In studies of probability discounting, the reduction in the value of an outcome as a result of its degree of uncertainty is calculated. Decision making studies suggest two issues with probability that may play a role in data obtained in probability discounting studies. The first issue involves the reduction of risk aversion via subdivision of risk. The second issue involves the differential results in studies of Bayesian reasoning when quantitative information is presented in terms of frequencies rather than percentages. A paper and pencil probability discounting procedure was conducted with 2 magnitudes. Two experimental conditions, 1 that subdivided risk with repeated probabilistic opportunities and 1 that presented information in terms of frequencies, were compared to a standard probability discounting procedure using percentages. The hyperbolic equation of Rachlin, Raineri, and Cross (1991) was used to determine discounting rate. While subdividing risk did not lead to discounting parameters different from other conditions, presenting information in terms of frequencies significantly reduced discounting relative to the standard probability discounting procedure. There was a reverse magnitude effect, with the large magnitude discounted more than the small magnitude, and the same pattern of results was exhibited in all conditions. These results suggest that individuals are less risk aversive when dealing with frequencies rather than percentages.