Donchian Breakout & Retest Signals [bitpunk]
Donchian Breakout & Retest Signals [bitpunk]
This indicator separates two ways to read a channel break: the first close outside the channel, and a later held retest. Use LB/SB to read a fresh long/short breakout, or LR/SR to wait for price to return and confirm again.
Two entry sequences 🧭
Long R watches the upper edge from above; short R mirrors it at the lower edge. A close beyond the pocket's adverse boundary cancels the watch; wick penetration alone does not. Same-side breakouts do not restart an active retest watch, while an opposite B replaces it.
Read the chart 📊
White ▲ LB / ▼ SB identify long/short breakouts; ▲ LR / ▼ SR identify long/short retest confirmations. The letters identify direction and timing, not signal quality. Signal Type defaults to Both; choose Breakout or Retest to select markers and alerts. If an accepted breakout and retest coincide in the same direction, Both reports R once.
By default, each direction receives one breakout and one independent retest signal until a completed close returns to the rolling channel midpoint. A breakout does not consume the later retest signal. This can suppress useful continuation entries; turn Structure Signal Reset off for the previous event stream.
Quiet lines show the rolling channel. The stronger fixed edge and shaded pocket show the retest watch; a dotted line is its confirmation gate. The panel reports the watch stage and remaining closes.
Useful settings ⚙️
Alerts and limits 🔔
Choose Any alert() function call, whose messages identify BREAKOUT / RETEST, or Long Channel Signal / Short Channel Signal. Signal Type applies to both routes. Enable Signals off hides entries and mutes alerts; Structure, panel and Normal / Enhanced remain display choices. Recreate alerts after changes.
Use standard time-based candles. All events wait for close. Breakouts can whipsaw; strict retests miss moves that never return. This independent MPL-2.0 implementation combines a prior-only Donchian breakout with a separate, time-limited held-test confirmation.
This indicator separates two ways to read a channel break: the first close outside the channel, and a later held retest. Use LB/SB to read a fresh long/short breakout, or LR/SR to wait for price to return and confirm again.
Two entry sequences 🧭
- B — Breakout. The channel uses the highest high and lowest low of the preceding 20 candles. A strict close outside it identifies a breakout. Wicks and equality do not count.
- R — Retest. A breakout freezes the broken edge and starts a watch. A later candle must touch its tolerance pocket and hold its adverse boundary on close. A still later candle must close beyond both the test candle's selected gate and the original edge. This is at least a three-candle sequence.
Long R watches the upper edge from above; short R mirrors it at the lower edge. A close beyond the pocket's adverse boundary cancels the watch; wick penetration alone does not. Same-side breakouts do not restart an active retest watch, while an opposite B replaces it.
Read the chart 📊
White ▲ LB / ▼ SB identify long/short breakouts; ▲ LR / ▼ SR identify long/short retest confirmations. The letters identify direction and timing, not signal quality. Signal Type defaults to Both; choose Breakout or Retest to select markers and alerts. If an accepted breakout and retest coincide in the same direction, Both reports R once.
By default, each direction receives one breakout and one independent retest signal until a completed close returns to the rolling channel midpoint. A breakout does not consume the later retest signal. This can suppress useful continuation entries; turn Structure Signal Reset off for the previous event stream.
Quiet lines show the rolling channel. The stronger fixed edge and shaded pocket show the retest watch; a dotted line is its confirmation gate. The panel reports the watch stage and remaining closes.
Useful settings ⚙️
- Retest Window — 10; Confirmation Window — 5. The first counts following candles after B to find a test. The second counts following candles after that test. Later contacts do not renew either clock.
- Retest Pocket (TR) — 0.25. Half-width around the edge, using mean true range captured before the breakout. Zero requires exact edge contact.
- Body Retest Gate — on. Uses the held test's body high/low instead of its wick extreme. Earlier confirmation can be weaker; turn it off for the wick rule.
- Max Retest Extension (TR) — 0. Zero disables the R distance cap. A positive value can skip wide tests and overextended R confirmations. B is never distance-capped; a skipped confirmation is not delayed into another R.
Alerts and limits 🔔
Choose Any alert() function call, whose messages identify BREAKOUT / RETEST, or Long Channel Signal / Short Channel Signal. Signal Type applies to both routes. Enable Signals off hides entries and mutes alerts; Structure, panel and Normal / Enhanced remain display choices. Recreate alerts after changes.
Use standard time-based candles. All events wait for close. Breakouts can whipsaw; strict retests miss moves that never return. This independent MPL-2.0 implementation combines a prior-only Donchian breakout with a separate, time-limited held-test confirmation.
Open research and indicators on stocks and crypto. Explore the work at bitpunk.io
Feragatname
Bilgiler ve yayınlar, TradingView tarafından sağlanan veya onaylanan finansal, yatırım, alım satım veya diğer türden tavsiye veya öneriler anlamına gelmez ve teşkil etmez. Kullanım Koşulları bölümünde daha fazlasını okuyun.
Open research and indicators on stocks and crypto. Explore the work at bitpunk.io
Feragatname
Bilgiler ve yayınlar, TradingView tarafından sağlanan veya onaylanan finansal, yatırım, alım satım veya diğer türden tavsiye veya öneriler anlamına gelmez ve teşkil etmez. Kullanım Koşulları bölümünde daha fazlasını okuyun.