We present a model for classification performance estimation for synthetic aperture radar (SAR) automatic target recognition. We adopt a model-based approach, in which classification is performed by comparing a feature vector extracted from a measured SAR image chip with a feature vector predicted from a hypothesized target class and pose. The feature vectors are compared using a Bayes likelihood match metric that incorporates uncertainty in both the predicted and extracted feature vectors. The feature vectors parameterize dominant scattering centers on the target, and include attributes that characterize frequency and angle dependence of scattering centers. We develop Bayes matchers that incorporate two different feature correspondence methods. Finally, we compare performance using measured SAR imagery for a 10-class problem under various match operating scenarios.