Results of a new poll conducted by Gravis were spread. The poll asked participants from Colorado for whom they will vote: Republican nominee Donald Trump or Democrat nominee Hillary Clinton.
Colorado is traditionally a swing state, where Republicans and Democrats have historically achieved similar voter support. This is why the election outcome in that state is considered crucial in determining which party will win the majority of electoral votes.
Gravis poll results
The results show that the two candidates have equal levels of support, each with 40.0% of the vote.
The poll was conducted from October 3 to October 4. A total of 1246 registered voters responded. There is a sampling error of +/-2.8 percentage points. Considering this error margin, the race is currently a statistical tie.
Putting the results in context
Single polls should be treated with caution, as they can include large biases. Rather than relying on results from single polls, we recommend to rely on combined polls or, even better, a combined forecast that includes forecasts from different methods, each of which includes different data.
In order to make the results comparable to forecasts from benchmark methods, one can translate them into two-party vote shares. The results of the actual poll mean 50.0 % for Clinton and 50.0 % for Trump in the two-party vote share. On July 8 Clinton received 51.2% in the Gravis poll and Trump received only 48.8%.
Results vs. other polls
Looking at an average of Colorado polls, Clinton's two-party vote share is currently at 54.5%. In comparison to the average forecast of other polls Clinton performed 4.5 percentage points worse in the poll. This deviation is outside the poll's margin of error, which suggests that the poll is an outlier.
Results compared to the combined PollyVote forecast
The most recent PollyVote predicts Clinton to gain 53.2% and Trump 46.9% of the two-party vote in Colorado. Clinton has 3.2 percentage points less when the results of the poll are compared to the combined PollyVote forecast for Colorado. The PollyVote forecast is thus outside the poll's sampling error.