A breakout is a close above a level you wrote down before the session, and a pullback is a pause that leaves the higher-low structure intact. Decide the level, the confirmation and the invalidation first, then judge each chart against them.
Define a breakout before it happens
A breakout is a daily close above a price level that you defined before the session, such as the highest high of a base or a prior swing high. Writing the level down first keeps the reading honest, because a level drawn after the move can always be placed just below the close. Trading above the level during the session does not count by this definition; the close decides. Many researchers also want volume above its recent average on the breakout session, as a sign of wider participation.
| Session | Close ($) | Low ($) | Close vs 50.00 (%) | Volume ratio |
|---|---|---|---|---|
| 32 | 50.90 | 49.48 | 1.80 | 2.72 |
| 33 | 51.60 | 50.56 | 3.20 | 1.70 |
| 34 | 52.30 | 51.45 | 4.60 | 1.28 |
| 37 | 50.65 | 50.15 | 1.30 | 0.90 |
| 39 | 51.55 | 50.77 | 3.10 | 0.66 |
Volume ratio is the session’s volume divided by its average over the last 20 sessions, the current one included, the same definition as the CleverScan field volumeRatio. Session 32 closes 1.80% above the level on 2.72 times average volume. The retest in session 37 trades down to 50.15 and closes above the level on lighter volume.
securityType = "Stock" and hi50FromHighPct >= -1 and volumeRatio >= 1.5Finds stocks closing within 1% of their highest high of the last 50 sessions on at least 1.5 times average volume. The 50-session window includes the current session, so hi50FromHighPct is never above zero; it is zero only when a session closes at its own high and that high is the highest of the window. Check each result on the chart to confirm that the high it cleared was a level you had marked.
Price compression before a breakout
Compression is a base in which the swings get smaller: lows rise toward a flat top, daily ranges narrow and volume tends to fade. In the figure the base lows rise from 46.20 to 47.30 to 48.40 while the highs stay between 49.90 and 50.00. Compression describes a narrowing range, not a direction, and the same tightening can resolve with a close below the lows.
- The height of the last 20 sessions’ range as a percent of its low, compared with earlier bases of the same stock.
- Average true range as a percent of the close,
atrPct14, compared with its own recent readings. - Swing lows that rise while swing highs stay level, the outline of an ascending triangle.
- Volume below its 20-session average on most sessions of the base.
- Narrowing Bollinger Bands, which measure the same contraction from closes.
securityType = "Stock" and hi20FromHighPct >= -3 and hi20FromLowPct <= 8 and atrPct14 <= 3Keeps stocks whose close is within 3% of the 20-session high and within 8% of the 20-session low, which caps the 20-session span at about 11.3% of the low, with an average true range of no more than 3% of the close.
Counter-example: a stock that drifts lower in a narrowing range for weeks is compressing too. A narrow range near the bottom of a decline answers a different question from a narrow range below a flat top in an uptrend, so read the structure around the base before reading the compression as strength.
How to wait for a breakout retest
A retest is a return toward the broken level after a breakout. If the level now acts as support, sessions close above it, or within a small tolerance of it, and the next swing low forms there. Waiting for a retest means writing three things in advance: the zone you expect price to revisit, the pattern that confirms the hold, and the close that cancels the idea.
- Set the retest zone, for example from the level down to 1% below it: 50.00 to 49.50 in the figure.
- Judge closes, not intraday lows: a dip into the zone that closes back above the level counts as a hold.
- Ask for confirmation, such as a close above the high of the retest session. That high is 51.23 in the figure, and session 39 closes above it at 51.55.
- Cancel the idea on a close below the zone, which turns the breakout into a failed one.
- Set a time limit in sessions, and record it when a breakout never offers a retest.
Not every breakout returns to its level. Waiting gives up the moves that never look back in exchange for a closer invalidation point, which is the trade-off the next section measures.
Breakout entry vs pullback entry
A breakout entry acts on the breakout close itself; a pullback entry waits for the retest and its confirmation. The breakout entry takes part in every follow-through but usually needs a wider invalidation, often below the last higher low of the base. The pullback entry can place its invalidation just below the retest low, and it misses the moves that never pull back.
| Entry rule | Entry close ($) | Invalidation ($) | Risk per share ($) | Risk (% of entry) |
|---|---|---|---|---|
| Breakout close, session 32 | 50.90 | 48.40 | 2.50 | 4.91 |
| Pullback confirmation, session 39 | 51.55 | 50.15 | 1.40 | 2.72 |
Here the pullback entry pays $0.65 more per share and risks $1.10 less. The comparison belongs to one hypothetical chart. How often breakouts offer a retest differs between stocks and periods, so measure it on your own sample before preferring either rule, and carry the risk per share into position sizing.
How to recognize a failed breakout
A failed breakout closes above the level and then closes back below it within a few sessions. The warning signs often appear together: a small margin above the level, ordinary volume, a long upper wick on the breakout session and a quick close back inside the base. The failure is confirmed when price also closes below the last higher low of the base, because the structure that supported the breakout is gone.
| Session | Close ($) | Close vs 50.00 (%) | Volume ratio | Reading |
|---|---|---|---|---|
| 23 | 50.35 | 0.70 | 1.09 | Close above the level on ordinary volume |
| 24 | 49.70 | −0.60 | 0.95 | Back below the level |
| 25 | 48.90 | −2.20 | 1.19 | Deeper inside the base |
| 26 | 47.35 | −5.30 | 1.69 | Below the last higher low of 47.60 |
Counter-example: a single close back below the level is not always a failure. Some breakouts dip back into the base for a session and recover inside the retest zone, which is why the definition above waits for a close below the last higher low before calling the breakout failed.
How to confirm a pullback in an uptrend
A pullback is a decline inside an uptrend that leaves the higher-low structure intact. Before treating a decline as a pullback, confirm the trend: a series of higher highs and higher lows, a close above a rising 50-session average, and the 50-session average above the 200-session average. Then measure how much of the last advance the decline gave back, and wait for the stock to show that it has stopped falling.
Retracement (%) = (swing high − pullback low) ÷ (swing high − swing low) × 100swing low- the low where the last advance started
swing high- the high where the advance ended; it must be above the swing low, otherwise there is no advance to retrace
pullback low- the lowest low of the decline so far
| Case | Swing low ($) | Swing high ($) | Pullback low ($) | Retracement (%) |
|---|---|---|---|---|
| Shallow pullback | 40.00 | 50.00 | 46.20 | 38.00 |
| Deep pullback | 40.00 | 50.00 | 41.50 | 85.00 |
The shallow case gives back 38% of the advance and holds well above the 40.00 swing low. The deep case gives back 85%: the structure is not broken yet, but it is close enough to the prior low that the next sessions decide whether the uptrend survives. Confirmation comes from price, such as a close above the prior session’s high, or a new higher low followed by a close above the last minor swing high.
securityType = "Stock" and close > sma200 and sma50 > sma200 and sma50 > Ago(1, SMA(close, 50)) and distSma50Pct >= -2 and distSma50Pct <= 2Lists stocks above their 200-session average whose 50-session average is above the 200-session one and higher than a session earlier, closing within 2% of that 50-session average. It finds candidates near a common pullback area; whether the swing structure is intact still has to be read on the chart.
Pullback or reversal?
A pullback and the start of a reversal look alike for the first few sessions. The difference shows in structure and participation: a pullback holds above the last higher low, while a reversal closes below it and then fails to make a new high. No single session settles the question, so compare the evidence side by side.
| Evidence | Pullback | Reversal |
|---|---|---|
| Last higher low | Holds on a closing basis | Closes below it |
| Next rally | Makes a higher high | Stops below the prior high, a lower high |
| Volume | Lighter on down sessions than on the advance | Heavier on down sessions than on up sessions |
| Depth | Similar to or shallower than earlier pullbacks | Deeper than any pullback of the trend |
| 50-session average | Still rising | Flattening or turning down, with closes below it |
Counter-example: a decline that stops at the 50-session average is not automatically a pullback. If it already closed below the last higher low on the way down, the average is the first support of a possible downtrend, not evidence that the uptrend continues.
Bullish divergence without a price reversal
A bullish divergence appears when price makes a lower low but a momentum oscillator such as RSI or MACD makes a higher low. It says that the decline lost speed between the two lows. It does not say that the decline is over: price can keep falling while the oscillator keeps rising, sometimes producing two or three divergences in a row.
- Price confirmation: a close above the last lower high, which ends the lower-high sequence.
- Structure confirmation: a higher low after the divergence low.
- Participation: volume that expands on up sessions after the low.
- Horizon: a divergence on a daily chart speaks to daily swings, not to a trend of several months.
Divergence is computed from the oscillator rather than from the swing structure, which is why it can persist through a downtrend. Read it on the chart: Stock Charts in TradeClever plots RSI and MACD in panes below the candles. CleverScan reads the current rsi14 value but does not compare two swing lows, so a divergence is something you mark in your notes, not a scan condition.
A research checklist for breakouts and pullbacks
Run the same steps on every candidate so that results from different charts can be compared. The list starts from a scan and ends with a written plan; it does not choose trades for you. Save the scan you use and keep a note of each rule’s version.
- Run a breakout or pullback rule in CleverScan and save it.
- Open each match in Stock Charts on a 1Y range and confirm the trend structure.
- Mark the level the breakout cleared and confirm that it existed before the breakout session.
- Measure the base: its height in percent, its length in sessions and whether its lows were rising.
- Check the breakout close: its margin above the level and its volume ratio.
- For a pullback, measure the retracement and locate the last higher low.
- Write the entry rule, the invalidation level and the time limit before the next session.
- Review the setup later with the same rule, whatever the outcome.
Continue with
Read our data and methodology notes for calculation context. This guide explains research tools and observations. It does not recommend a trade or investment.