Handicapping Minnesota's Battleground House Districts

Based on the available polling data and my expectations of the parochial tendencies and recent localized voting patterns in Minnesota, I figured I’d take a stab at handicapping what I think the most realistic path to victory looks like in the four primary House race battlegrounds in the state as well as the 7th district which is not expected to be a battleground this year but wasn’t in 2016 either yet ended up being a five-point race.  Without further adieu, let’s get rolling…..

MN-01 (Dan Feehan vs. Jim Hagedorn)

I’ve stated before that I think Democrat Dan Feehan, a political novice who moved to southern Minnesota since the 2016 election to run for office, seems to be a strong political talent but is probably suffering right now from a name ID disadvantage against his Republican challenger Jim Hagedorn.  There’s a month left for Feehan to raise his profile, but Hagedorn begins this race with a couple of advantages.  Despite being considered too conservative for the district, Hagedorn has run in the past two cycles and overperformed expectations both times, especially in 2016 when he came within a point of winning.  His father was a Republican representative in the central portion of the district in the 1970s and that legacy surname clearly moved some votes even when Hagedorn was challenging the incumbent Walz.  With a lesser-known Democratic challenger on the ballot, it’s reasonable to believe the Hagedorn surname will move even more votes, particularly in the district’s western counties.

So we’ll start with the defensive posture counties for Feehan.  I expect Hagedorn to do at least as well as he did in 2016 in counties such as Brown, Martin, Faribault, Watonwan, and Jackson, which are more rural counties in the western and central portions of MN-01.  Most of these are politically elastic counties but Hagedorn likely moved thousands of votes away from Walz there in 2014 and 2016 because they were part of his father’s district, and I expect he will take away a significant number of would-be Feehan votes in 2018 as well.  And while I expect Feehan to win, and perhaps decisively, in Blue Earth and Nicollet counties in this same region which make up the immediate Mankato area, it’s still reasonable to believe his margins will be weaker than they would be if not for Hagedorn’s familial legacy representing the area.

Moving east, Presidential year turnout in Rochester and the college city of Winona was the difference in Walz keeping his job by about 3,000 votes in 2016, and Rochester and Winona will be integral in a path to victory for Feehan too.  Rochester has the kinds of upscale demographics that could really see a shift towards Democrats this cycle, but it’s still optimistic to believe that in a midterm cycle, Rochester can itself make up the number of raw votes in Feehan’s favor that Hagedorn’s parochial advantage in the greater Mankato area will produce for him.  Feehan almost certainly needs more than Rochester if he’s gonna win district-wide.

Traditionally Democratic and union-heavy Freeborn County (Albert Lea) and Mower County (Austin) voted for Walz in 2016 but with softer margins than usual.  If this race ends up competitive at all I suspect Feehan will win in those counties, and probably better than Walz did in 2016.

That brings me to the five counties which I suspect will be the epicenter of the 2018 battleground.  Four of them are in between Mankato and Rochester, with little to no influence from Hagedorn’s legacy vote….

The first two, and in my opinion the counties that will likely decide the winner, are Waseca County (Waseca) and Steele County (Owatonna).  Steele County in particular has a significant number of voters while Waseca is the more rural of the two but still highly relevant.  Both are German-settled and ancestrally Republican, but have shown signs in the last decade of becoming more bipartisan.  Walz won both in 2014 while losing them both by five points in 2016.  I don’t think Feehan has the luxury of narrowly losing them in 2018 and still prevailing, which makes them the election night bellwethers I’ll be checking out on the November 6th map.  If Waseca and Steele counties are blue on election night, Feehan probably wins.  If they’re red, my money’s on Hagedorn.

The next pair of bellwether counties I’ll point to are just north of Waseca and Steele…..Le Sueur County and the non-Northfield portion of Rice County.  MN-01 inherited these two counties in 2012.  Unlike Waseca and Steele, they were historically Democrat-tilting but have grown more conservative as they’ve transformed to exurbia in the last 20 years.  Walz narrowly won the two counties in 2014 but they shifted hard to Hagedorn in 2016.  I’m inclined to think Hagedorn wins them both again this year, but Feehan probably needs to shave those numbers down from the 10+ point victories for Hagedorn in 2016.  If either of them does go Feehan, it’s hard to imagine Hagedorn is winning district-wide.

The final bellwether to watch is Nobles County in southwestern Minnesota, which is outside the orbit of Tom Hagedorn’s old Congressional district, and is ancestrally Democratic and Irish Catholic heavy, with a population center of the meatpacking town of Worthington which is majority-minority.  Walz won Nobles County in 2014 and Hagedorn won in 2016.  Without having any parochial biases and a generally elastic and wave-prone electorate, Nobles County is very up-for-grabs this year.



MN-02 (Angie Craig vs. Jason Lewis)

Minnesota was full of surprises on November 8, 2016, and one of the biggest was the victory of conservative firebrand Jason Lewis over his Democratic challenger in a race where “Hillary’s coattails” were considered certain to push Craig over the finish line.  Instead, Lewis won the district by a wider margin than Trump, although he got a major assist from a third-party candidate not on the ballot this year who I suspect played the role of the spoiler.  I remain pretty confident that with the district’s demographics and Lewis’ position of weakness from the outset that it’s ripe for a decisive mid-to-high single-digit victory.  Let’s break it down….

Dakota County is the center of gravity in this district with about half of its population and three distinct political identities.  The northern portion of the county (South St. Paul, Inver Grove Heights) is blue-collar and traditionally Democratic, but saw a slight shift in Trump’s direction in 2016 as most WWC-heavy areas did, but I still expect it to come out soundly for Craig this year.  The central portion of the county (Eagan, Burnsville, Apple Valley) has always been its primary political battleground, with a moderate and upscale profile and has been trending Democratic for several cycles now.  Hillary outpaced Obama’s margins in this part of the county and I expect Angie Craig to really boost her margins compared to 2016.  The exurban southern portion of the district (Lakeville, Farmington) has always been Republican and much of it shifted a little further towards Trump in 2016 but coming out even stronger for Jason Lewis.  The southern portion of Dakota County is by no means its most heavily populated region, but it’s definitely not a small number of voters we’re talking about either, which is why Craig won the northern and central precincts of Dakota County soundly in 2016 yet only prevailed by 2 points countywide.  I’ll predict a slight shift towards Craig in this portion of the district coupled with more considerable movement towards her in the northern and especially the central part of the county, leading to an Angie Craig win of around 10 points in Dakota County.

Bordering Dakota County to the west is Scott County, which has somewhat similar demographics to Dakota yet far more conservative politics, being the GOP voter base of the 2nd district.  It was Jason Lewis’ best of the six counties in 2016 and I fully expect him to win it again in 2018.  But particularly in the more moderate and educated quasi-suburbs Shakopee and Savage in the northeastern part of the county, I expect considerable shift towards Craig.  If Lewis only wins Scott County by single digits, which I figure is a real possibility, then he’s definitely toast district-wide.

The small portion of southern Washington County in MN-02 is working-class and Democrat-leaning, but probably won’t produce the deciding margins of the election unless it ends up being really close.  The Rice County portion of MN-02 includes the very liberal college town of Northfield and will be Craig’s base as it was last time.  But I don’t see her netting too many if any more votes out of Rice County than in 2016, since she probably tapped out the college vote in a Presidential year.

And certainly worth keeping an eye on are the more rural Goodhue and Wabasha counties on the district’s southeast side.  They lean Republican and went strongly for Lewis, but are also quite elastic and if Craig ends up really pulling away in the end, it wouldn’t surprise me to see either or both of Goodhue and Wabasha going blue as well.  Al Franken won Wabasha in 2014, after all.  If either of them do go blue, it’s certainly game over for Lewis, but my guess is the baseline here will be in central Dakota County.  If Eagan goes Craig by more than 15 points, while Burnsville and Apple Valley go Craig by more than 10 points, and maybe throw in a narrow Craig win in Shakopee, then I think she’s a safe bet for victory.


MN-03 (Dean Phillips vs. Erik Paulsen)

I will have quite a bit less to say about this race than most of the others I’m profiling because the district’s demographics are uniform enough that I expect it will largely swing pretty evenly from precinct to precinct in the Democrat’s direction if the polls are right and Paulsen is in as big of trouble as it seems.   One of the reasons Paulsen has been so hard to beat is that the Democrats have very few reliable strongholds in MN-03.  The eastern half of Bloomington and the majority minority city of Brooklyn Park are it, and a decade ago even they were quite a bit more competitive.  But they’ve trended Democratic and I suspect have a lot more room for Democratic growth as evidenced by the degree to which Hillary dominated Trump there in 2016.

Beyond that, I predict the swing in Phillips’ direction to be fairly consistent from a PVI standpoint whether we’re talking about Democrat-leaning Minnetonka, Republican-leaning Maple Grove, or the tony lakefront suburbs on Lake Minnetonka which have always been staunchly Republican but don’t fit the demographics of Trump’s Republican Party.  Even Chanhassen in Carver County, the brightest red county in the metro area, is very likely to go for Phillips if the early polls prove correct.   The only exceptions where Paulsen might hold up better compared to his past wins include his long-time home base of Eden Prairie (leaning Republican but anti-Trump), the more working-class Coon Rapids across the river in Anoka County which actually bucked the district and flipped from Obama to Trump in 2016, and the exurbs on the western fringe of Hennepin County like Rogers and Medina, which make up the core vote that keeps Republican gubernatorial candidate Jeff Johnson on the Hennepin County Board.   Barring a huge polling error or a late-breaking disqualifier against Phillips, I suspect the MN-03 race will closely reflect the 2016 Presidential outcome.


MN-07 (Collin Peterson vs. Dave Hughes)

Nobody expected Collin Peterson, the most conservative Democrat in Congress, to be at any risk in 2016 against his fourth-rate GOP challenger Hughes….but nobody expected he’d be facing a headwind of 2-1 Trump in what had previously been a Republican-leaning but competitive district in western Minnesota farm country.  It seems less likely that Peterson will prove as vulnerable in this environment but I figure it’s worth analyzing to point out his 2016 weaknesses and where the potential lies for him to get back to the kinds of dominating margins he posted in the district during the Bush years.

Peterson’s base going back 30 years and then some is the Red River Valley of northwestern Minnesota, home to a lot of sugar beet and wheat farms along with the Minnesota side of the Fargo-Moorhead and Grand Forks “metro areas”, and I suspect he’ll get his strongest support there in 2018 as he has in the past.  As for the southern half of the district south of I-94, Peterson inherited most of that turf after 2002 and has never had quite the connection to that part of the district as he has in the Red River Valley, but has still prevailed in most of southwest Minnesota’s thinly populated farm counties, even in 2016.  It’d be surprising if he didn’t get better numbers than in 2016 or even 2014 this year.

Part of the reason MN-07 is such a tough nut for a Democrat to crack is because its most heavily populated counties (Otter Tail, Douglas, Stearns) are its most Republican.  Even though Peterson won the district by more than five points in 2014 and 2016, he still lost these three counties both times.  It’ll be interesting to see if he wins them back this time.  My guess is he does, but I suspect Douglas (Alexandria) will be the least likely of the three as for some reason he’s always underperformed there.  The biggest shift away from Peterson in 2016 came from the counties in the southeastern portion of the district, and since they’re also among the more populated counties of the district, the impact was felt in the overall margin districtwide.  Already Republican McLeod County (Hutchinson) went narrowly for Torrey Westrom in 2014 but Dave Hughes won it by 15 points.  Meeker County (Litchfield) went from Peterson +4 in 2014 to Hughes +10 in 2016.  And Kandiyohi County (Willmar) went from Peterson +9 to Peterson +3.  I’ll be curious to see if those counties swing back Peterson’s direction this year.

My best guess is that Peterson wins decisively, somewhere between 15 and 20 points.  I’d love to see him crest 60% again and it wouldn’t completely blow my mind if he did, but I feel like there’s been some conservative consolidation in this district since the Bush years and that the days of 2-1 Peterson victories are likely over.  I hope he holds off retirement until 2022 though, ensuring the Republicans don’t have an incumbent heading into the next district configuration, whatever happens with that.


MN-08 (Joe Radinovich vs. Pete Stauber)

I suspect of all the battleground House races in Minnesota, Democrat Joe Radinovich in the open seat in northeastern Minnesota has the narrowest and most complex path to victory.  In the last two cycles, Democrat Rick Nolan eked out narrow victories to hang onto a seat that was a Democratic stronghold only a decade ago, but there are significant fractures in the coalition that made Democrats so formidable in the district for decades.  In particular, the heavily unionized mine and steelworkers on the Iron Range are trending Republican for some general reasons and some very region-specific reasons, even as the areas of MN-08 seeing population growth are heavily Republican and getting more so.

With that in mind, I think Democrat Radinovich is gonna lose quite a bit of the support Nolan got in St. Louis County, particularly since his challenger is a county commissioner who has scored a number of endorsements from DFL mayors on the Iron Range.  Nolan won St. Louis County by 23 points in 2016 while Hillary won it by 12, mostly on the strength of Duluth.  And Stauber’s county commissioner district is actually in the portion of the county surrounding Duluth, meaning he’ll likely overperform near Duluth as well as the Iron Range.  My guess is Radinovich’s margin in the county will fall somewhere between the margin of Nolan and Hillary, but probably closer to that of Hillary.  I’m betting a 15-point Radinovich win in St. Louis County.

If I’m right, Radinovich needs to make up votes elsewhere in the district to make up for lost votes on the Iron Range.  His best bet is the upscale and heavily populated Brainerd Lakes area.  Stewart Mills, the Republican nominee in 2014 and 2016, was from the Brainerd area and scored strong margins there.  If Trump fatigue is gonna be felt anywhere in the otherwise demographically favorable MN-08, it’s likely to be among the McMansions around Brainerd.  The fact that Radinovich briefly held a House seat in this area gives him a competitive advantage that could prove very handy in peeling away some of those Mills voters from 2016.

Another option for Radinovich to outperform Nolan is the Twin Cities’ northern exurbs in Isanti and Chisago counties.  These are ancestrally Democratic counties whose population has more than doubled with mostly conservative transplants in the past generation.  The problem for Radinovich is that Nolan was already unusually strong in these counties, performing well above the recent Democratic baseline.  Nolan won in 2014 and 2016 largely because he didn’t completely collapse here in MN-08’s southeast side.  It’s expecting a lot to believe Radinovich will outpace Nolan here and net too many additional votes from 2016.

More promising for Radinovich is the northwest side of MN-08 which remains Democratic and went narrowly for Nolan in 2016 despite flipping decisively for Trump at the top of the ticket.  Koochiching County (International Falls) is thinly populated and won’t net too many votes, but Itasca County (Grand Rapids) could theoretically net more if Radinovich scores numbers consistent with the Democratic baseline.  There’s an Iron Range influence in Itasca County, however, that is a wild card.

Duluth is the largest city in MN-08 but it seems unlikely that Radinovich can net more votes in the heavily Democratic city than Nolan did in a Presidential cycle and as I said before his ties to Duluth are more likely to result in Stauber inroads in the area. Ditto for the other counties on the north shore of Lake Superior.  Carlton County south of Duluth has significant potential to become bluer, however, and could nudge Radinovich incrementally closer to the finish line.

This is the race I’ve been by far the most bearish about amongst the Minnesota battlegrounds.  It strikes me that there will be a lot of moving parts needed for Radinovich to eke out a win as the Democratic Party attempts to squeeze out the last drops of residual strength it has to salvage a seat slipping away from them before the next round of redistricting.  Even if they win it in 2018, it’s a tall order to hold it in 2020.  Maybe I’m overestimating how much Democratic bleeding we can expect from the Iron Range, however, as they surprised me with how well they held up in 2014.  If Radinovich did as well as Nolan did in St. Louis County, then he would quickly become the favorite in this race.  That will be far from easy against Stauber though.

And Radinovich has an additional obstacle on his path to victory from Independence Party candidate Ray “Skip” Sandman, a perennial candidate from the left in this district who ran with the Green Party in 2014.   Sandman’s candidacy was a headwind for Rick Nolan in 2014 as the 4.3% of the vote that Sandman scored very likely came at Nolan’s expense, and it was extremely lucky for Nolan that Sandman didn’t run in 2016 because he would probably have been the difference in such a challenging cycle for Democrats.  Sandman’s candidacy is exactly what Radinovich does not need this year as it likely means thousands of diverted votes far more likely to go in his column than in Stauber’s.