Double Top and Double Bottom — Reversal Patterns

Twice the market pushes up to the same high, and twice it gets rejected. Between those attempts a trough is left behind, and that trough, not the high itself, is the real signal. This is the double top, one of the oldest reversal patterns on the chart; its mirror image, the double bottom, works after a decline. Below I show how to confirm it and when to walk away.

What exactly is a double top and a double bottom?

A double top, which looks like the letter M, forms when price rises to a level, pulls back, returns to roughly the same place and turns away again. Two failed attempts at the same barrier say it plainly: buyers tried twice and twice came up short. A double bottom, shaped like the letter W, is the exact opposite — two failed attacks by sellers on the same level after a decline.

The element between the peaks or troughs matters most. With a double top it is the trough separating the two attempts; with a double bottom, the local high between the lows. We call that level the confirmation line, or neckline. Until price breaks through it, the pattern does not exist yet — you only have two turning points that may still be a pause inside a trend.

Why doesn't the second peak have to match the first exactly?

The textbook drawing shows two peaks at the same height, but the market is rarely that polite. In practice the second top often sits a little lower, and the second bottom a little higher — and that is perfectly acceptable. A slightly lower second top is frequently the stronger signal: it shows buyers lacked the strength even to repeat their previous result.

My rule of thumb: a difference between the peaks of up to about three percent of the pattern's height is fine. If the second top clearly exceeds the first, the pattern is invalidated — a continuation, not a reversal. What matters more than symmetry is the reaction of price: both turning points should look like places where supply genuinely halted the move.

"A double top occurs when prices reach a new high, decline, then return to within a few percentage points of the prior high before tumbling again." — Thomas N. Bulkowski, Encyclopedia of Chart Patterns, Wiley, 2005

How do you confirm the pattern and measure the target?

Confirmation is a candle closing beyond the confirmation line, not a wick merely piercing it. With a double top I wait for a close below the trough dividing the two peaks; with a double bottom, for a close above the local high between the lows. That distinction between "price touched" and "price closed beyond" filters out most false moves. Since support and resistance underpins every pattern, it pays to first learn how to draw support and resistance levels properly.

The target comes from the measured-move method: measure the height from the peaks to the confirmation line, and project it from the break point in the direction of the reversal. The stop goes just beyond the second peak (or trough), because a return into that zone means the scenario has failed. Many traders prefer a more patient entry on the retest, when price returns to the confirmation line, bounces off it from the other side, and only then opens the position with a tighter stop.

A hypothetical worked example, step by step

Here is the logic on daily EUR/USD — hypothetical only. After weeks of gains the pair stalls at 1.0900, pulls back to 1.0820, then returns and forms a second peak at 1.0890, a touch lower. The trough at 1.0820 is the confirmation line, so the pattern is roughly 80 pips tall. While price stays above 1.0820 I do nothing; the signal is only a daily close below it. The measured-move target then sits near 1.0740 (the break level less 80 pips), with the stop just above 1.0890. If price climbs back above both peaks, the pattern is void.

What is the honest success rate of this pattern?

Here I have to be candid, because the internet promises miracles. Research by Thomas Bulkowski shows a typical Adam and Adam double top has a break-even failure rate near 25 percent and ranks 19th of 36. Some variants do better — the Eve and Eve double bottom fails in only about twelve percent of cases. Good numbers for technical analysis, but far from a sure thing.

Most false signals are born in consolidation, when the market drifts sideways and keeps producing "almost patterns" that are really just range noise. That is why two filters help: the pattern should end a clear trend rather than appear mid-drift, and it is better when the break comes with rising activity. The pattern is also stronger when it agrees with the bigger picture, fitting the logic of trend-following systems by signalling that a trend has run out of fuel. Our technical analysis course places chart patterns alongside indicators and price action.

How is it different from head and shoulders?

Both patterns belong to the same family of reversal signals. A double top rests on two peaks at a similar level, while head and shoulders rests on three, the middle one (the head) being the highest. The double top is easier to spot and appears more often, while head and shoulders is often considered slightly more reliable because its structure is more developed. For a detailed comparison, see the head and shoulders pattern. The decision mechanics are identical: a close beyond the confirmation line, a measured-move target, and a stop beyond the pattern's extreme.

The most common mistakes

The most dangerous mistake is entering before price has closed beyond the confirmation line. Close behind come mistaking range noise for a real pattern, rejecting valid setups because the second top is fractionally lower, and trading very low timeframes where stray "double tops" lose all meaning. The last trap is dropping the stop-loss: if even the best variants fail in a double-digit share of cases, going unprotected is only a matter of time.

What to do tomorrow

1. Open the daily chart of a pair you follow, scroll back about a year and mark every spot with two peaks or two troughs at a similar level, learning to tell genuine structures from sideways drift.
2. For each structure, draw the confirmation line through the trough between the peaks or the high between the troughs, then check whether price truly closed beyond it, counting how many confirmed versus false alarms.
3. Mark the measured-move target and the stop-loss beyond the extreme peak or trough on those examples, then compare reward against risk and reject any setup where the potential gain is not clearly larger than the risk.
4. Write your own rule for the acceptable difference between peaks into your trading journal — for instance up to three percent of the pattern height — and apply it consistently rather than judging each formation by eye.
5. Before going live, test the whole routine on a demo account across a dozen or so signals, noting every entry, exit and reason for the decision, because only a repeatable demo result earns the right to risk capital.