The majority of research examining prosecutorial discretion has focused on legal factors such as the seriousness of the offense or the extra-legal characteristics of the accused including race/ethnicity and gender. The amount of variance explained by court researchers, however, remains quite low. The present study extends previous research examining the primary determinants of prosecutor's decision to dismiss or fully prosecute focusing on driving while intoxicated cases. The authors focus on the predictive contribution of the strength of evidence relative to legal and extra-legal variables. The data consist of 2,358 driving while intoxicated cases filed in Harris County, TX during the first 8 months of 1999. The findings strongly support the inclusion of strength of evidence variables in court research and further suggest their past omission may have attributed significance to spurious relationships. Abstract published by arrangement with Springer.