Directional Probability + Trajectory (AI)

Directional Probability + Trajectory (AI) is an adaptive artificial intelligence script that analyzes
real-time price action to estimate directional probability and project the most likely future price trajectory directly on your chart. It continuously learns from every candle, scanning historical patterns, calculating probabilities, and drawing the projected path forward — complete with a dynamic confidence zone.

HOW IT WORKS

The model operates in two stages:

1. Pattern Recognition
Scans a configurable lookback window, extracting normalized features from price action. It identifies the most similar historical patterns to the current market structure using distance-based matching.

2. Adaptive Probability Estimation
A self-learning model trains online — updating its internal parameters on every single bar. It estimates the probability of the next candle closing bullish or bearish, and uses that probability to reweight the historical pattern matches. Patterns that agree with the AI's current directional bias receive significantly more weight. Patterns that disagree are suppressed. This creates a trajectory projection that is shaped by both historical evidence AND real-time learned probability.

The result is a projected trajectory that adapts dynamically to changing market conditions.

FEATURES

• Future price trajectory projected on chart
• Dynamic confidence cone (upper/lower bounds)
• Real-time probability estimation (Bullish % / Bearish %)
• Confidence bar — visual gauge of directional conviction
• Adjustable AI influence strength over pattern matching
• Info panel
• Background coloring based on directional signals

INTERPRETATION

Trajectory Line
The main projected line represents the weighted average of the most similar historical outcomes, biased by the AI's real-time directional probability estimate.

Green = Bullish projection
Red = Bearish projection

Confidence Cone
The dashed boundaries represent standard deviations from the mean trajectory. A wider cone indicates higher uncertainty among the matched patterns. A narrow cone suggests strong agreement.

Probability Panel
• Prob. Green — Estimated probability of next candle closing bullish
• Prob. Red — Estimated probability of next candle closing bearish
• Direction — Current AI directional bias
• Confidence — How far the probability is from 50/50 (0-100%)

End-Point Label
Shows the projected price and percentage change at the end of the projection window.

HOW THE AI INFLUENCES DIRECTION

The probability model doesn't just display a signal — it actively reshapes the trajectory projection.

When the AI estimates bullish probability:
→ Historical neighbors that resulted in upward moves receive
amplified weight
→ Bearish neighbors are suppressed

When the AI estimates bearish probability:
→ Downward-moving neighbors are amplified
→ Bullish neighbors are suppressed

BEST PRACTICES
• Works on any timeframe and any asset
• Higher timeframes tend to produce smoother projections
• Increase K Neighbors for smoother trajectories
• Decrease K Neighbors for more reactive projections
• Increase Lookback for more historical data (needs sufficient bars)
• Use the confidence cone width as a measure of projection reliability
• Combine with your own analysis — this is a probability tool, not a guarantee

LIMITATIONS

• Past patterns do not guarantee future results
• The projection updates on every new bar
• Requires sufficient historical bars for the configured lookback
• Best used as a confluence tool alongside other analysis methods