This paper briefly examines the nature of hedging and its role in the formulation of categorical and probabilistic forecasts. Hedging is defined in terms of the difference between a forecaster's judgment and his forecast. It is then argued that a judgment cannot accurately reflect the forecaster's true state of knowledge unless the uncertainties inherent in the formulation of this judgment are described in a qualitative and/or quantitative manner. Since categorical forecasting does not provide the forecaster with a means of making his forecasts correspond to such judgments, a categorical forecast is generally a hedge. Probabilistic forecasting, on the other hand, presents the forecaster with an opportunity to eliminate hedging by making his (probabilistic) forecasts correspond exactly to his judgments. Thus, contrary to popular belief, the desire to eliminate hedging should encourage forecasters to express more rather than fewer forecasts in probabilistic terms.
1 The National Center for Atmospheric Research is sponsored by the National Science Foundation.