It’s a general cultural problem I’ve noted in late-stage capitalist America, especially amongst economists like Silver, that influential people and decision-makers have increasingly insisted on trying to force reality to accommodate their predictive models rather than trying to adjust the models to better reflect reality.
There may be an argument that the 2016 projections accurately informed the electorate, but it’s not going to be convincingly made by someone whose main purpose is trying to cover up his screw-ups and salvage his public reputation.