Space structures are subject to thermal and mechanical loads. Matrix cracks can form in composite components, which results in a change in their thermal and elastic properties. The objective of this study is to develop a method to predict transverse microcracking in general composite laminates subject to thermal and mechanical loads. The approach combines probabilistic and analytical components in an incremental damage method. The probabilistic components include a distribution of flaws characterized by a Weibull probability function, seeding of flaws at random locations, inspection of flaws in random order for crack initiation, and inspection of cracks in random order for extension. The analytical components include fracture mechanics based energy criteria that uses a shear lag derivation of the stress and displacement fields. Degradation of material properties, temperature-dependent material properties, and a material variations model are incorporated into the method. This method is implemented through a computer program that predicts crack densities, crack distributions, and degraded laminate properties as functions of an arbitrary thermomechanical load profile. Parametric analyses are used to understand the behaviors predicted by the method and their sensitivities to model parameters. Predictions are compared to previously collected data and observations for different laminate configurations and material systems. For both thermal and mechanical loads, crack density predictions capture general trends and agree with much of the data. The method shows improvements on the current state of the art in several areas. The effective flaw model predicts the initiation and gradual accumulation of cracks. The material variations model allows the method to emulate the intrinsic variability of the crack data. The method predicts crack distributions and their evolution as cracking progresses. This evolution can include the formation of different crack types and patterns. The effect of ply thickness on this evolution is correctly predicted. The success of the method shows its superiority as a tool for predicting cracking. By replicating complex observed behavior using a relatively simple method, the work supports the physical soundness of the method and increases our understanding of the mechanisms of microcracking in composite laminates.