Thread by @mcpli

Democrats could win a majority in the U.S. House this November, but thanks to gerrymandering, doing so would require a near unprecedented win in the national vote. @lauraroyden and I explain in today's NYT: nytimes.com/2018/03/25/opi… #fairmaps 1/

For more detail, here's the study that @lauraroyden, @yurijrudensky, and I wrote looking at the impact of extreme gerrymandering on the 2018 midterm. #fairmaps 2/

We find that to be favored to win back a majority, Democrats would need to win the national vote by nearly 11 points (i.e., 55.5% to 45.5%). That would be a larger margin than either the 2006 D wave or the 2010 R wave. #fairmaps 3/

Which is not to say, it couldn't happen. It absolutely could. Lots of signs that 2018 is an unusually pro-D year. But a win by that large a margin could net Ds only a thin majority. #fairmaps 4/

And 2018 aside, if 2020 proves to be a more normal year - even a still pretty strong D year but one closer to the election range you expect, gerrymanders could come back to bite with a vengeance. #fairmaps 5/

This stands in contrast to Rs lucky break in the 2010 wave. Not only did Rs win a huge number of seats (63 in all), redistricting took place the next year, enabling them to lock in those gains in states like NC, PA, and elsewhere. Ds may not have that luck. #fairmaps 6/

But while undoubtedly many will focus on the 2018 election angle, there are several bigger stories in our report. Take for example, the strong evidence of how much fairer the map drawn by California's independent redistricting commission is. #fairmaps 7/

This chart from the report compares California infamous bipartisan incumbent protection gerrymander from the 2000s (light blue) with the commission map (dark blue). Under the gerrymandered map, each party safely gets its seats. #fairmaps 8/

By contrast, under CA's commission drawn map, lots of seats are in play at every level. As a party (D or R) wins more votes, it wins more seats. Basically what most people think democracy should more or less look like. #fairmaps 9/

New York's court-drawn map also is highly responsive to electoral shifts. As a party wins more votes, it wins more seats. #fairmaps 10/

The responsiveness of New York's court drawn map is a powerful rejoinder to the argument that maps favor Rs because all the Ds live in cities. NY has large & dense concentrations of Ds in NYC, but map is still highly responsive. #fairmaps 11/

That's in stark contrast to the New York senate map, which was the product of a backroom gerrymander deal - but more on that later. Let's stick with Congress for now. #fairmaps 12/

Another big story out of the report is how adverse treatment of minorities is often a key to creating unresponsive maps (i.e. ones where winning more votes doesn't net a party more seats). Consider Virginia. #fairmaps 13/

The Virginia congressional map drawn by Republicans is shown in the chart in light blue. The map drawn by a court-appointed special master is in dark blue. #fairmaps 14/

Under the GOP map, African-American voters were crammed (packed in redistricting lingo) into a district stretching from the Tidewater region to Richmond. Federal courts struck that down as a racial gerrymander & ordered map redrawn. #fairmaps 15/

When that happened, not only did African-Americans get treated more fairly, but lo and behold the map became more responsive, enabling each party to win more seats as its statewide performance improved. #fairmaps 16/

Texas is another example of this. The GOP drawn map failed to create new minority districts and that failure was key to producing an unresponsive map. The court-ordered tweaks partially fixed this, but Texas' map is still pretty unresponsive. #fairmaps 17/

By contrast, a map proposed by Latino groups would have increased Latino representation more significantly (Latinos after all made up 2/3 of Texas' population gain last decade), but also produced a much more electorally responsive map. #fairmaps 18/