By NICK GOSNELL
Hutch Post
HUTCHINSON, Kan. — University of Kansas Political Science professor Patrick Miller said there isn't really good polling data on the governor's race in the state, despite how close national pundits perceive it to be.
"We have three or four polls of the gubernatorial race, all of which I would absolutely characterize as cheap and low quality," Miller said. "All of them show a toss up, a very close race among undecided voters. They don't tell us much else that's interesting."
Republican Derek Schmidt, despite his high profile in Kansas, doesn't create the kind of national interest the previous nominee did.
"We haven't had much polling of our race this year," Miller said. "I think in 2018, we were an attractive race for a lot of out of state pollsters to poll, because of Kris Kobach being the Republican nominee and all the attention that he brought to that. We haven't had much attraction of those really established and good pollsters to our race this year, unfortunately."
With a race this close and the inherent play in the joints in even the best polling, it's easy to see how the race could go either way.
"Polling is not an absolute," Miller said. "Polling is always a prediction with a degree of error," Miller said. "Let's say that a candidate has, let's make this simple, 50% of the vote. The wrong way to interpret that poll is to say the candidate has 50% of the vote, end of story. That would just be incorrect. Polling always comes with that margin of error. Let's say it's four percent. If the poll says that a candidate has 50% of the vote, with a four percent margin of error, then the correct way to interpret that is that, statistically speaking, there's a 95% chance that the candidate's true level of support is between 46% and 54%."
It really comes down to who shows up at the polls. Variables like weather and national trends that go up and down completely out of the control of both campaigns could make enough difference to decide a race that is within the margin of error.