I would think the other way around. When one has sufficient documentation, one can extrapolate minor details. When there is little in the way of primary sources to go on, how can one make large assumptions?

I suppose "never" is the wrong word, though.