OPEN-SOURCE SCRIPT
Adaptive ML VWAP v1.0
Overview
Adaptive ML VWAP is a next-generation "Smart Indicator" that moves beyond static deviations (Standard Deviation). Instead of assuming market volatility is distributed normally (Bell Curve), this indicator uses a k-Nearest Neighbors (k-NN) machine learning engine to learn the specific volatility behavior of the asset you are trading.
It answers the question: "When price extends away from VWAP, how far does it actually go before reversing?"
The Adaptive ML Engine
This script features a 5-Dimensional ML Engine that tracks every major extension or pullback event. It records:
Deviation Depth (Normalized to ATR)
Trend Slope (Is the trend steep or flat?)
ADX (Trend Strength)
VWAP Deviation (Relative Position)
Time of Day (Session Context)
When a new setup occurs, the k-NN engine instantly searches its memory for the 5 most similar historical events and calculates the probability of success based on what happened last time.
Two Strategy Modes
You can toggle the logic to suit your trading style:
1. Mean Reversion Mode (Default)
"Fade The Move"
Goal: Catch price at an exhaustion point returning to VWAP.
Signal: Triggers when price touches a Smart Band and reverses back toward the center.
k-NN Learning: Learns which conditions favor a snap-back.
Best For: Ranging markets, Lunch hours, Choppy sessions.
2. Trend Following Mode
"Ride The Move"
Goal: Catch breakouts that are launching away from value.
Signal: Triggers when price breaks out of the Inner Band (1.0).
k-NN Learning: Learns which breakouts tend to extend to the Outer Bands.
Best For: Morning Drives, News Events, Strong Trends.
Visual Guide
The indicator uses a Dynamic Gradient system to visualize risk/reward:
Cyan Mist (0.5 - 1.0): The Value Zone. Noise area. Safe for trend entries.
Deep Cyan (1.0 - 2.0): The Trend Zone. Price is moving proactively.
Orange Glow (2.0 - 3.0): The Danger Zone. Price is statistically overextended. Reversals are highly probable here.
"Fractal" Math
Unlike standard indicators that break when you change timeframes, Adaptive ML VWAP uses Fractal Normalization.
A "2.0 Band" on a 15-second chart means the same statistical extreme as a "2.0 Band" on a 4-hour chart.
Auto-Adaptive Lookback: The indicator automatically boosts the ML memory (Lookback) on lower timeframes (seconds/minutes) where more noise requires larger sample sizes, ensuring robust predictions without manual tweaking.
Settings
Auto-Adapting Lookback: (Default: True) automatically increases Lookback to 100+ for seconds charts and 50+ for minute charts.
Lookback (Events): Manual override base value (Default: 100).
Strategy Mode: Toggle between Mean Reversion and Trend Following.
k-Neighbors: The number of similar past events to structurally compare (Default: 5).
Disclaimer: This tool is for educational purposes. Machine learning performance is dependent on market conditions and historical recursion.
Adaptive ML VWAP is a next-generation "Smart Indicator" that moves beyond static deviations (Standard Deviation). Instead of assuming market volatility is distributed normally (Bell Curve), this indicator uses a k-Nearest Neighbors (k-NN) machine learning engine to learn the specific volatility behavior of the asset you are trading.
It answers the question: "When price extends away from VWAP, how far does it actually go before reversing?"
The Adaptive ML Engine
This script features a 5-Dimensional ML Engine that tracks every major extension or pullback event. It records:
Deviation Depth (Normalized to ATR)
Trend Slope (Is the trend steep or flat?)
ADX (Trend Strength)
VWAP Deviation (Relative Position)
Time of Day (Session Context)
When a new setup occurs, the k-NN engine instantly searches its memory for the 5 most similar historical events and calculates the probability of success based on what happened last time.
Two Strategy Modes
You can toggle the logic to suit your trading style:
1. Mean Reversion Mode (Default)
"Fade The Move"
Goal: Catch price at an exhaustion point returning to VWAP.
Signal: Triggers when price touches a Smart Band and reverses back toward the center.
k-NN Learning: Learns which conditions favor a snap-back.
Best For: Ranging markets, Lunch hours, Choppy sessions.
2. Trend Following Mode
"Ride The Move"
Goal: Catch breakouts that are launching away from value.
Signal: Triggers when price breaks out of the Inner Band (1.0).
k-NN Learning: Learns which breakouts tend to extend to the Outer Bands.
Best For: Morning Drives, News Events, Strong Trends.
Visual Guide
The indicator uses a Dynamic Gradient system to visualize risk/reward:
Cyan Mist (0.5 - 1.0): The Value Zone. Noise area. Safe for trend entries.
Deep Cyan (1.0 - 2.0): The Trend Zone. Price is moving proactively.
Orange Glow (2.0 - 3.0): The Danger Zone. Price is statistically overextended. Reversals are highly probable here.
"Fractal" Math
Unlike standard indicators that break when you change timeframes, Adaptive ML VWAP uses Fractal Normalization.
A "2.0 Band" on a 15-second chart means the same statistical extreme as a "2.0 Band" on a 4-hour chart.
Auto-Adaptive Lookback: The indicator automatically boosts the ML memory (Lookback) on lower timeframes (seconds/minutes) where more noise requires larger sample sizes, ensuring robust predictions without manual tweaking.
Settings
Auto-Adapting Lookback: (Default: True) automatically increases Lookback to 100+ for seconds charts and 50+ for minute charts.
Lookback (Events): Manual override base value (Default: 100).
Strategy Mode: Toggle between Mean Reversion and Trend Following.
k-Neighbors: The number of similar past events to structurally compare (Default: 5).
Disclaimer: This tool is for educational purposes. Machine learning performance is dependent on market conditions and historical recursion.
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.