If the last lesson was the rule that fails, this is the one that works.
A breakout is defined here with no judgement at all: today's close is above the highest high of the previous 20 sessions. That is it. No trend lines, no patterns, no interpretation.
| sample | touched +20% | vs the market | |
|---|---|---|---|
| Close above the 20-day high | 1,895 | 24.9% | 1.42× |
| Held-out years | 1,946 | 34.7% | 1.20× |
| Close above the 52-week high | 602 | 33.6% | 1.91× |
| Held-out years | 922 | 37.1% | 1.29× |
Breaking a 52-week high, the thing that feels most like "too late", was the strongest single classic signal we measured, in both periods.
Volume decides whether the breakout is real
Splitting the same breakouts by how much volume came with them:
| Breakout on... | sample | touched +20% | vs the market |
|---|---|---|---|
| 2× normal volume or more | 699 / 877 | 28.6% / 36.3% | 1.63× / 1.26× |
| Below-normal volume | 629 / 520 | 20.2% / 32.7% | 1.15× / 1.13× |
A breakout on heavy volume beat a breakout on light volume in both periods. The price crossing the line is the claim; the volume is the evidence. A quiet breakout is a stock drifting through a number nobody defended.
The best combination we found
| Setup | sample | touched +20% | vs the market |
|---|---|---|---|
| Breakout and above the 200-day average | 1,426 / 1,755 | 27.2% / 36.0% | 1.55× / 1.25× |
Note what this is not: it is not "breakout + RSI below 70", which measured 1.02× and 1.00× exactly nothing. Filtering a breakout for "not yet overbought" removed the entire edge. We will come back to why in the RSI lesson, because the reason is the most counter-intuitive number in this course.
Do this now, practical activity
- Open SuperCharts and load any stock in today's signals.
- Set the range to 1 year and find the highest point on the chart.
- Ask: is price above it, at it, or below it? That single reading places the stock in one of the rows above.
- Now turn on the volume pane at the bottom. On the day price crossed its recent high, was the volume bar taller than the ones around it, or shorter?
The two-part check you now own: did it break, and did anyone show up? Those two questions separate a 1.63× setup from a 1.13× one.
Measured: a close above the 52-week high touched +20% within 30 sessions in 33.6% of 602 cases (1.91×) and 37.1% of 922 held-out cases (1.29×). On 2×+ volume, 20-day breakouts measured 1.63× and 1.26× against 1.15× and 1.13× on light volume.