Thursday, December 28, 2023

To make Wisconsin maps fairer, how much should we care about who people vote for?

Since last week's Wisconsin Supreme Court ruling that will require new legislative maps for 2024, I've been doing a bit of noodling at possible maps beyond the stuff I did before.

My original maps are still the ones that I think make the most sense overall, as they are based on keeping similar communities together (so cities like Sheboygan, Wauwatosa and Whitewater don't get split up as they do under the GOP's current gerrymander), has an adequate number of majority-minority districts, and is mostly compact and contiguous. Here's what they look like, as a review, starting with the state Assembly.

And now the state Senate.

I drew these using the excellent Dave's Redistricting website, and you can certainly use the site to do the same for yourself. When using the vote totals from last year's Governor's election, which Dem Tony Evers won by 3.5 points, we can look at which districts Evers got the most votes in, and which districts Republican Tim Michels got the most votes in.

Here's the Assembly distribution with my maps, with the vertical dashed line showing a 50-50 split in seats. I'll focus in on the districts that are within 15 points of each other.

Republicans would still be favored to be in control of the Assembly, but at around a 55-44 level as opposed to the 63-36 advantage the Republicans hold today.

And notice all of those districts that are between around 46.5% to 53.5% of Dem vote (21 in all)? Compare that to the current GOP gerrymander.

Fewer seats between 46.5% and 53.5% (15 vs 21), and instead of the mostly straight line my maps had, the GOP gerrymander has significant dropoffs to protect its members.

The State Senate is a similar story, although this leans harder to the GOP due to the larger district sizes.

This means the "base scenario" under my maps would be 19-14 GOP, but that's not the 22-11 supermajority we have today. And again, note the mostly straight line of 8 districts between 47% and 53.5%. Then compare it to the current GOP gerrymander.

There are half the districts between 47% and 53.5% than there was in my map, and 3 of those 4 favor Republicans (again, in a chart based on Dems getting 51.5% of the vote). Dems would have to pull around 57% of the vote to flip the Senate, instead of the 54-55% they'd need with my maps.

But what if you want a map with most of the districts being competitive and a 50-50 coin flip for control? Dave's has that, in its "most competitive" category. Unfortunately, that type of map would be a complete mess in Wisconsin. Here's the Senate version.

Lots of slicing, This is especially true when you get to the Milwaukee area.

In addition, none of these Senate districts have a non-white voting age population at or above 40%. That would likely be found to be illegal under the Voting Rights Act, given the population distributions in this state. So let's throw that out.

I did some fooling around with my base Senate map, and I found a way to make it near that 50-50 split. But I still maintain 3 majority-minority districts, and I don't think the map looks too contrived.

Among the differences are what you see in the Fox Valley, where District 18 gets more of the Appleton area and gets rid of the parts in Fond du Lac County, and District 19 goes further north into the Oneida Reservation and the western suburbs of Green Bay. Elsewhere in the state, District 24 goes north from Stevens Point into into Wausau, District 13 is moved further south and west. And in the Milwaukee area, District 8 goes back to looking like it does today, with Dem-voting Milwaukee suburbs like Whitefish Bay and Fox Point included, and parts of Waueksha and Washington Counties are taken out.

So what does it look like for competitiveness?

18-15 Republican as a base, but 13 of the 33 districts are in the "competitive" zone, with a near-equal split in which party is favored in those races, and 11 of those 13 have vote margins of 5% or less. That's a situation that wouldn't allow either party to be able to ignore voters, and have to continue to work to get power and/or stay in power.

With only two weeks to go before maps are submitted to the Wisconsin Supreme Court and its two academic consultants, it reminds us that there are a couple of ways we can base "fair maps"on. Do we try to completely ignore how people vote in a certain area, and base districts on similarity of communities and geography? Or do we take partisanship into account, with a goal of having more competitive districts, and maximizing the chance of having control of the Legislature flip if public opinion flips?

I'm going to get back to work on a "competitive" Assembly map, and likely report in the near future on what that gives me.


  1. As i recall, no less an, ahem, authority than Sir Robbin' Vos declared that partisanship was an appropriate criterion in map drawing. So there should be few Republican objections on that score.

    1. Why yes, that is what Robbin' said when presenting the maps. After all, everything is "partisan" in GOPWorld, so why shouldn't we account for outcomes on it?

      And from what I read about the two PoliSci consultants, they agree with using election results as an input on putting the maps together. We'll know soon enough.