Request: Clarify meaning of judging a prediction "right' or "wrong" to refer to whether event occurred

[vivianlwd]

I know at least one person who interprets the question of whether a prediction is "right' or "wrong" as an evaluation of their accuracy with respect to the prediction as opposed to whether the predicted event occurred or not. E.g. if the prediction is "X will happen" at 1% confidence, and X does not happen, they will judge their prediction as "right." This causes their Brier score to get worse, though, because what the "right" and "wrong" are referring to is whether or not the predicted event occurred. I suspect quite a few people misunderstand the question that way. I suggest adding a clarifying sentence stating something like, "A prediction is 'right' if the predicted event occurred, and 'wrong' if it did not occur, no matter the ascribed confidence level," above the "unknown," "right," and "wrong" buttons. Thanks!