This paper proposes a novel method for estimating future operational risk capital: incremental value-at-risk (IVaR). The method can be used either for predicting VaR in the short term or as a “sense check” for a capital value that has already been calculated. Its foundation is the difference in data between one capital calculation and the next using any established procedure. For a constant length data window, there is only a minor change in data when that window moves forward in time, assuming that data entering the window presents no marked deviations from data leaving the window. There is no change for the bulk of the data. Following a capital calculation, we use a Bayesian analysis to generate “synthetic” data prior to new data arriving. That “synthetic” data is used as a proxy for actual data in advance of its arrival. Capital is then estimated using an ordinate on the quantile distribution proposed by C. R. Rao in 1973. The IVaR method requires an assumption of data homogeneity, and if this is met, the accuracy of the prediction is high; 5% accuracy is achievable for a one-month-ahead prediction. The figure is similar for predictions three months ahead, although data disparities are apparent with this horizon.