The 2016 election results were surprising for a number of reasons, but perhaps most surprising was the outcome in Minnesota. The state, which Republican presidential candidates have not carried since 1972, swung heavily toward Donald Trump. His 1.5-percentage-point loss represented the best showing for a Republican presidential candidate since Ronald Reagan narrowly missed carrying the state in 1984.

Given this, we might have expected Minnesota to continue trending toward Republicans. The 2018 midterms were held in a very hostile environment for Republicans overall, yet their performance in Midwestern states was surprisingly respectable. They defeated a Democratic incumbent in Indiana, held the Iowa and Ohio governors’ mansions, nearly held the Wisconsin governor’s mansion, and put up surprisingly strong performances with unheralded candidates in the Michigan and Ohio Senate races.

Minnesota did not follow this trend, however. Sen. Amy Klobuchar won reelection to the Senate by 24 points, while appointed Sen. Tina Smith held her seat by 10. Democrat Tim Walz (pictured, with Klobuchar) won the Minnesota gubernatorial race by 11 points. Overall, Minnesota and Illinois were two Midwestern states that didn’t even offer Republicans a silver lining on a bad election night.

In 2020, the state once again seems to be lagging others in presidential polling. Before a recent poll from Emerson College showing a close three-point presidential race (about which, more later), the RealClearPolitics average had Joe Biden leading Trump in Minnesota by nine points, compared to 6.7 points in Michigan, 5.5 points in Wisconsin, and 6.4 points in Pennsylvania.

You can see the overall trendlines over the past decade here:

This shows the partisan lean of the state – that is, how much more Republican or Democratic the state was than the country as a whole in a given election. (If Donald Trump got 49% of the vote in a state, it has a one-point Republican lean, since he received about 48% of the vote nationally.) As you can see, most of the states have swung precipitously toward Republicans in the past few cycles. This has been particularly pronounced in Iowa, Ohio and Wisconsin. Illinois has continued to trend away from the GOP. Minnesota has moved toward Republicans, but it has been a much more gradual shift than those of other states.

Why is this the case? To understand this, it may be useful to return to a series of articles that David Byler and I wrote in the wake of the 2016 election. In that series we observed that American elections could increasingly be explained by a geographic component – rural areas were increasingly becoming Republican while America’s megapolises were swinging toward the Democrats. This explains the shifts in the Great Lakes states.

In the following series of charts, I’ve taken each state and broken it down by county into six categories, based upon the metro/micropolitan area the county is in: Mega Cities (think New York, Chicago, Houston), Large Cities (e.g., Columbus, Ohio), Small Cities (e.g., Youngstown, Ohio), Large Towns, Small Towns, and Rural Areas. We can look at each state over the last eight presidential elections, and see some obvious trends (click to enlarge):

In every state, rural areas and small towns have broken heavily against the Democratic Party. They are, generally speaking, about 10%-20% more Republican than the country as a whole now. Note that this was not always the case; in most of our states, these areas were right in the middle back in 1988. Large towns have also trended toward Republicans, although there is more heterogeneity here, especially in Indiana.

Small cities and large cities, on the other hand, have been fairly consistent over the past few decades, with the notable exception of Wisconsin (where Madison continues to trend leftward). The Chicago metropolitan area – the lone “mega city” in the region – has moved toward Democrats, except for the exurban sliver in Wisconsin.

The point is, across all states in the region, the rural/town/city areas have been behaving in similar ways. So what does this tell us about why Republican performances in Minnesota have lagged? It has to do with the composition of the states. Recall Illinois, which we observed moving away from Republicans. We noted that the mega city of Chicago was moving toward Democrats, with the rest of the state moving toward Republicans. But Chicago and various large cities in Illinois comprise 69% of the vote in the states. In Michigan, 55% of the vote was cast in the Detroit metro area (the large city) with the remainder in small cities, towns, and rural areas. In a place like Michigan, the sharp Republican trend in those latter areas packs a lot more punch.

The two Republican-leaning states in the region – Indiana and Ohio – are even more rural, with big cities casting 48% and 51% of the state’s vote, respectively. In other words, the less punch big cities pack, the more Republican the state becomes. The exception is Wisconsin, where metro Milwaukee packs a relatively light electoral punch at just 32% of the state, yet the state remains purple. But recall that Wisconsin was also the state where small cities were overwhelmingly Democratic; because of Madison and its environs, Republicans don’t reap the same electoral benefit from towns and cities that they do elsewhere in the region.

Iowa stands as an extreme case of this effect. It was the most Democratic state in the region in 1988, but by 2016 was one of the most heavily Republican. Its rural regions are not unusually Republican, and its small towns lean Republican, but they were the most heavily Democratic of our group in 2016. What distinguishes Iowa is the complete absence of a mega city or a large city; 26% of Iowa is located outside a metropolitan or micropolitan statistical area.

So where does Minnesota fit into this? It is the second-most Democratic state in the region, nestled somewhere between Illinois and Wisconsin in terms of its politics, and as we might expect, big cities cast the second-largest share of the vote of any state in the region. Minneapolis-St. Paul in 2016 cast about 63% of the vote in the state in 2016. In other words, Minnesota has been tough for Republicans to make gains because it is actually a particularly urban Midwestern state, dominated by a metro area. Without improving their appeal to big cities, Republicans are likely to find it difficult to make solid gains in the Land of 10,000 Lakes.

That still requires us to talk about the Emerson Poll, showing close presidential and Senate races. One possibility is that the poll is simply wrong. We shouldn’t dismiss this possibility, although the credibility of the poll is boosted somewhat by a similar result from the Republican polling firm Trafalgar Group. Another possibility is that Trump’s response to urban discontent is resonating somewhat with voters in the region, giving Republicans a boost in big cities. We’ll know more as more polling comes in, but regardless, for Republicans to really put Minnesota in play, they’ll have to figure out how to appeal to places like Minneapolis.