NBC-WSJ-Marist published the results of a new poll. In this poll, respondents from Colorado were asked for whom they will vote: Republican nominee Donald Trump or Democrat nominee Hillary Clinton.
Historically, Colorado has been a swing state, in which neither the Democrats nor the Republicans have had overwhelming support to clinch its electoral college votes. This is why predictions here are of particular importance.
NBC-WSJ-Marist poll results
The results show that 46.0% of interviewees would cast a ballot for former First Lady Hillary Clinton, while 32.0% would vote for real estate developer Donald Trump.
The poll was carried out from August 4 to August 10 among 899 registered voters. Taking into account the poll's error margin of +/-3.3 percentage points, the gap between both candidates is statistically significant.
Putting the results in context
As a general rule, however, don't have too much faith in the results of single polls, as they sometimes contain large errors. Rather than trusting the results from single polls, one should look at combined polls or, even better, the combined PollyVote forecast that relies on different methods and data.
For the following comparison, we convert the candidates' raw poll numbers into two-party vote shares. This procedure yields figures of 59.0% for Clinton and 41.0% for Trump. For comparison: Only 55.1% was obtained by Clinton in the NBC-WSJ-Marist poll on August 1, for Trump this result was 44.9%.
Results in comparison to other polls
If we look at an average of Colorado polls, Clinton's two-party vote share is currently at 57.0%. This value is 2 percentage points lower than her respective numbers in the NBC-WSJ-Marist poll. This difference is within the poll's error margin, which suggests that the poll is not an outlier.
The poll in comparison with PollyVote's prediction
The latest PollyVote expects Clinton to gain 56.4% of the two-party vote in Colorado. This means that Polly's combined forecast is 2.6 points below her polling numbers. The PollyVote forecast is thus in line with the poll's sampling error.