I wasn't drawing a dichotomy between "an outlier and a point that is far out of the distribution of the other points because of unmodeled variation". I was sugg

I wasn't drawing a dichotomy between "an outlier and a point that is far out of the distribution of the other points because of unmodeled variation". I was suggesting that they may be the same thing.