Some should stop with the Brexit analogy. The US is fundamentally different. First, UK polling is notorious for being unreliable, this is due to the nature of their elections, smaller baseline pool and lower quality analysis. UK pollsters are famous for using novel methods, due to the difficulty of accurately predicting results, whereas the US has more conventional voting and a massive number of elections, creating a reliable sample.
Also, the primary vote in the UK swings conservative, hence they often outperform polls. The US is the opposite, both on issues and primary vote, there is a huge Democrat advantage. Hence predicting Trump will do a Brexit seems weird. That proportionally large, latent pool of conservative voters isn't there. I can't predict if he will win, but so far in the primaries he underperformed based on 538 polls and that was amongst conservatives. Given the wider population Dems outnumber Repubs by like 40% and they massively outpoll Repubs on key issues, guess we will need to be patient with the election and see if any events counter the natural advantages.