Velocity Bands

Velocity Bands creates dynamic deviation-based support and resistance levels that adapt to market conditions. Using outlier-filtered standard deviation calculations, it identifies extreme price conditions and provides signals for mean reversion and reversal setups.
How It Works
Velocity Bands calculates:
A weighted moving average (WMA) as the center line
Normalized price deviation from the average
Outlier filtering to remove extreme spikes
Standard deviation bands at +1/+2 and -1/-2 levels
The bands dynamically adjust to market volatility, expanding in volatile conditions and contracting in calm markets.
Key Features
Dynamic Bands
Velocity Bands creates four levels:
+2 Band (Upper Outer): Extreme overbought level
+1 Band (Upper Inner): Moderate overbought level
-1 Band (Lower Inner): Moderate oversold level
-2 Band (Lower Outer): Extreme oversold level
The center line (mean) is displayed as a dotted line, providing a reference point.
Outlier Filtering
Before calculating bands, Velocity Bands filters out outliers (price moves beyond 4 standard deviations). This ensures the bands aren't distorted by extreme spikes or data errors, creating more stable and reliable levels.
Gradient Fills
The bands include gradient fills:
Base fills: Subtle gradients between +1/+2 and -1/-2 bands
Enhanced fills: When price is outside the bands, enhanced gradients highlight the extreme condition
Re-entry Signals
When price returns from extreme zones:
▼ (Down Arrow): Price re-enters from above the +2 band (bearish mean reversion)
▲ (Up Arrow): Price re-enters from below the -2 band (bullish mean reversion)
These signals suggest price is returning to normal levels after an extreme move.
Rejection Signals
When price enters a band but fails to hold:
Rejection from upper band: Price entered the upper band but closed below it (bearish reversal)
Rejection from lower band: Price entered the lower band but closed above it (bullish reversal)
These signals indicate potential reversals when price can't sustain moves into the bands.
Last updated
Was this helpful?