1. Quick summary (one-line)

Each signal object represents a coin detected by the screener with raw metrics (price, volume, RSI), derived analytics (health_score, trend_strength, momentum_phase), and a smart_confidence score which the enggine uses to rank signals. Use health_score and trend_strength for quality checks; follow risk_level and momentum_phase for tactical actions.

2. Example signal (JSON)

{
  "symbol": "ABC",
  "name": "ABC Token",
  "current_price": 0.1234,
  "current_score": 48,
  "enhanced_score": 62.1,
  "price_change_1h": 3.2,
  "price_change_4h": 8.5,
  "price_change_24h": 12.4,
  "market_cap": 25000000,
  "total_volume": 1500000,
  "volume_spike_ratio": 2.3,
  "volume_acceleration": 0.18,
  "volume_consistency": 0.82,
  "volume_surge": 1.6,
  "rsi_fast": 78.2,
  "rsi_slow": 56.7,
  "rsi_delta": 21.5,
  "momentum_regime": "STRONG_BULL",
  "momentum_phase": "ACCELERATION",
  "smart_confidence": 78,
  "health_score": 84,
  "trend_strength": 72,
  "risk_level": "LOW",
  "appearance_count": 4,
  "performance_since_first": 18.3,
  "score_improvement": 14.1,
  "smart_alerts": ["💎 Excellent health score", "⚡ Strong momentum with confirmation"],
  "order_blocks_count": 2,
  "order_block_alignment": "NEAR_SUPPORT",
  "order_block_score": 0.72,
  "multi_tf_confirmation": "BULL",
  "binance_ratio": 0.4,
  "liquidity_score": 0.7,
  "market_context_score": 0.62
}

This is a condensed example of the structure produced by get_display_data().

3. Field-by-field explanation & interpretation

4. How the enggine decides priority / urgency

The enggine computes a combined priority using enhanced_score, smart_confidence, health_score, trend_strength, absolute performance and appearance count. The aggregator returns a category (e.g. 🔥 HIGHEST, 🚀 HIGH, 📈 MEDIUM-HIGH, ...). Use HIGHEST/HIGH to trigger immediate attention and possible allocation; MEDIUM to watch; LOW to ignore.

5. Practical reading — decision rules (examples)

1. Buy candidate (conservative):
   • health_score > 75 AND trend_strength > 60
   • smart_confidence > 65 AND multi_tf_confirmation == "BULL"
   • order_block_alignment == "NEAR_SUPPORT" preferred
2. Speculative entry (higher risk):
   • enhanced_score > 65 AND volume_spike_ratio > 2.0 AND health_score > 60
   • If momentum_phase == "PARABOLIC" → consider taking partial profits or setting tight SL
3. Avoid / exit:
   • risk_level == "HIGH" OR health_score < 45
   • performance_since_first > 40% with momentum_phase == "DISTRIBUTION" (possible top)

These are examples — adapt thresholds to your strategy and position sizing rules.

6. Where values come from in the code (reference)

• Display payload / DB fields: function get_display_data(signal) composes all display fields.
• Health score: computed in _calculate_enhanced_health_score — weighted factors: performance, score improvement, volume trend, momentum trend, price trend, appearance_count, consistency.
• Trend strength: computed in _calculate_enhanced_trend_strength, uses performance magnitude, volume strength, momentum strength, RSI delta, appearance strength.
• Smart confidence: computed in calculate_enhanced_smart_confidence — aggregates market_context, volume_profile, momentum regime, order blocks, etc.

7. Quick troubleshooting

• Missing fields: If a signal lacks a field, check the data pipeline step that extracts klines/volumes — default fallbacks exist in the code (e.g. default RSI=50, volume_spike=1).
• Low confidence even with pump: Market context (BTC trend & Fear&Greed) can reduce smart_confidence. See MarketContextAnalyzer.
• Order block detection noisy: OrderBlockLiteAnalyzer requires min lookback; if klines < threshold it returns empty. Increase lookback data.
• Signals not showing in UI: The screener saves UI signals into self.last_ui_signals and DB via the database manager — check DB insert logs for errors.

8. Glossary (short)

• Health score: composite gauge of signal robustness (0–100).
• Trend strength: how strong the detected trend is (0–100).
• Smart confidence: model confidence combining market, volume and orderblock signals.
• Order block: historical area of price rejection/absorption (demand/supply).
• Momentum phase: lifecycle stage of the move (guides trade urgency).

1. Overview of the Performance Tracking System

The performance engine records every coin from the moment it first appears in the screener. Over time, it updates:

• price evolution and percent change
• score progression
• momentum phase transitions
• appearance frequency
• volume and trend behaviors

This forms the tracking_performance object inside get_display_data(), and is used by dashboard data, and priority ranking.

2. Example – Tracking Performance JSON

{
  "first_price": 0.091,
  "current_price": 0.127,
  "performance_since_first": 39.56,
  "score_improvement": 18.2,
  "appearance_count": 5,
  "normalized_performance": 0.68,
  "momentum_phase": "ACCELERATION",
  "alerts": [
    "🔥 Strong normalized growth",
    "📈 Momentum accelerating",
    "⭐ Multiple reappearances confirm trend"
  ]
}

3. Field-by-Field Explanation & Interpretation

4. How Momentum Phases Are Determined

The enggine uses a multi-factor approach:

• normalized_performance (0–1 scale)
• score_improvement trend
• appearance_count growth
• volume & RSI behavior

Phase Rules

• ACCUMULATION: normalized_performance < 0.20 and appearance_count ≤ 2
• ACCELERATION: 0.20–0.55 normalized_performance AND positive score_improvement
• PARABOLIC: normalized_performance > 0.55 OR performance_since_first > 60%
  High risk of reversal
• DISTRIBUTION: performance_since_first high but score_improvement declining
• CAPITULATION: Strong negative performance or severe volume drop

5. Alerts Generated by the Tracking Engine

The enggine creates alerts based on performance context to help users quickly understand what is happening.

• 🔥 Strong normalized growth — performance rising smoothly
• ⚡ Acceleration detected — score up + performance up together
• 📈 Trend continuation likely — multiple reappearances
• 🚨 Overextended / parabolic — potential top
• ⚠️ Distribution phase — momentum slowing
• 💀 Capitulation risk — collapse likely

Alerts combine performance, appearance_count, and volatility conditions to create accurate warnings.

6. Decision Guide — How to Use Tracking in Trades

Best Opportunities (High Probability)

• ACCUMULATION + low volatility
• ACCELERATION + rising score
• appearance_count ≥ 3

Short-Term Scalping Zones

• PARABOLIC
• performance_since_first > 40%
• Telegram may send “Trend Still Climbing” updates

Risky or Exit Zones

• DISTRIBUTION
• CAPITULATION
• score_improvement turning negative

7. How the enggine Stores and Updates Performance Data

Each time the screener detects the same coin:

1. If first appearance → create tracking entry
2. Else → update fields:
   • current_price
   • performance_since_first
   • score_improvement
   • appearance_count
   • momentum_phase
   • alerts
3. Store updated snapshot to database

This allows historical tracking even if the enggine restarts.