📊 Trader Behavior Insights: An Analysis of Performance vs. Market Sentiment

This project analyzes historical trader data from Hyperliquid and the Bitcoin Fear & Greed Index to explore how trader behavior aligns with market sentiment. The goal is to uncover actionable patterns that can lead to smarter, data-driven trading strategies.

---

🧠 Methodology

🔹 Data Loading

• Two primary datasets were used:
  • Hyperliquid trader data (PnL, leverage, side, execution details, etc.)
  • Bitcoin Fear & Greed Index (daily sentiment classification and scores)

🔹 Data Preprocessing

• Dates were standardized to daily granularity to enable accurate merging.
• The Timestamp IST field in the trader dataset was converted to date-only format.
• Both datasets were merged using this date field as a key.
• Numerical fields like Closed PnL and Fee were cleaned and converted to appropriate types.

---

📊 Analysis & Visualizations

📈 1. Fear & Greed Index Over Time

• Visualization: Line chart of the sentiment score over time.
• Insight: Shows how market emotions evolved over the observed period. Peaks and dips indicate shifts in market psychology.

---

🎨 2. Sentiment Classifications Over Time

• Visualization: Scatterplot of sentiment scores color-coded by classification.
• Insight: Highlights distinct sentiment phases and their frequency over time.

---

📊 3. Distribution of Sentiment Classifications

• Visualization: Bar chart showing the count of each sentiment type.
• Insight: Fear and Extreme Fear dominate the dataset, indicating a market frequently in pessimistic phases.

---

💹 4. Average Trader PnL by Market Sentiment

• Visualization: Bar chart of average Closed PnL by sentiment classification.
• Insight:
  • Traders tend to perform better during 'Fear' and 'Extreme Fear'.
  • Performance significantly drops during Greed phases.

---

💸 5. Average Fee Paid by Sentiment

• Visualization: Bar chart of average trading fees by sentiment.
• Insight:
  • Higher fees are observed during Greed and Extreme Greed, indicating increased trading activity and possibly emotional, FOMO-driven trades.

---

🟢🔴 6. Buy vs. Sell: Average PnL by Sentiment

• Visualization: Grouped bar chart comparing Buy vs. Sell performance across sentiments.
• Insight:
  • Buy orders during Fear are consistently profitable.
  • Sell orders during Greed outperform buy orders.
  • Validates a contrarian strategy: "Buy the Fear, Sell the Greed."

---

📈 7. Trader Activity by Market Sentiment

• Visualization: Bar chart of the number of trades executed under each sentiment.
• Insight:
  • Trading activity spikes during both Extreme Fear and Extreme Greed.
  • Traders are most active during emotional extremes, often when markets are most volatile.

---

🧐 Insights Summary

• Traders tend to perform better on average during Fear periods, especially with Buy orders.
• Fees and risk appetite increase significantly during Greed, indicating heightened emotional trading.
• A Contrarian Strategy (Buy during Fear, Sell during Greed) statistically yields better performance.
• Trader activity clusters around emotional extremes, where market participants are most active and potentially most vulnerable.

This analysis demonstrates that market psychology directly affects trader behavior and profitability. Incorporating sentiment-aware strategies can significantly improve risk-adjusted returns.

---

🧭 Strategic Recommendations

• Contrarian Signal Model: Focus on buying in "Extreme Fear" and taking profits or short positions in "Extreme Greed."
• Dynamic Leverage Management: Advise traders to reduce leverage in greedy markets to minimize exposure.
• Trader Education: Use sentiment-driven insights to help traders avoid FOMO and emotional decision-making.

---

📁 Project Files

• historical_trader_data.csv – Raw trader data
• fear_greed_index.csv – Bitcoin sentiment data
• trader_sentiment_analysis_full.ipynb – Full analysis notebook
• visual_1_fear_greed_over_time.png
• visual_2_sentiment_classifications.png
• visual_3_sentiment_distribution.png
• visual_4_avg_pnl_by_sentiment.png
• visual_5_trader_activity_by_sentiment.png 📄 Download Full Analysis Report (PDF)

---

👨‍💻 Author

Ayush Raj
Data Science Project: Trader Behavior and Market Sentiment

---

✅ Thank you for reviewing my project! I look forward to the opportunity to discuss it further.