I thought an interesting Wisconsin analysis would be to look at how the State Senate (upper chamber) and State Assembly (lower chamber) have changed overtime with respect to presidential election results. Both chambers were drawn to be (and still are) Republican gerrymanders), but in 2012, when the current maps were first put into place, the State Assembly seemed to be the more egregious gerrymander of the two. While Barack Obama failed to carry a majority of districts in either chamber (while winning statewide by almost 7 points), he did worse in the State Assembly (where he carried about just 43% of the districts) than in the State Senate (where he carried about 48% of the districts, which is actually just one district short of a majority). However, the opposite was true for Hillary Clinton (in 2016), who carried about only 30% of State Senate districts, while carrying a slightly higher 36% of State Assembly districts. (Given that Clinton narrowly lost Wisconsin, it’s not surprising that she carried far fewer districts than Obama in either chamber.) Joe Biden, who narrowly won statewide, did similar to Clinton, but carried one additional district in each chamber. (To be precise, he carried two assembly districts that Clinton lost, but lost one that she carried.) All of the legislative districts that flipped from Obama to Trump in 2016, voted for Trump again in 2020.
Clearly, the Republican gerrymanders in both chambers have held up very well (and have gotten worse for Dems in the Trump era) but what has caused the State Senate map to now become worse for Dems than the State Assembly map? In Wisconsin, each senate district is geographically comprised of three assembly districts, so each assembly district is entirely contained in one senate district. Let’s look at the number of “cross-party” assembly districts with regards to the 2012 and 2020 presidential elections. By “cross-party” I mean an assembly district (AD) that voted for a different president than its corresponding senate district (SD).
With regards to 2012, there are nine Romney ADs contained in an Obama SD, but only four Obama ADs contained in a Romney SD. This explains why the Dems were (initially) not as disadvantaged in the upper chamber, since more Romney assembly districts become absorbed into an Obama senate district than vice versa, negating the effect of some of the Republican gerrymandering in the lower chamber. It’s worth noting that there are three Obama SDs that each contain one Romney AD, and three Obama SDs that each contain two Romney ADs. The fact that Obama failed to carry a majority of assembly districts (2 or more) in several of the senate districts he won is more evidence showing that the assembly gerrymander was (drawn to be) worse for Democrats. By contrast, Romney carried a majority of the nested assembly districts in all of the senate districts that he won. (The four “cross-party” Obama ADs are all contained in different senate districts.)
In the case of 2020, however, there are five Trump ADs contained in a Biden SD, but eight Biden ADs contained in a Trump SD, so now the Dems are at even greater disadvantage in the upper chamber, since more Biden assembly districts are absorbed into Trump senate districts than the other way around. It’s worth noting that a lot of the “cross-party” Romney and Biden assembly districts are located in Obama-Trump senate districts. The mere fact that the senate districts flipped from Obama to Trump has disadvantaged the Dems in the upper chamber, since the Republican parts of the senate districts are now red enough to offset the Democratic parts (which are still large enough for an assembly district), whereas in the past they weren’t red enough to offset the Democratic parts (but were red enough for a Republican assembly district, or two in more gerrymandered cases.)
Interestingly, the results of 2018 midterms fall somewhere in between 2012 and 2016/2020 in terms of which WI chamber disadvantaged Dems more. Tony Evers, who narrowly won statewide, didn’t do significantly worse in one chamber than the other (in each chamber he carried about 36% of the districts.) Likewise, Tammy Baldwin, who won by a large enough margin statewide to carry a majority of districts didn’t really do much better in one chamber than the other. (She carried around 55% of the districts in each chamber, which is actually close to her statewide win percentage.) This makes some sense, given that the Dem/Republican coalitions in 2018 were also somewhere in between 2012 and 2016/2020.
It’s worth noting that Democrats are already at a geographic disadvantage in WI (with Dems naturally packed into the Milwaukee and Madison areas) and this has gotten worse with urban areas continually becoming bluer and rural areas becoming redder. It also doesn’t help the Dems that the parts of state that are trending blue the most (Milwaukee suburbs) were amongst the reddest parts of the state to begin with. There are small pockets of the Democratic strength outside of the Milwaukee and Madison areas big enough for an assembly district, but not a senate districts, since the surrounding areas are super red.
Thanks to Daily Kos Elections for their 2012, 2016, and 2018 statewide calculations and Aaron Moriak (twitter handle @MaronAaoriak) for all the 2020 statewide calculations.