A new poll out by the University of Colorado today has casual bystanders freaking out, but the numbers may not actually reflect what happens at the ballot box. So, everyone take a deep breath and stick with us.
We’re not in the business of debunking polls due to turnout models because the truth is that we don’t know who will show up at the polls this year. That said, we can certainly compare the composition of this poll to past years’ voter turnout and we can, with certainty, say that the poll numbers touted by the University of Colorado poll depend on historic, and we mean historic, Democratic turnout.
We’re not saying it can’t be done. We’re just saying that it hasn’t been done.
Here is a historical look at who has shown up at the polls over the past seven cycles:
Let’s take a look at who was in the sample from the University of Colorado poll. Here is the party affiliation they sampled:
- Republican: 32%
- Democratic: 37%
- Unaffiliated: 26%
Right off the bat, we can see that, per past elections, this poll is overweight on Democrats. Even in Democrats’ best years, they never break 34%. We’re not saying it won’t or can’t happen, but that it never has before. Similarly, even in Republicans worst years, they’re never below 33%. Plus, this poll seems to skew younger than our impression of the voting population.
Not to mention, we’ve heard that each of these ballot measures are polling much differently than what is represented in this poll. For example, we’ve heard that support for Amendment 74, the property rights amendment is in the high 70s, not 63%.
With a poll as wildly skewed (according to past performance) as this, Coloradans should ask whether the University of Colorado sought to inform or influence in its release.