Those who are right in the end are most often those who changed their minds a bunch of times
There are forecasting tournaments wherein Professional Forecasters register predictions for specific, measurable, time-bound real-world results. The Top 2 correlating factors to those forecasters who do the best are raw intelligence (as one would expect) as the 2nd highest-correlated, but by far and away the highest-correlated trait to “being right” is how frequently the predictors changed their minds. This is a reflection of them not being “tied” to their current prediction in light of new information. They don’t hold predictions too closely, and actively seek out disconfirming information.