OPEN-SOURCE SCRIPT
Diupdate Auction & Liquidity Command Center Volume Profile, Measured
Auction & Liquidity Command Center — Volume Profile, Measured
The levels traders already use — prior POC, value area, naked POCs, prior day high/low, session AVWAP, HVN/LVN — each scored by its measured reaction on this chart: how often price rejects vs breaks, and what the fade has been worth in R. Levels with evidence, not levels with vibes. Never a buy or sell.
What it does
Every structure tool draws levels. None of them measures what happens when price gets there. This tool builds the session-anchored auction map with profile-grade accuracy, detects qualified touches of every level, resolves each touch through a triple-barrier outcome, and pools the results by level TYPE into a live scoreboard: pPOC +0.01R · rej 50% · n156. You see not just where the levels are, but which kinds of levels have actually meant something on this chart — and which are coin flips.
The components, and why they are combined
This is a deliberate synthesis of four parts, each covering the previous one's weakness:
A profile-grade level engine (Market Profile — J. P. Steidlmayer). Nine level types from the session volume-at-price profile and session extremes: prior POC, prior VAH/VAL (classical two-row 70% expansion), naked POCs (prior POCs never revisited), prior day high/low, the session's anchored VWAP, and HVN/LVN volume nodes (prominence-filtered local extremes). Accuracy choices: each bar's volume is distributed range-proportionally across the rows it overlaps (not binned at one point); POC ties break toward the session center. Weakness left open: a drawn level says nothing about whether it matters.
A qualified-touch detector. A level must be ARMED — price fully away from it by at least k×ATR — before a touch of it can count, and it disarms after every touch. Chop sitting on a line cannot enter the record. Approach direction is stored with every event. Weakness left open: a touch is not an outcome.
Triple-barrier outcome resolution (outcome labelling — M. López de Prado). From each touch: REJECT if price moves m×ATR back the way it came first, BREAK if it moves m×ATR through first, TIMEOUT after T bars. Purity rules: barriers are fixed at the ATR of the touch moment; evaluation starts the bar after the touch; a bar hitting both barriers is a timeout, never a guess. Weakness left open: one level's history is n = 1.
Per-TYPE pooling with honesty gates. Statistics pool by level type, never by individual line — a type is a real sample. A type shows no score until a minimum number of its touches have resolved (default 20); until then it reads BUILDING with its count. Timeouts are reported in n but excluded from the reject/break ratio. Fade expectancy = (rejects − breaks) / (rejects + breaks), in R.
How to read it
Rails are colored and styled by type (solid profile levels, dashed day levels, dotted volume nodes, violet naked POCs); each label carries its type's live score or its BUILDING count.
Evidence on the chart: a gray • at every qualified touch, then ○ (teal) where the touch rejected and ✕ (amber) where it broke. Every number on the scoreboard can be audited against the chart.
Dashboard: nearest level and its score, with a plain-language verdict (tends to hold / coin flip / tends to break) so the read needs no statistics background; per-type scoreboard (fade R · reject % · n) for all nine types; touch counts; the exact engine settings in the NOTE row.
Honest expectations: most types on most charts score near zero — that is the truthful baseline, and seeing it protects you from folklore. The value is in the exceptions this chart's own history reveals (for example, day extremes often carry a modest positive fade expectancy while POC retests are a coin flip), and in knowing the difference.
How to use it
Use the scoreboard to weight your own playbook: give more respect to touches of types that have measured well here, less to types that grade as noise — and size accordingly. The "Touch of a MEASURED level" alert fires only when price reaches a type with a real sample behind it. This is context about where price reactions have had structure — never a direction, never an entry signal.
Non-repaint & universality
Profiles, POC/VA/nodes and day levels commit only at session close on confirmed bars; touches and outcomes resolve on confirmed bars; the AVWAP is cumulative within its session. Nothing repaints. The script requests no external data of any kind — no lower timeframes, no security calls — so it runs identically on every plan and every symbol with volume.
Use on any market
Volume source, profile rows, value-area %, node thresholds, arm distance, barriers and sample gates are all inputs. Defaults suit liquid intraday index futures; intraday timeframes give the engine the most touches to learn from.
Originality & credits
The synthesis — a range-proportional session profile, qualified-touch detection, touch-time-ATR triple-barrier outcomes, and per-type pooled reaction statistics displayed as a live scoreboard — is original work for this publication. Concept credits: Market Profile / point of control / value area — J. Peter Steidlmayer; naked (virgin) POC — market-profile literature; anchored VWAP — as popularised in modern trading literature; triple-barrier outcome labelling — M. López de Prado. Implementation and charting design are the author's own.
Disclaimer
Research and education only. NOT financial advice, NOT a signal service, NOT a guarantee of future results. Reaction statistics are empirical frequencies from this chart's limited history, pooled per level type; they change with regime and sample, and a positive expectancy is not a promise. Validate independently and manage your own risk.
The levels traders already use — prior POC, value area, naked POCs, prior day high/low, session AVWAP, HVN/LVN — each scored by its measured reaction on this chart: how often price rejects vs breaks, and what the fade has been worth in R. Levels with evidence, not levels with vibes. Never a buy or sell.
What it does
Every structure tool draws levels. None of them measures what happens when price gets there. This tool builds the session-anchored auction map with profile-grade accuracy, detects qualified touches of every level, resolves each touch through a triple-barrier outcome, and pools the results by level TYPE into a live scoreboard: pPOC +0.01R · rej 50% · n156. You see not just where the levels are, but which kinds of levels have actually meant something on this chart — and which are coin flips.
The components, and why they are combined
This is a deliberate synthesis of four parts, each covering the previous one's weakness:
A profile-grade level engine (Market Profile — J. P. Steidlmayer). Nine level types from the session volume-at-price profile and session extremes: prior POC, prior VAH/VAL (classical two-row 70% expansion), naked POCs (prior POCs never revisited), prior day high/low, the session's anchored VWAP, and HVN/LVN volume nodes (prominence-filtered local extremes). Accuracy choices: each bar's volume is distributed range-proportionally across the rows it overlaps (not binned at one point); POC ties break toward the session center. Weakness left open: a drawn level says nothing about whether it matters.
A qualified-touch detector. A level must be ARMED — price fully away from it by at least k×ATR — before a touch of it can count, and it disarms after every touch. Chop sitting on a line cannot enter the record. Approach direction is stored with every event. Weakness left open: a touch is not an outcome.
Triple-barrier outcome resolution (outcome labelling — M. López de Prado). From each touch: REJECT if price moves m×ATR back the way it came first, BREAK if it moves m×ATR through first, TIMEOUT after T bars. Purity rules: barriers are fixed at the ATR of the touch moment; evaluation starts the bar after the touch; a bar hitting both barriers is a timeout, never a guess. Weakness left open: one level's history is n = 1.
Per-TYPE pooling with honesty gates. Statistics pool by level type, never by individual line — a type is a real sample. A type shows no score until a minimum number of its touches have resolved (default 20); until then it reads BUILDING with its count. Timeouts are reported in n but excluded from the reject/break ratio. Fade expectancy = (rejects − breaks) / (rejects + breaks), in R.
How to read it
Rails are colored and styled by type (solid profile levels, dashed day levels, dotted volume nodes, violet naked POCs); each label carries its type's live score or its BUILDING count.
Evidence on the chart: a gray • at every qualified touch, then ○ (teal) where the touch rejected and ✕ (amber) where it broke. Every number on the scoreboard can be audited against the chart.
Dashboard: nearest level and its score, with a plain-language verdict (tends to hold / coin flip / tends to break) so the read needs no statistics background; per-type scoreboard (fade R · reject % · n) for all nine types; touch counts; the exact engine settings in the NOTE row.
Honest expectations: most types on most charts score near zero — that is the truthful baseline, and seeing it protects you from folklore. The value is in the exceptions this chart's own history reveals (for example, day extremes often carry a modest positive fade expectancy while POC retests are a coin flip), and in knowing the difference.
How to use it
Use the scoreboard to weight your own playbook: give more respect to touches of types that have measured well here, less to types that grade as noise — and size accordingly. The "Touch of a MEASURED level" alert fires only when price reaches a type with a real sample behind it. This is context about where price reactions have had structure — never a direction, never an entry signal.
Non-repaint & universality
Profiles, POC/VA/nodes and day levels commit only at session close on confirmed bars; touches and outcomes resolve on confirmed bars; the AVWAP is cumulative within its session. Nothing repaints. The script requests no external data of any kind — no lower timeframes, no security calls — so it runs identically on every plan and every symbol with volume.
Use on any market
Volume source, profile rows, value-area %, node thresholds, arm distance, barriers and sample gates are all inputs. Defaults suit liquid intraday index futures; intraday timeframes give the engine the most touches to learn from.
Originality & credits
The synthesis — a range-proportional session profile, qualified-touch detection, touch-time-ATR triple-barrier outcomes, and per-type pooled reaction statistics displayed as a live scoreboard — is original work for this publication. Concept credits: Market Profile / point of control / value area — J. Peter Steidlmayer; naked (virgin) POC — market-profile literature; anchored VWAP — as popularised in modern trading literature; triple-barrier outcome labelling — M. López de Prado. Implementation and charting design are the author's own.
Disclaimer
Research and education only. NOT financial advice, NOT a signal service, NOT a guarantee of future results. Reaction statistics are empirical frequencies from this chart's limited history, pooled per level type; they change with regime and sample, and a positive expectancy is not a promise. Validate independently and manage your own risk.
Catatan Rilis
Auction & Liquidity Command Center v1.1: bounded scoring and net expectancySCORES ARE BOUNDED, NOT ASSERTED
The outcome test is a symmetric triple barrier — from a qualified touch, price must travel the same distance in either direction. If a level type carried no information, rejections and breaks would land fifty-fifty. That makes 50% the honest null, and it makes a bare rejection percentage misleading: 60% on twenty touches carries a 95% interval of roughly 38% to 78%, which straddles the null completely.
Every level type now carries a Wilson 95% interval on its rejection rate, shown alongside the score. More importantly, the verbal call is now decided by the bound rather than the average. A level type "tends to hold" only when the lower bound clears 50%, and "tends to break" only when the upper bound falls below it. Everything in between is reported as a coin flip — which, at realistic sample sizes, is most of the time. That is the honest answer, not a shortcoming of the measurement.
EXPECTANCY IS NET OF COST
With symmetric barriers a rejection pays one R and a break costs one R, so gross expectancy is simply rejections minus breaks over the total. The round trip, however, is paid on winners and losers alike, so it subtracts directly from expectancy rather than scaling it.
The reported figure is now net of a round-trip cost expressed in R, defaulting to 0.10R. The consequence is worth stating plainly: a level type showing plus 0.10R gross is a losing one once a 0.15R round trip is charged. Set the figure to your own execution reality — a tighter outcome barrier makes cost a larger share of each R.
Everything else is unchanged: the range-proportional volume profile, classical value-area construction, qualified-touch arming, session anchoring and non-repaint commitment all behave exactly as before. The script still requests no external data and runs identically on every plan.
Unchanged framing: levels are descriptive structures from this chart's own history, touch records are counts rather than predictions, and this is research and education only — not financial advice, not a signal service.
Skrip open-source
Dengan semangat TradingView yang sesungguhnya, pembuat skrip ini telah menjadikannya sebagai sumber terbuka, sehingga para trader dapat meninjau dan memverifikasi fungsinya. Salut untuk penulisnya! Meskipun Anda dapat menggunakannya secara gratis, perlu diingat bahwa penerbitan ulang kode ini tunduk pada Tata Tertib kami.
Pernyataan Penyangkalan
Informasi dan publikasi ini tidak dimaksudkan, dan bukan merupakan, saran atau rekomendasi keuangan, investasi, trading, atau jenis lainnya yang diberikan atau didukung oleh TradingView. Baca selengkapnya di Ketentuan Penggunaan.
Skrip open-source
Dengan semangat TradingView yang sesungguhnya, pembuat skrip ini telah menjadikannya sebagai sumber terbuka, sehingga para trader dapat meninjau dan memverifikasi fungsinya. Salut untuk penulisnya! Meskipun Anda dapat menggunakannya secara gratis, perlu diingat bahwa penerbitan ulang kode ini tunduk pada Tata Tertib kami.
Pernyataan Penyangkalan
Informasi dan publikasi ini tidak dimaksudkan, dan bukan merupakan, saran atau rekomendasi keuangan, investasi, trading, atau jenis lainnya yang diberikan atau didukung oleh TradingView. Baca selengkapnya di Ketentuan Penggunaan.