In the NY times today: When Polling is More Like Guessing, by Nate Cohen.
He leads with this opinion:
“Election analysts and forecastersdepend on accurate polling. Unfortunately, there’s not much of it so far this cycle.
Many of the surveys to date have been conducted by firms that use automated phone surveys and combine deficient sampling with baffling weighting practices.”
He goes on to provide evidence of some of the disconnects between polling results in terms of projected demographics versus likely demographics about who will vote. Differences in demographics will alter predictions.
To the extent that polling predictions affect actual votes–which is an interesting research question–it is perhaps more than just an argument about demographics.