OPEN-SOURCE SCRIPT
Adaptive Volumetric Reversion Channel Fade Validator
Adaptive Volumetric Reversion Channel — Fade Validator (AVRC)
What it is
AVRC is an anchored, volume-weighted regression channel that frames mean-reversion ("fade") setups and then gates, scores and validates them — so you can see whether fading stretched price actually has an edge on your symbol and timeframe instead of taking it on faith. It is a study / analysis framework, not a strategy and not a signal service.
Why these components are combined (the mashup rationale)
Fading an extreme asks three different questions, and no single classic tool answers all three. AVRC coordinates several non-redundant lenses on one shared geometry (an anchored regression channel) and one shared volatility unit (residual σ), so each lens can check the others rather than echoing it:
WHERE is price stretched? A volume-weighted regression centerline with residual-σ bands is drawn against a slower macro trend-relative volume map — volume binned by σ-distance from a longer regression line. Whether the tactical band sits in a thin (fast-traversed) or thick (heavily-traded) macro node tells you if a fade is likely clean or absorbed. This cross-read is the connective tissue between the two layers.
Is a reversion ACTUALLY firing? Independent "tells" at the band — a close-back rejection, a band-confluent momentum divergence, an equal-high/low liquidity sweep, and the macro-density read. Because these tells are correlated, their agreement is shrunk by a design-effect correction so echoes can't masquerade as independent confirmation.
Is the market in a reverting STATE? A regime gate (variance-ratio test + a reversion-trust correlation) only passes fades when price increments offset rather than compound. An entry-time ride-risk score (macro-trend alignment, the two-centerline spread, an already-walking band, momentum, mean-reversion half-life, and multi-timeframe trend consensus) flags fades likely to be "walked" rather than reverted.
The components share one geometry and one volatility unit, and each can veto the others. The goal is to suppress low-quality fades more than to generate them.
How the validation layer works (what makes this more than a drawing)
Every fade is logged and, a fixed horizon later, resolved: its forward return is measured in ATR units and tabulated Gate ON vs Gate OFF — follow-through %, a Wilson 95% interval, whipsaw %, and mean R per fade. Outcomes are additionally split Reverted vs Rode, by macro node (thin/thick), and by ride-risk (low/high at the running median). The panel's Edge line synthesizes this into a single read: is Gate ON's follow-through interval clearing the ungated baseline with positive mean R and enough samples? Per-fade rows also export to the Data Window for your own analysis. Every filter has to earn its place against the ungated baseline.
How to use it
Set the Price source (top of settings). Defaults are tuned for an intraday index future; the source is user-selectable so the framework runs on any symbol or market. Volume-based parts (heatmap, profile, POC) need a real volume feed.
Read the panel top-down: Now (live setup) → State (regime + spread + compression) → the A/B scoreboard (Gate OFF, Gate ON, Revert, Ride) → Edge verdict.
A fade arms when price tags the outer band and at least one tell prints, then passes only if the regime (and optional ride-risk) gate agrees. Target is the centerline or the nearest untested POC.
If Gate ON does not beat Gate OFF on follow-through and mean R with non-overlapping intervals and enough samples, the edge isn't there on this symbol/timeframe — change them, don't force it. The signal is clearest on higher intraday timeframes; 1-minute is mostly noise.
What is original here
The original work is the coordination: a shared-σ, timeframe-adaptive regression channel used as a reversion frame; a trend-relative volume map cross-read against the band; decorrelated tells fused by a design-effect shrink; a statistical regime gate; an entry-time ride-risk score; and a built-in A/B + forward-return validation harness — combined so each lens can veto the others and the tool reports its own hit rate. It is not a re-skin of any single indicator.
Concept credits (techniques are standard; this implementation is original)
Volume-weighted least-squares & polynomial regression; residual-σ channels; anchored VWAP (all standard); Volume Profile / Value Area / Point of Control — Market Profile (Steidlmayer / CBOT); Variance-Ratio test — Lo & MacKinlay (1988); design effect / effective sample size — Kish (1965); proportion confidence interval — Wilson (1927); mean-reversion half-life — Ornstein–Uhlenbeck process; ATR, RSI, Parabolic SAR — Wilder; Stochastic — Lane; Supertrend (classic, MTF context). Builds on established open-source regression-channel and anchored-VWAP techniques.
Settings (all defaults are on; tuned for an intraday index future)
Data/Source · Volatility unit · Macro volume heatmap · Tactical channel & bands · Interrelation & band-walk · Ride-risk filter · Density cross-read · Reversion tells · Regime gate · POC targets · Fade signal · Validation & export · Dashboard & theme (auto light/dark) · MTF trend context. The two signal-suppression gates (walk-gate, ride-gate) ship off so the indicator shows its signals and lets the validation panel tell you whether enabling them helps.
Disclaimer
For research and education only. NOT financial advice, NOT a recommendation, and NOT a guarantee of future results. All statistics shown are in-sample on loaded history, close-to-close at the horizon, without costs or slippage — a study aid, not a backtest. Mean reversion fails in trends and during regime breaks. Do your own research and manage your own risk.
What it is
AVRC is an anchored, volume-weighted regression channel that frames mean-reversion ("fade") setups and then gates, scores and validates them — so you can see whether fading stretched price actually has an edge on your symbol and timeframe instead of taking it on faith. It is a study / analysis framework, not a strategy and not a signal service.
Why these components are combined (the mashup rationale)
Fading an extreme asks three different questions, and no single classic tool answers all three. AVRC coordinates several non-redundant lenses on one shared geometry (an anchored regression channel) and one shared volatility unit (residual σ), so each lens can check the others rather than echoing it:
WHERE is price stretched? A volume-weighted regression centerline with residual-σ bands is drawn against a slower macro trend-relative volume map — volume binned by σ-distance from a longer regression line. Whether the tactical band sits in a thin (fast-traversed) or thick (heavily-traded) macro node tells you if a fade is likely clean or absorbed. This cross-read is the connective tissue between the two layers.
Is a reversion ACTUALLY firing? Independent "tells" at the band — a close-back rejection, a band-confluent momentum divergence, an equal-high/low liquidity sweep, and the macro-density read. Because these tells are correlated, their agreement is shrunk by a design-effect correction so echoes can't masquerade as independent confirmation.
Is the market in a reverting STATE? A regime gate (variance-ratio test + a reversion-trust correlation) only passes fades when price increments offset rather than compound. An entry-time ride-risk score (macro-trend alignment, the two-centerline spread, an already-walking band, momentum, mean-reversion half-life, and multi-timeframe trend consensus) flags fades likely to be "walked" rather than reverted.
The components share one geometry and one volatility unit, and each can veto the others. The goal is to suppress low-quality fades more than to generate them.
How the validation layer works (what makes this more than a drawing)
Every fade is logged and, a fixed horizon later, resolved: its forward return is measured in ATR units and tabulated Gate ON vs Gate OFF — follow-through %, a Wilson 95% interval, whipsaw %, and mean R per fade. Outcomes are additionally split Reverted vs Rode, by macro node (thin/thick), and by ride-risk (low/high at the running median). The panel's Edge line synthesizes this into a single read: is Gate ON's follow-through interval clearing the ungated baseline with positive mean R and enough samples? Per-fade rows also export to the Data Window for your own analysis. Every filter has to earn its place against the ungated baseline.
How to use it
Set the Price source (top of settings). Defaults are tuned for an intraday index future; the source is user-selectable so the framework runs on any symbol or market. Volume-based parts (heatmap, profile, POC) need a real volume feed.
Read the panel top-down: Now (live setup) → State (regime + spread + compression) → the A/B scoreboard (Gate OFF, Gate ON, Revert, Ride) → Edge verdict.
A fade arms when price tags the outer band and at least one tell prints, then passes only if the regime (and optional ride-risk) gate agrees. Target is the centerline or the nearest untested POC.
If Gate ON does not beat Gate OFF on follow-through and mean R with non-overlapping intervals and enough samples, the edge isn't there on this symbol/timeframe — change them, don't force it. The signal is clearest on higher intraday timeframes; 1-minute is mostly noise.
What is original here
The original work is the coordination: a shared-σ, timeframe-adaptive regression channel used as a reversion frame; a trend-relative volume map cross-read against the band; decorrelated tells fused by a design-effect shrink; a statistical regime gate; an entry-time ride-risk score; and a built-in A/B + forward-return validation harness — combined so each lens can veto the others and the tool reports its own hit rate. It is not a re-skin of any single indicator.
Concept credits (techniques are standard; this implementation is original)
Volume-weighted least-squares & polynomial regression; residual-σ channels; anchored VWAP (all standard); Volume Profile / Value Area / Point of Control — Market Profile (Steidlmayer / CBOT); Variance-Ratio test — Lo & MacKinlay (1988); design effect / effective sample size — Kish (1965); proportion confidence interval — Wilson (1927); mean-reversion half-life — Ornstein–Uhlenbeck process; ATR, RSI, Parabolic SAR — Wilder; Stochastic — Lane; Supertrend (classic, MTF context). Builds on established open-source regression-channel and anchored-VWAP techniques.
Settings (all defaults are on; tuned for an intraday index future)
Data/Source · Volatility unit · Macro volume heatmap · Tactical channel & bands · Interrelation & band-walk · Ride-risk filter · Density cross-read · Reversion tells · Regime gate · POC targets · Fade signal · Validation & export · Dashboard & theme (auto light/dark) · MTF trend context. The two signal-suppression gates (walk-gate, ride-gate) ship off so the indicator shows its signals and lets the validation panel tell you whether enabling them helps.
Disclaimer
For research and education only. NOT financial advice, NOT a recommendation, and NOT a guarantee of future results. All statistics shown are in-sample on loaded history, close-to-close at the horizon, without costs or slippage — a study aid, not a backtest. Mean reversion fails in trends and during regime breaks. Do your own research and manage your own risk.
Açık kaynak kodlu komut dosyası
Gerçek TradingView ruhuyla, bu komut dosyasının mimarı, yatırımcıların işlevselliğini inceleyip doğrulayabilmesi için onu açık kaynaklı hale getirdi. Yazarı tebrik ederiz! Ücretsiz olarak kullanabilseniz de, kodu yeniden yayınlamanın Topluluk Kurallarımıza tabi olduğunu unutmayın.
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.
Açık kaynak kodlu komut dosyası
Gerçek TradingView ruhuyla, bu komut dosyasının mimarı, yatırımcıların işlevselliğini inceleyip doğrulayabilmesi için onu açık kaynaklı hale getirdi. Yazarı tebrik ederiz! Ücretsiz olarak kullanabilseniz de, kodu yeniden yayınlamanın Topluluk Kurallarımıza tabi olduğunu unutmayın.
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.