The WNBA’s commercial growth has collided with strict salary caps, leaving front offices without statistical tools to optimize rosters. In this project, we developed a data-driven Fair Market Value (FMV) engine that predicts a player's contract valuation based on on-court production. We engineered lagged performance features from 150 raw variables to mimic front-office decision-making. Using Lasso, SelectKBest, and Random Forest feature selection, we isolated 25 key drivers, identifying start rate and points per game as the most salient indicators. To prevent temporal leakage, we implemented a chronological validation scheme grouped by player identity, reserving 2025 as a holdout set. Our final model is a tuned Random Forest Regressor which achieves an RMSE of $44k, outperforming linear and heuristic baselines. Ultimately, evaluating contracts against an FMV ±15 threshold exposed sharp structural inefficiencies, revealing under-valued rookie contracts and over-valued veterans.