We analyze a broad set of buy-side trading biases identified in leading finance journals over the past 75 years. Using 14.5 years of trade-level data for retail and institutional investors, we find that roughly 20 percent of documented biases are no more prevalent among retail investors than would be expected from random trading, while retail investors display about three times as many biases as institutions. Among retail trades, each additional bias is associated with a 128-basis-point lower return over the following 12 months, and this underperformance largely reflects within-investor behavior. Attention-related biases are particularly costly. The incidence and cost of biases vary over time, becoming more common and more detrimental during periods of heightened market volatility. Female investors exhibit more biases per trade, especially those linked to familiarity and peer influence. Overall, our results provide a unified empirical framework quantifying the magnitude, persistence, and dynamics of retail trading biases and their performance consequences