There is contemporary anecdotal evidence that defense counsel appointed in capital cases are frequently inexperienced, underpaid, and lacking in resources to adequately defend their clients. The article presents a reanalysis of the data of Baldus, Woodworth, and Pulaski (1990) on 606 cases of capital murder in Georgia in the 1970s, the only empirical study linking type of counsel to sentence. Controlling for variables describing the character of the defendant and the circumstances of the crime, a death sentence was more likely when defense counsel was appointed rather than retained privately. This was a consequence primarily of the prosecutor's decision to seek a death sentence rather than jury bias in sentencing. Results show that like crimes do not receive like punishments, and that the kind of trial a defendant receives depends to a significant degree on whether he or she has enough money to retain private counsel. The data support the conclusion that sentencing under the Georgia statute was, in the 1970s, and is today to some degree, arbitrary and capricious. Notes, figures, tables, references