The chart above reflects the results of the Data for Progress Poll in Nevada -- done between February 12th and 15th of 766 likely Nevada caucus voters, with a 3.4 point margin of error.
The chart above reflects the results of the Point Blank Political Poll in Nevada -- done between February 13th and 15th of 266 likely caucus voters, with a 5.6 point margin of error.
Cable news commenters have remarked about how hard it is for polls to predict the outcome of the Democratic caucus in Nevada. The two polls above show this is true. One has Sanders leading, while the other has Steyer leading.
Why is this? One reason is the small samples used in the Nevada polls (making it important just who they questioned). Another is that many voters seem to be having trouble making up their minds. And then their is the unique nature of the Nevada voting -- with each early voter making up to 5 ranked choices. If their first choice is not viable enough to qualify for delegates, then the vote-counters go to the second choice (and so on).
Since delegates are not awarded on just the first choice of voters, it makes it difficult to see how delegates will be awarded -- and it could be very late before we know who won how many delegates in the state.
