Harper (R)Harper (R) poll in Colorado: Trump trails by a clear margin
Results of a new poll administered by Harper (R)Harper (R) were published. The poll asked participants from Colorado for whom they will vote: Republican Donald Trump or Democrat Hillary Clinton.
In Colorado, the popular vote is usually decided by a narrow margin. Therefore, the state is commonly viewed as a battleground state, which makes it particularly interesting from a forecasting perspective.
Harper (R)Harper (R) poll results
Of those who replied, 45.0% said that they plan to vote for former Secretary of State Hillary Clinton, whereas 38.0% indicated that they would give their vote to billionaire Donald Trump.
The poll was carried out from July 7 to July 9 among 500 likely voters. There is a sampling error of +/-4.4 percentage points. Considering this error margin, the race is currently a statistical tie.
Putting the results in context
In general, however, don't have too much faith in the results of single polls, since they sometimes incorporate large errors. Rather than trusting the results from single polls, one should rely on combined polls or, even better, a combined forecast that incorporates different methods and data.
To make the results comparable to benchmark forecasts, one can convert them into two-party vote shares. This yields figures of 54.2% for Clinton and 45.8% for Trump. For comparison: Only 45.8% was obtained by Clinton in the Harper (R)Harper (R) poll on August 7, for Trump this result was only 0.0%.
Comparison to other polls
An average of recent polls in Colorado sees Clinton at 55.9% of the two-party vote. Compared to her numbers in the Harper (R)Harper (R) poll Clinton's poll average is 1.7 percentage points better. This deviation is within the poll's sampling error, which means that the poll is not an outlier.
Results compared to the combined PollyVote forecast
The current PollyVote anticipates Clinton to gain 53.4% of the two-party vote in Colorado. This means that Polly's prediction is 0.8 points below her polling numbers. Again, a look at the poll's sampling error reveals that this difference is negligible.