pattern-height and stop. Sometimes it falls short, sometimes it blows through. Use the measured move as a rough sense of whether the trade offers enough room to be worth the risk, then let a trailing exit or your own risk-reward math decide when you are actually done. A measured target that sits barely past your stop distance is telling you the setup is not worth it.

Setting invalidation before you have an opinion

This is the part that separates trading the pattern from gambling on it. Before you enter, you decide the single price at which the pattern is simply wrong. Not the price where you panic, the price where the structure that made you interested no longer exists.

For a double top, that price is above the higher of the two peaks. The entire premise is that the level twice rejected price. If price closes back above that level, it did not reject; it consolidated and broke out. Your reason for being short is gone, so you leave. For a double bottom, invalidation sits below the lower of the two troughs, same logic in reverse.

Define that level first, before you think about targets or size. Then work backwards. The distance from your entry to invalidation is your risk per unit, and that distance drives how large your position should be. A common rule of thumb is to risk a small, fixed slice of your account on any one trade, often cited around one percent, so that a string of failed patterns is an annoyance rather than a crater. The pattern gives you a clean invalidation point; use it. Setting a proper stop-loss is not optional decoration on a reversal trade, it is the thing that lets you take the next one.

Trading is risky and most reversal attempts do not follow through, so size for the losses, not the wins.

Where these patterns fail

A few honest failure modes worth knowing:

• The triple top in disguise. What looks like a double top sometimes just adds a third touch and keeps ranging. Your invalidation above the peaks handles this, which is exactly why you set it.
• The trend that never asked your permission. In a very strong trend, price can slice through a supposed neckline and keep going the original direction. Two touches against a freight train is not a lot of evidence.
• The too-small pattern. A double top that is only a few candles tall on a fast time frame is often just noise dressed up as structure. Bigger, cleaner patterns on higher time frames tend to behave better.

If you want to get sharper at telling a real one from a mirage, it is the same muscle as spotting any chart pattern: you are looking for a clean level, a clear failure, and a decisive break, not a vague vibe.

How this fits a trend-first approach

Double tops and bottoms are reversal patterns, which means by definition they fight the prior trend. That is fine, but it is worth being honest that fighting the trend is the harder side of the trade. Many traders only take these patterns when they line up with a bigger-picture reason, such as a double top forming right into a major resistance zone on a higher time frame, or a double bottom at long-standing support.

This is also where a systematic tool can keep you honest. It is very easy to see an M on the chart, decide the trend is over, and short into an uptrend that has plenty of life left. A trend-reading tool that simply states long, short, or flat, and stays flat most of the time, gives you a second opinion on whether the reversal you think you see has actually shown up in the trend itself. Vektor is built around that idea: it reads the trend on gold and Bitcoin, waits through most of the chop, and plots its exit as a trailing stop that follows the move rather than guessing the turn. It will not draw your necklines for you, but it can stop you from confusing a normal pullback with a real reversal.

If reversal patterns are not clicking yet, it is often because the underlying concept is shaky rather than the pattern itself. The head and shoulders pattern works on the same neckline logic and is worth studying alongside this one, and reversals in general make a lot more sense once you are comfortable with the idea of trend following and why most of the money is made by staying with a move rather than fading it.

The short version

Double tops and double bottoms are two failures at the same level, an M and a W. They are not patterns until the neckline breaks on a closing basis, so wait for that close instead of front-running it. Measure the target from the pattern height if you want a rough sense of room, but do not treat it as a magnet. Set your invalidation above the peaks (or below the troughs) before you do anything else, size the trade off that distance, and accept that plenty of these will fail. The whole edge of the pattern is that it hands you a clean place to be wrong. Use it.

FAQ

How reliable is a double top or double bottom?

Reliable enough to trade, not reliable enough to trust blindly. The pattern only counts once price closes through the neckline, and even then plenty of them fail. Treat it as a setup with a defined risk point, not a promise.

Do the two peaks or troughs have to be exactly equal?

No. Rough equality is fine, and a second peak slightly higher or lower than the first still qualifies as long as price failed near the same zone. Treat the level as a band, not a single price.

Where do I put the stop on a double top?

A common approach is just beyond the higher of the two peaks, so a break back above the failed level takes you out. On a double bottom, mirror it below the lower trough.

What time frame works best for these patterns?

Higher time frames like the 4-hour and daily tend to produce cleaner patterns with fewer false necklines. Lower time frames print more of them, but a larger share are noise. Pick one you can actually watch and stay consistent.