This paper proposes a prediction method for the thermal characteristics and static and dynamic per formances of tilting pad journal bearings (TPJBs) with ball-socket pivots under four lubricant feeding methods: sin gle orifice (SO), spray bar (SB), spray bar blocker (SBB), and leading edge groove (LEG). The model incorporates pivot stiffness derived from Hertzian contact theory and employs a mixing coefficient to represent the proportion of lubricant carried from the upstream to the downstream pad, excluding leakage. To improve thermal prediction accuracy, we calibrated the mixing coefficients for each feeding method using published experimental temperature data and compared the predictions with those of the test data. With increasing rotor speed, maximum temperature increased, and each lubricant feeding method affected cooling differently. Among the four methods, LEG cooled the bearing most effectively, followed by SB, SBB, and SO. The model showed clear trends in the eccentricity ratio for different oil supply methods. As rotor speed increased, direct stiffness increased, whereas direct damping decreased. These changes were more noticeable depending on the type of lubricant feed, and the results matched the experimental data well. The results clearly show that the method of supplying oil has a strong effect on both the temperature and dynamic behavior of the bearing. Changing the mixing coefficient for each case enhanced model prediction performance under various conditions. This method helps improve the design and operation of TPJBs, especially in high-speed and high-load situations where controlling temperature is very important.