If you look at national polling, Donald Trump and Kamala Harris are in a flat-footed tie. However, district-level polling—which, so far, hasn’t been contaminated by pro-Trump pollsters trying to push the narrative of a “red wave” for Trump—suggests that Harris is actually leading, albeit narrowly.
That’s why I was initially surprised to see a poll from Fort Hays State University’s Docking Institute for Public Affairs showing Trump leading by only single digits among likely voters in traditionally crimson-red Kansas. Specifically, five points—48.2% to Harris’ 43.2%. Prairiecrat diaried it here, but I thought a deeper dive into this poll was necessary. While there’s a lot to suggest this poll may be believable, it would also buck decades of voting history in Kansas, a state which has normally long since been sewn up for the GOP at this point. Conventional wisdom would suggest that if Kansas is within single digits this close to Election Day, Harris may be her way to a bigger win than even the most optimistic among us have believed.
Check out the crosstabs here. The folks at Docking applied their likely voter screen after excluding those who “do not plan to vote or cannot vote.” From the looks of it, the race is close because Trump loses five percent of self-identified Republicans compared to Harris losing only 3.7% of self-identified Democrats. They’re almost tied among independents—41.5% Harris, 41.3% Trump. The one crosstab that doesn’t seem to make sense is voters who are 18-34. Trump wins them 53.9-34.4; in most of the rest of the country, that figure would be reversed.
On one hand, Docking appears to be a fairly accurate pollster. It has five polls in FiveThirtyEight’s database. Most recently, as Prairiecrat noted, its final poll of 2020 had Trump leading Biden by 14.4; he won by 14.6. In 2018, it had Laura Kelly leading Kris Kobach by four; Kelly won by six.
On the other hand, though, conventional wisdom would make it hard to square Trump only leading by single digits in Kansas with all the indications that this race is close. Kansas has never even potentially been on the board this close to Election Day in my lifetime. Indeed, the only time it was close since it was swept up in LBJ’s landslide of 1964 was in 1976, when Carter only lost it to Ford by eight. In that election, Carter took 44 percent of the vote—one of only four times the Democrats have done so since 1964. Combined with our current hyperpolarized environment, conventional wisdom would suggest that if Trump is only leading in Kansas by single digits this close to Election Day in the absence of a substantive third-party candidate (which is why Bush 41’s five-point win over Clinton in 1992 doesn’t count), he’s potentially getting Goldwatered. Imagine if Harris were only leading by single digits in Rhode Island. We’d be in full panic mode right now. Plus, none of Kansas’ three Republican-held congressional seats are on the board. That includes the only even potentially competitive Republican-held seat in the state, Topeka-based KS-02, where former Rep. Nancy Boyda has struggled to gain traction for an open seat.
But one of my longtime friends who lives in Wichita and has long been active in the local League of Women Voters thinks there may be something to this poll. She told me that Docking has a sterling reputation in Kansas. Plus, it’s based in Hays, in the more rural area of the state west of Wichita. She recalled that most Kansas Republicans are old school types, cut from the same cloth as Bob Dole—in other words, the kind of Republican who, on paper, would recoil at the horror of Jan. 6. She knows; she’s nominally registered as a Republican, but has voted Democratic for years in part because wingnuts like Kobach and Sam Brownback have seized control of the state party.
Another old friend of mine recalled having a major client in the Topeka area during his days in the financial services industry. From his time there, he remembered that most of the people there were “salt of the earth types” who were pretty conservative, but “nothing like MAGA.”
All told, I really don’t know what to make of this. I’m a very partisan Democrat, but I deal in truth, and I’m inclined to be skeptical of any poll that may favor us but doesn’t pass the smell test. For instance, in early October, ActiVote rolled out a poll showing Harris leading in my home state of North Carolina even as it had Trump winning nationally in a poll that was in the field around the same time. That literally does not happen; as we all know, a Republican cannot win without North Carolina. But on the other hand, Docking’s track record suggests it doesn’t miss by much. If this poll is accurate, to put it mildly, it’s another potential canary in the coal mine.
