Cup and handle: a top-ranked pattern

The shape, and its measured criteria

Cup and handle, criteria included A decline then a recovery tracing a rounded bottom, a U and not a V, over 7 to 65 weeks. Then a handle of at least one week, formed in the upper half of the cup, before the upward exit. The target is the height of the cup, projected from the breakout point: it is reached in 61% of recorded cases.

A cup and handle reads in two stages. First a decline then a recovery drawing a rounded bottom, the cup. Then a small consolidation near the top, the handle, before the upward exit.

The published criteria are precise, and they are what separates the pattern ranked 3rd from an approximate drawing.

The cup must be U-shaped, not V-shaped. A pointed bottom is not a cup: it describes a violent reversal, not patient accumulation. It is the most commonly ignored criterion, and the most discriminating.

Duration runs from 7 to 65 weeks, variations allowed. This is a longer-term pattern, not an intraday one. If you think you see one on five minutes, you are looking at something else.

The handle lasts at least one week, with no maximum, and it forms in the upper half of the cup. A consolidation that drops back into the lower half invalidates the pattern: the market did not hold its gain.

U or V, high or low handle On the left, the shape the published criteria require: a rounded U bottom and a handle in the upper half. In the middle, a sharp V bottom tells of a brutal reversal, not patient accumulation. On the right, a handle dropping into the lower half: the market did not hold its gain, and the pattern is invalidated.

The numbers

Study by Thomas Bulkowski across 913 perfect trades, quoted as it stands.

Performance rank: 3 out of 39. That is the podium.

Failure rate: 5%. Compare that with the 25% and 37% of the symmetrical triangle. One pattern in twenty goes nowhere, against one in four for the triangle.

Average rise: 54%. Price target met: 61%.

Return to the pattern after the break: 62%. That is the one disappointing figure in the set, and it sits at the same level as the triangle's. The pattern is excellent, the entry still gets shaken two times out of three, and that is what ruins most trades.

Source: thepatternsite.com, consulted in September 2026. The cases studied cover US stocks.

Cup and handle versus symmetrical triangle, in % The two extremes of the category, upward breakout, in percent. The cup fails one time in twenty, against one in four for the symmetrical triangle, and rises 54% on average against 34%. Only one number does not move: 62% returns to the broken line for both. The pattern is excellent, the entry still gets shaken two times in three.

The trap that ruins the pattern

Put the two numbers that matter together: 5% failure and 62% returns after the break. They say something precise. The pattern almost always works, and it almost always goes through a moment where it looks like it is not working.

Concretely: you enter on the break of the handle, price comes back down to touch the level you have just broken, your stop placed just below it is taken out, and the 54% rise happens without you. That is not bad luck, it is the normal behaviour of the pattern, documented in nearly two cases out of three.

The two possible answers. Enter on the return rather than on the break, which costs the comfort of a clean entry but gives a far better price. Or place the stop below the low of the handle rather than just under the breakout line, which costs position size but lets the pattern breathe.

What you must not do is enter on the break with a tight stop. That is the losing combination, and it is predictable in advance.

The cup's trap: the return to the line You enter on the handle breakout, at 1, with a tight stop just under the line. Price comes back to touch that line, as in 62% of cases, the stop is hit, and the rise happens without you. The two fixes: enter on that return, at 2, or place the stop below the low of the handle, which lets the pattern breathe at the cost of a smaller position.

What it gives in your own journal

This pattern raises a different journal question from the others, and it is the most interesting in the category: how many cups have you actually traded, and how many have you spotted after the fact?

A pattern taking seven to sixty-five weeks to form is far easier to recognise looking back than waiting in front of a screen. If your journal holds no entry on a cup while you recognise them on every chart you review, the problem is not the pattern, it is that you do not handle it live.

The second measurement is the maximum adverse excursion of your entries on cups. Given the 62% of returns, it should be large on many of your winners. If your losers show an adverse excursion barely larger than your winners, your stop sits in the noise zone of the pattern and it needs to move further out, even at the cost of position size.

Tradoshi computes that measurement on every imported trade, and the label isolates the population.

Frequently asked questions

Can the cup be V-shaped? The published criteria ask for a U. A V describes different market behaviour and the figures quoted here do not apply to it.

Is there an inverted cup? Yes, the bearish version, often called an inverse cup and handle. It is markedly less documented, so treat it as unquantified.

Which timeframe? The pattern forms over seven to sixty-five weeks. That is daily or weekly. On low timeframes, what you see is not this pattern.

Where does the target go? The height of the cup, projected from the breakout point. The target is met in 61% of recorded cases.

And if the handle drops into the lower half? The pattern is invalidated. That is a criterion, not a preference.

Key takeaways

A U-shaped cup and not a V, from 7 to 65 weeks, with a handle of at least one week forming in the upper half. Those are criteria, not preferences.
Ranked 3rd out of 39 with 5% failure and a 54% average rise: the best ranking in this whole file.
But 62% of returns to the broken line. The pattern almost always works and almost always goes through a moment where it looks like it fails.
Entering on the break with a tight stop is the losing combination, and it is predictable before you enter.
In your journal, first count how many cups you traded live rather than recognised after the fact.

Going further

These blog articles dig into this lesson's ideas, one subject per article.

Patterns in trading: spotting what makes you win or lose