After the 2016 election, something weird happened in commentary-land. Trump had just won the presidency by trading some traditionally Republican white-collar suburbanites for more typically Democratic blue-collar white voters in the Upper Midwest. And some pundits immediately began talking as if these new Romney-to-Clinton and Obama-to-Trump voters were basically the heart and soul of their new parties.
There are a lot of problems with that sort of narrative. But one big one is that it takes a short-term trend and assumes that itâs necessarily set in stone. We donât know if blue-collar whites in the Upper Midwest are now Forever Republicans, and white-collar voters who hate Trump now arenât guaranteed to stay Democratic forever. In fact, the voters who flipped from one party to another between 2012 and 2016 might be fairly characterized as (gasp!) swing-y voters who will have different midterm turnout rates.
And thatâs a problem for Republicans in the Midwest. Trumpâs overall approval is low, and thatâs going to have reverberations in places with a lot swing-ier voters. In fact (and Iâm far from the first person to point this out) Democrats are poised to post some of their most significant wins (in senate, governor and House races) in Michigan, Wisconsin, Illinois, Minnesota, Iowa, and other nearby states.
And these problems threaten to topple one of the biggest names in conservative circles: Scott Walker.
Walker is a solid conservative who managed to win the governorship of Wisconsin in 2010 on the heels of a big Democratic presidential win and at a time when the conventional wisdom held that Wisconsin was Unchangeably Blue. Then he won a recall in 2012 and re-election in 2014. Election wonks (me included) thought Walker would have a serious shot at the GOP nomination in 2016. But money problems, some tough polls, and a desire to clear the field so that an ideological conservative could take on Trump one-on-one (rather than have Trump run the table on delegates while his opponents divided up the anti-Trump voteâwhich is what happened) drove him out of the field.
And Trump could cause Walker to lose yet another race. Because Scott Walker has a Trump problem. Iâll explain.
To understand Walkerâs problem, we need start with Wisconsinâs Senate race. Democratic Sen. Tammy Baldwin looks like sheâs going to win that race handily. My Senate model, SwingSeat, thinks that her win probability is somewhere around 98 percent based on a combination of polling and other data (such as Trumpâs approval rating, incumbency, whether itâs a midterm, etc.). The polls alone are doing most of the work in this modelâher opponent, Leah Vukmir, has never led in a public, non-partisan poll and the RealClearPolitics average currently puts Baldwin ahead by double digits. Vukmir could theoretically come back (through a massive movement in the polls combined with a favorable polling error) but the odds are very long.
Those numbers have a lot more to do with Trump and Baldwin than they do with Vukmir. I ran a Senate election model that only looks at the basics (presidential approval, past results in the state, etc.) and tried to figure out where the race should go without using head-to-head polling or highly-detailed information about the candidates. Basically the model found that an incumbent Democratis Wisconsin senator running with an unpopular Republican president is in the White House should have a roughly 94 percent probability of beating a replacement-level Republican this year.
So Wisconsin is largely off the Senate map simply because Baldwin is an incumbent from a swing state and Trump is unpopular. And this second factorâTrumpâs unpopularityâis a big problem for Walker.
Walker canât rely just on people who are voting straight-ticket Republican. Heâll have to win a substantial number of Baldwin votersâmany of whom will be expressing their frustration with Trump via a Baldwin vote. I theorized that this might be a difficulty for Walker back in August. Now we have more data to back that theory up.
| Poll | Survey Dates | Vukmir Margin | Walker Margin | Difference |
| Ipsos | Oct. 12 – Oct. 18 | -15 | -3 | 12 |
| Marquette | Oct. 3 – Oct. 7 | -10 | 1 | 11 |
| Marist | Sept. 30 – Oct. 4 | -14 | -8 | 6 |
| Ipsos | Sept. 14 – Sept. 21 | -13 | -7 | 6 |
| Marquette | Sept. 12 – Sept. 16 | -11 | -5 | 6 |
| Suffolk | Aug. 18 – Aug. 24 | -8 | -2 | 6 |
| Marquette | Aug. 15 – Aug. 19 | -8 | 0 | 8 |
| Emerson | Jul. 26 – Jul. 28 | -14 | -7 | 7 |
| Marist | Jul. 15 – Jul. 19 | -17 | -13 | 4 |
| Marquette | Jun. 13 – Jun. 17 | -9 | 4 | 13 |
This table compares Walkerâs margin and Vukmirâs margin in recent polls that took stock of both the Senate and governor races.
As you can see, Walker is consistently outperforming Vukmir. His margin is on average about 8 points higher than hers in any given poll. Thatâs good news for Walkerâthe math is a little fuzzy because of undecided and third-party voters, but itâs not a stretch to think that Walker is wining over some voters who are inclined towards Baldwin in the Senate race.
But Walker still trails Tony Evers, his Democratic opponent, by a single-digit margin in most polls. RealClearPolitics gives Evers a 3.6 point lead, and FiveThirtyEightâs forecast gives Evers a 60 percent chance of winning (Eversâ win probability is 72 percent in their more poll-reliant model). Basically Walker is outperforming Vukmir (which he needs to do), but still falling short.
Not every indicator is negative for Walker. The expert handicappers are a bit more bullish on Walker than the polls. Cook Political Report, Sabatoâs Crystal Ball, Inside Elections, RealClearPolitics and Governing all agree that Wisconsinâs gubernatorial race is a âtoss-up.â And unlike the Senate conest, this race isnât overâit fits in the “Leans-” or “Tilts-” Democratic zone in my view. Many other candidates have come back from this sort of deficit, and Walker has shown that he can be a political survivor.
But the numbers suggest that Walker is in worse shape now than heâs been in any of his three other governor races. In 2010, 2014 and the 2012 recall he lead in pre-election polls, sometimes by a solid margin.
Right now, he’s behind.
