I investigate the determinants of settlement and plaintiff payouts in federal civil trial litigation. Using a process theory framework, I analyze the impacts of multiple litigation phases on the outcomes of cases and on one another. While most work on litigation offers uncertain inference due to the possibility of using a selected sample, I include both settled and adjudged cases in my sample. This allows me to more closely approximate a random sample, making inferences to the population of disputes more defensible. I find that there’s very little evidence that judicial ideology plays a substantial role in trial-court outcomes, and some evidence of strategic behavior among trial judges. The primary determinants of both whether a case settles, and the outcome to the plaintiff, is the relevant facts and law. The major theoretical contribution of this dissertation is the integration of dialogue among the litigants and the court. Empirically, it innovates on its use of payouts instead of a simple win/loss metric, using events other than the final outcome to measure determinants, and using multiple ideological measures. My findings suggest that our analyses of trial courts should be predicated on the uniqueness of that institutional setting rather than importing models from collegial courts.