Complete Guide to the Cup-and-Handle Pattern
— William O'Neil's Surge Criteria
The Cup and Handle is a flagship bullish pattern that William O'Neil formalized as part of his CANSLIM strategy. We cover cup depth, handle conditions, and pivot-point entry on a practical basis.
Pattern Structure
The Cup: a U-shaped move where the price falls from a peak, forms a bottom, and recovers back to roughly the old peak level. It takes a few weeks to a few months. The ideal depth is −15 to 33% below the peak.
The Handle: a shallow pullback of −5 to 15% from the right-side peak of the cup. It lasts 1 to 4 weeks. Ideally, volume shrinks during this pullback. The handle's high is the pivot point.
Entry Conditions
- A breakout above the pivot point (the handle high + 0.1 to 0.5%)
- Volume on the breakout day at 1.5x or more the 50-day average
- The broader market in an uptrend (confirm the Nasdaq and S&P)
Worked Example — Checking a Cup and Handle with Numbers
Example (illustrative numbers — not a call on any real stock): say a stock peaks at $100, then declines to a $70 low. The cup depth is (100−70)/100 = 30%, inside the ideal −15 to 33% range. The price then recovers to roughly $100 and pulls back to $92. The handle pullback is (100−92)/100 = 8%, within the −5 to 15% guideline. With the handle high (the pivot point) at $100, applying the pivot+0.1–0.5% rule puts the entry in the $100.10–$100.50 range. If the stock breaks out at $100.30 on volume at least 1.5x the 50-day average (say, 1 million shares typically vs. 1.5 million that day), the entry condition is met. If the price later falls below the handle low ($92), you'd stop out — by O'Neil's rule (−7 to 8% from the pivot price), that's roughly $92.80, a −7.5% stop from the $100.30 entry.
Stop-Loss on Failure
If the price falls below the handle's low after a pivot entry, stop out. O'Neil's own standard is −7 to 8% from the pivot entry price.
When This Pattern Is Wrong
Even when every cup-and-handle condition is met (−15 to 33% depth, a −5 to 15% handle, a 1.5x-volume breakout), the move can still fail to follow through. Common causes: ① a one-day volume spike on the breakout that fades the very next session ("one-shot volume"), ② a breakout that happens while the broader market (Nasdaq, S&P) has already turned down — the market's direction often overrides an individual stock's pattern, and ③ a handle that pulls back much deeper than the ideal −5 to 15% (−20% or more), which can mean buying interest is already exhausted by the time of the breakout.
DawnScan does not track "cup and handle" itself as a discrete signal with a published measured success rate (recognizing this shape from a chart is somewhat subjective — where exactly the cup and handle begin and end can differ from one reader to the next, which makes it hard to define consistently enough for code to tally). The quantitative signals that often co-occur with a cup-and-handle breakout, like approaching a 52-week high or a Bollinger squeeze, are ones DawnScan actually tracks, with their hit rate against the full stock universe updated daily on the Base Rate page — this article does not invent a "cup and handle succeeds X% of the time" figure.
How This Connects to DawnScan
The DawnScan scanner catches stocks showing a Bollinger squeeze or OBV accumulation near a 52-week high. Confirming a cup-and-handle pattern on the chart for a stock that already meets these conditions can help with the entry decision.
What to Check Before Trusting the Pattern
There are so many chart patterns that testing several at once makes it look like at least one of them happens to fit by chance. The procedure for filtering out this illusion is multiple-testing correction, and continually tuning criteria like cup depth and handle length to fit historical data leads straight into overfitting. These two are exactly the gates DawnScan runs a signal through before adopting it.