AC contactors are widely used in power systems. The accurate reliability assessment of AC contactors is essential for ensuring the safe and stable operation of the system. The existing reliability studies on AC contactors did not consider the competing failure and degradation correlation among the three-phase contacts as well as the randomness of the failure threshold of each contact, which may lead to inaccurate reliability evaluation result. To solve these problems, a performance degradation model incorporating the competing failure, the degradation correlation, and the randomness of the failure thresholds is formulated by using the cumulative arc erosion amount to characterize the performance state. A multistage parameter estimation method based on the maximum likelihood estimation is presented for the problem that there are many model parameters and it is difficult to estimate all the parameters simultaneously. A reliability approximation method is provided based on the Monte Carlo technique, and an approximate expression of the mean time to failure is derived based on the technique of Riemann sum. Finally, the effectiveness of the proposed method is evaluated by simulation and real case studies. The real case study shows that the proposed reliability evaluation method has high accuracy, and the goodness of fit is about 45% better than the existing methods.