Coated bubbles arise in flotation columns and solvent extraction contactors; however, no unified correlation predicts their drag coefficient from dimensionless parameters across multiple fluid systems. This work develops an empirical power-law correlation from 2,268 observations spanning three continuous phases (water, 26 wt% NaCl, 30 wt% sucrose) and two organic phases (kerosene and sunflower oil), covering viscosity ratios from 0.34 to 58 and Reynolds numbers from 56 to 894. Drag modification by oil coating depends critically on the viscosity ratio: coated bubbles exhibit higher drag than uncoated bubbles in water and salt systems but lower drag in the sucrose system, where the coating is less viscous than the continuous phase. Three power-law correlations were developed: an uncoated bubble model (Mean Absolute Percentage Error (MAPE) = 1.2%), a coated-bubble model (MAPE = 6.9%), and a unified model (MAPE = 2%) that reduces to the uncoated bubble model when no coating is present. The Reynolds and Eötvös numbers are the dominant parameters; Morton number, coating fraction, and viscosity ratio are secondary within the tested parameter window (0.34 ≤ η ≤ 58). Validation against 25 independent external datasets (2,882 uncoated bubble and 28 coated-bubble measurements, 2,910 total) yielded a mean R² = 0.97 and a MAPE of 5–6% for the uncoated bubble subset; coated-bubble external validation, while consistent with model predictions, is based on a limited number of data points and should be interpreted as preliminary. The model is valid for 30 < Re < 1000 and 0.1 < Eo < 10, providing empirical drag closure for drift-flux and multiphase flow models applied to single-bubble systems under quiescent conditions within this parameter window.