Flip‐Flop Index: Quantifying revision stability for fixed‐event forecasts

Traditional verification statistics focus on measures that inform us of forecast skill. Another aspect of the forecast is its stability or, to take the opposite view, how much it "flip-flops". Weather forecasters report that they are reluctant to change a published forecast if they judge there is a risk of it being changed back again as they consider that such instability detracts from the message delivered and likely response. This aspect of forecasts is quantified using the Flip-Flop Index [1].