Three of the jurisdictions (El Paso, Tex.; New Orleans, La.; Seattle, Wash.) had recently attempted to eliminate or greatly reduce plea bargaining; three (Pima County, Ariz.; Norfolk, Va.; Delaware County, Pa.) had few restrictions on plea bargaining and were characterized by more decentralized prosecutorial control. Analysis indicates that differences in sentence severity for defendants who went to trial or pleaded guilty were explained primarily by the fact that the more serious cases were more likely to be tried. When acquittals also were considered, defendants who pleaded guilty received sentences equal to or more severe than those who were tried. Further, more experienced defendants and those who pleaded guilty at the first opportunity did not receive sentence leniency. To a large extent, the same variables (e.g., offense seriousness, prior record) predicted sentence severity for guilty pleas and trials in all jurisdictions. There was some evidence to suggest that the three jurisdictions with tighter controls on plea bargaining succeeded in tightening the fit between case characteristics and sentencing for both plea and trial cases. Tables, 47 references, 16 footnotes. (Author abstract modified)