3 ms·
For the individual prediction it's impossible to say, so you reward the people who got it right based on how likely they said it was, and you penalize the peopl
by dllthomas 3y ago
For the individual prediction it's impossible to say, so you reward the people who got it right based on how likely they said it was, and you penalize the people who got it wrong based on how unlikely they said it was. If we apply a proper scoring rule, over time the you score better by giving better calibrated guesses.
- khazhoux 3y agoI agree with this. But the original statement is poorly phrased, that "He was right." He was perhaps more right than others who gave Trump lower odds. But there's not an absolute right/wrong (unless someone gave 100:0 odds).
- dllthomas 3y agoHe seems to have been right that Trump's odds were higher because prediction error for different states wasn't independent as some other models were assuming. I agree that we don't have a way of assessing whether the percentage given was "correct", but over a bunch of predictions we can keep track and see how he does when he gives various percentages, or compare against others with a proper scoring rule. I don't make any claim that the odds he gave were right, but just that if they were we should still expect to see his "prediction" be wrong pretty often, so the fact that his prediction was wrong this time only counts so much against him (and to make up for that, only counts so much in is favor if he's right with low confidence).