With Money and Government in the Roman Empire (1994), Richard Duncan-Jones revolutionized ancient economic history by estimating the value of coinage in circulation in the mid-second century CE Roman empire through numismatic and statistical methods. His work implied an anomalously high monetization ratio for an ancient economy, given past and current estimates of Roman GDP. This paper offers a new model of the money supply. Based on recent die studies of Hadrian’s gold coinage, it uses Monte Carlo simulation to estimate the value of centrally-minted precious metal coins produced under Hadrian, and then of the total coinage in circulation ca 160 CE. The results suggest a somewhat lower money supply, with much less gold, but more silver than expected. However, the results are not quite low enough to explain away the high monetization ratio, implying a more prosperous and trade-oriented Roman economy than currently thought.