This paper introduces IvyDB TradeFlow, a framework for classifying options trades as buyer- or seller-initiated and analyzing how that activity varies by trade type, price, and exchange venue. Using SPY options on June 17, 2026, the day of an FOMC decision, it shows how trade condition codes and venue data improve on a simple bid–ask classification method. Midpoint trades accounted for 10.2% of the day’s volume, and TradeFlow assigned a direction to 99.9% of that volume. The framework also separates retail and aggressive institutional activity and provides data for studying order imbalance, execution costs, and changes in trading behavior around market events.