Methodology
I went looking and couldn’t find much. Iirc, they require you to join to go deeper. They didn’t say you’d get access to the internals. I assume that is included.
I don’t trust polls that don’t reveal internals. Too easy to turn on one candidate or another when it is politically helpful.
This is the exact same team which developed RAND in 2012 and correctly predicted the 2012 election based on their own polling + panelists, so until events dictate otherwise I am going to error on the side of trusting their methods. Although I do wish they would account for the other candidates and publish the “Other” or not voting options.
http://www.latimes.com/politics/la-na-pol-usc-daybreak-poll-methodology-20160714-snap-story.html
It's the same methodology as the Rand Poll that was #4 in 2012. It's innovative methodology, but since Gallup was so bad in 2012 that they gave up polling presidential elections, "something" new is needed in the wireless era. Is this "it"? We will see, "in the fullness of time".