I feel pretty torn about Myers-Briggs, because while I recognise its shortcomings as a "scientific hot mess", it still provides a useful heuristic model for per

I feel pretty torn about Myers-Briggs, because while I recognise its shortcomings as a "scientific hot mess", it still provides a useful heuristic model for personality types. At its core, it is a model about binary preferences, and people can often be understood by their sharpest preferences, the places where they are most imbalanced. I don't think it's fair to compare this to astrology, which as far as I know isn't accompanied by any kind of reasonable or observable mechanism. An example which comes to mind from recent media is John McAfee. Anyone who has put much thought into MBTI will no doubt all agree that he was an ENTP, who exemplified the charming chaos of extraverted intuition, while becoming paranoid about the parts of life that he repressed: the down-to-earth organisation of introverted sensing. Knowing his Big Five percentages wouldn't give you the same quick insight into how exactly he was always his own worst enemy.