majority, but certainly not as large of a majority assay a 13-point advantage would. so you talk about not taking one pollen running with it. that was a problem. the other problem though was polling didn t accurately reflect what people, how people were going to vote. there were a lot of trump voters who didn t answer polls. they didn t feel comfortable sharing their inclinations. do you think we can see that again? taking poll averages, if people aren t telling us who they are going to vote for, what do they matter? sure. look, polls garbage in, garbage out. but if you look at the 2018 cycle, we have special elections most predominantly in alabama. and i looked at the numbers and what you have found is the democrats is out performing the polling not republicans. so i think one of the key lis lessons from 2016 isn t so much to think republicans will out think the polls, that polls have margin of error and we should adhere to that and it could