The difference between LA times "survey" and the average of traditional polls has been 5-6 pts (and rather consistent). Again, difference between surveying and polling is predicting turnout and elimination of lying (since surveys are anonymous and polling is not). Don't you see why there is a large delta between the LA times survey and the polls when Trump is in the race? Finally do you really think the poll people are going to admit this while they are getting paid? They will admit it after the election and then say 2016 is one-off that won't be repeated.