OPEN-SOURCE SCRIPT
VWAP AI - Statistical Bands & Touch Stats [Dots3Red]
⚓ VWAP AI - STATISTICAL BANDS & TOUCH STATS [Dots3Red]
VWAP's standard deviation bands are treated more or less as reliable support and resistance — on faith. This script checks that faith against the actual chart in front of you: every band touch is graded, every break beyond a band is graded, and the results accumulate into a running, honest record.
✨ WHY THIS MATTERS
VWAP tells you the volume-weighted average price — where the "center of gravity" of trading has actually been. The bands around it are meant to show how far price typically wanders from that center before snapping back. But "typically" varies enormously by instrument, session, and market condition, and no plain VWAP tool tells you what's actually been happening on your chart.
This script tracks it directly:
📊 +1σ | 62% rejected (n=41)
That means 41 touches of the +1σ band have been recorded on this chart, and 62% of them resulted in price genuinely rejecting back toward VWAP. Measured history, not an assumption baked into the tool.
⚙️ HOW IT WORKS
⚓ Anchoring — VWAP resets at the start of each new period. Session is the classic intraday default; Week and Month extend the same logic to longer views. Custom Bar anchors once, permanently, to a specific historical point you choose — useful for anchoring to an earnings date, a gap, or any event you want to measure from, rather than the calendar.
📏 Two-tier statistical bands — Band 1 and Band 2 are both standard-deviation multiples of VWAP, computed from a proper running variance (not an ATR approximation). Defaults are ±1σ and ±2σ, both fully adjustable.
🎯 Touch grading — when price wicks into a band without closing beyond it, that's logged as a touch. Within a configurable window, it resolves as:
• Rejection — price moved back toward VWAP by a meaningful distance
• Break — price closed convincingly through the band
• Timeout — neither happened clearly enough to call
🔄 Break-to-reversion tracking — separately, when price actually closes beyond Band 1, the script watches whether that move reverts back toward VWAP or continues away from it. This answers a different question than touch grading: not "did the band hold," but "once it didn't, did price come back anyway?"
🔒 Non-repainting — all grading happens strictly on confirmed bars.
🧭 HOW TO USE
1️⃣ Check the band stats before treating a level as reliable. "+1σ: 71% rejected (n=38)" and "+1σ: 44% rejected (n=12)" look like the same line on the chart but mean very different things about how much to lean on it.
2️⃣ Use break-reversion stats to judge a breakout beyond VWAP's range. If breaks above Band 1 have reverted back 65% of the time on this chart, that's useful context before assuming a fresh breakout will keep running.
3️⃣ Read Price vs VWAP as the simplest possible bias check. Above VWAP means the average buyer today is in profit; below means the average buyer is underwater. It's a blunt but genuinely useful read on crowd positioning.
4️⃣ Let sample sizes build before trusting the percentages. Every stat shows its N= specifically so you can judge reliability yourself — a handful of touches is not yet a pattern.
5️⃣ Match the anchor mode to what you're actually measuring. Session for pure intraday structure, Week or Month for a longer view, Custom Bar when you want to measure from one specific moment forward.
⏱️ WHICH TIMEFRAMES WORK BEST
Session-anchored VWAP is fundamentally an intraday tool — it was built for, and is most meaningful on, timeframes where a full session contains enough bars to form a real distribution: 1-minute through 1-hour is the classic and most effective range, which is exactly where VWAP sees the heaviest institutional and day-trading use.
On daily or weekly charts, a Session anchor resets so frequently relative to the bar size that it stops being meaningful — you'd see very few bars per session. For higher-timeframe or swing-style use, switch the anchor to Week, Month, or Custom Bar instead, so the accumulation window actually spans enough bars to produce a meaningful VWAP and band structure.
The touch and break statistics also need enough occurrences to mean anything — a fast-moving intraday chart will accumulate a useful sample size in days; a slow higher-timeframe anchor will take considerably longer.
🛠️ SETTINGS
⚓ Anchoring — Session / Week / Month / Custom Bar, source price
📏 Bands — Band 1 and Band 2 standard-deviation multipliers, Band 2 visibility toggle
🎯 Touch Statistics — Touch Tolerance, Rejection Distance, Reversion Distance, Outcome Window
🎨 Visualization — independent Band 1 / Band 2 touch marker toggles, Dot or Triangle marker style, marker size, VWAP and band line widths, independent fill transparency per band tier
🎨 Colors — VWAP line, Band 1 lines, Band 2 lines, upper/lower touch markers, Price Above/Below VWAP indicator, and full dashboard color control (background, border, header, row styling)
🖥️ Dashboard — show/hide, position — current VWAP value, price position, all four band stats, and both break-reversion stats in one place
📝 NOTES
Statistics accumulate from when the indicator is added to the chart and reset only when explicitly cleared by reloading. A Custom Bar anchor never resets on its own, it measures continuously from the point you chose. Band 2 statistics take meaningfully longer to build a useful sample than Band 1, simply because price reaches ±2σ far less often than ±1σ.
⚠️ DISCLAIMER
This is an analytical and visualization tool. It does not generate trade signals and does not constitute financial advice. Historical rejection and reversion rates do not guarantee future performance.
VWAP's standard deviation bands are treated more or less as reliable support and resistance — on faith. This script checks that faith against the actual chart in front of you: every band touch is graded, every break beyond a band is graded, and the results accumulate into a running, honest record.
✨ WHY THIS MATTERS
VWAP tells you the volume-weighted average price — where the "center of gravity" of trading has actually been. The bands around it are meant to show how far price typically wanders from that center before snapping back. But "typically" varies enormously by instrument, session, and market condition, and no plain VWAP tool tells you what's actually been happening on your chart.
This script tracks it directly:
📊 +1σ | 62% rejected (n=41)
That means 41 touches of the +1σ band have been recorded on this chart, and 62% of them resulted in price genuinely rejecting back toward VWAP. Measured history, not an assumption baked into the tool.
⚙️ HOW IT WORKS
⚓ Anchoring — VWAP resets at the start of each new period. Session is the classic intraday default; Week and Month extend the same logic to longer views. Custom Bar anchors once, permanently, to a specific historical point you choose — useful for anchoring to an earnings date, a gap, or any event you want to measure from, rather than the calendar.
📏 Two-tier statistical bands — Band 1 and Band 2 are both standard-deviation multiples of VWAP, computed from a proper running variance (not an ATR approximation). Defaults are ±1σ and ±2σ, both fully adjustable.
🎯 Touch grading — when price wicks into a band without closing beyond it, that's logged as a touch. Within a configurable window, it resolves as:
• Rejection — price moved back toward VWAP by a meaningful distance
• Break — price closed convincingly through the band
• Timeout — neither happened clearly enough to call
🔄 Break-to-reversion tracking — separately, when price actually closes beyond Band 1, the script watches whether that move reverts back toward VWAP or continues away from it. This answers a different question than touch grading: not "did the band hold," but "once it didn't, did price come back anyway?"
🔒 Non-repainting — all grading happens strictly on confirmed bars.
🧭 HOW TO USE
1️⃣ Check the band stats before treating a level as reliable. "+1σ: 71% rejected (n=38)" and "+1σ: 44% rejected (n=12)" look like the same line on the chart but mean very different things about how much to lean on it.
2️⃣ Use break-reversion stats to judge a breakout beyond VWAP's range. If breaks above Band 1 have reverted back 65% of the time on this chart, that's useful context before assuming a fresh breakout will keep running.
3️⃣ Read Price vs VWAP as the simplest possible bias check. Above VWAP means the average buyer today is in profit; below means the average buyer is underwater. It's a blunt but genuinely useful read on crowd positioning.
4️⃣ Let sample sizes build before trusting the percentages. Every stat shows its N= specifically so you can judge reliability yourself — a handful of touches is not yet a pattern.
5️⃣ Match the anchor mode to what you're actually measuring. Session for pure intraday structure, Week or Month for a longer view, Custom Bar when you want to measure from one specific moment forward.
⏱️ WHICH TIMEFRAMES WORK BEST
Session-anchored VWAP is fundamentally an intraday tool — it was built for, and is most meaningful on, timeframes where a full session contains enough bars to form a real distribution: 1-minute through 1-hour is the classic and most effective range, which is exactly where VWAP sees the heaviest institutional and day-trading use.
On daily or weekly charts, a Session anchor resets so frequently relative to the bar size that it stops being meaningful — you'd see very few bars per session. For higher-timeframe or swing-style use, switch the anchor to Week, Month, or Custom Bar instead, so the accumulation window actually spans enough bars to produce a meaningful VWAP and band structure.
The touch and break statistics also need enough occurrences to mean anything — a fast-moving intraday chart will accumulate a useful sample size in days; a slow higher-timeframe anchor will take considerably longer.
🛠️ SETTINGS
⚓ Anchoring — Session / Week / Month / Custom Bar, source price
📏 Bands — Band 1 and Band 2 standard-deviation multipliers, Band 2 visibility toggle
🎯 Touch Statistics — Touch Tolerance, Rejection Distance, Reversion Distance, Outcome Window
🎨 Visualization — independent Band 1 / Band 2 touch marker toggles, Dot or Triangle marker style, marker size, VWAP and band line widths, independent fill transparency per band tier
🎨 Colors — VWAP line, Band 1 lines, Band 2 lines, upper/lower touch markers, Price Above/Below VWAP indicator, and full dashboard color control (background, border, header, row styling)
🖥️ Dashboard — show/hide, position — current VWAP value, price position, all four band stats, and both break-reversion stats in one place
📝 NOTES
Statistics accumulate from when the indicator is added to the chart and reset only when explicitly cleared by reloading. A Custom Bar anchor never resets on its own, it measures continuously from the point you chose. Band 2 statistics take meaningfully longer to build a useful sample than Band 1, simply because price reaches ±2σ far less often than ±1σ.
⚠️ DISCLAIMER
This is an analytical and visualization tool. It does not generate trade signals and does not constitute financial advice. Historical rejection and reversion rates do not guarantee future performance.
สคริปต์แบบโอเพนซอร์ส
ด้วยเจตนารมณ์หลักของ TradingView ผู้สร้างสคริปต์นี้ได้ทำให้เป็นโอเพนซอร์ส เพื่อให้เทรดเดอร์สามารถตรวจสอบและยืนยันฟังก์ชันการทำงานของมันได้ ขอชื่นชมผู้เขียน! แม้ว่าคุณจะใช้งานได้ฟรี แต่โปรดจำไว้ว่าการเผยแพร่โค้ดซ้ำจะต้องเป็นไปตาม กฎระเบียบการใช้งาน ของเรา
Connect the dots to elevate your trading.
คำจำกัดสิทธิ์ความรับผิดชอบ
ข้อมูลและบทความไม่ได้มีวัตถุประสงค์เพื่อก่อให้เกิดกิจกรรมทางการเงิน, การลงทุน, การซื้อขาย, ข้อเสนอแนะ หรือคำแนะนำประเภทอื่น ๆ ที่ให้หรือรับรองโดย TradingView อ่านเพิ่มเติมใน ข้อกำหนดการใช้งาน
สคริปต์แบบโอเพนซอร์ส
ด้วยเจตนารมณ์หลักของ TradingView ผู้สร้างสคริปต์นี้ได้ทำให้เป็นโอเพนซอร์ส เพื่อให้เทรดเดอร์สามารถตรวจสอบและยืนยันฟังก์ชันการทำงานของมันได้ ขอชื่นชมผู้เขียน! แม้ว่าคุณจะใช้งานได้ฟรี แต่โปรดจำไว้ว่าการเผยแพร่โค้ดซ้ำจะต้องเป็นไปตาม กฎระเบียบการใช้งาน ของเรา
Connect the dots to elevate your trading.
คำจำกัดสิทธิ์ความรับผิดชอบ
ข้อมูลและบทความไม่ได้มีวัตถุประสงค์เพื่อก่อให้เกิดกิจกรรมทางการเงิน, การลงทุน, การซื้อขาย, ข้อเสนอแนะ หรือคำแนะนำประเภทอื่น ๆ ที่ให้หรือรับรองโดย TradingView อ่านเพิ่มเติมใน ข้อกำหนดการใช้งาน