Louisiana Congressional Redistricting Analysis

Analyzes Louisiana's 2026 congressional redistricting bill by comparing the original and amended maps against 5,000 simulated neutral redistricting plans.

Data

The following files are required to run the script:

• LA_2026_01_BLOCK_DATA.txt: block demographics
• LA_2026_01_VTD_DATA.txt: precinct demographics
• 2026 Precinct Shapefiles (01-27-2026)/: precinct geometries

Block equivalency files for SB 121 (original and amended) are downloaded automatically from the Louisiana Legislature's redistricting portal.

What It Does

1. Assigns each census block to its district under both the original and amended SB 121 plans.
2. Aggregates demographics and voter registration to the district level.
3. Simulates 5,000 neutral redistricting plans using redist_smc.
4. Compares both SB 121 plans against the simulated ensemble on metrics including compactness, population deviation, Black VAP share, partisan lean, and parish splits.

Analysis

In this data, there is a single Black-majority district (district 2) in terms of total population (60% of the total population of the district), voting-age population (58%), and registered voter population (59%). This is also the only minority-majority district, with minority voters compacted into this district. This is also the only Democratic-majority district, with the other districts not being competitive (all less than 35% democrat).

Using 5,000 simulations we see:

• Obvious racial gerrymandering, packing minority groups into a single district. District 2 is at the ceiling in both plans (100th percentile BVAP), meaning no simulated plan draws a district that is as concentrated.
• This reengrosed, and the original plan sits at the 22nd percentile for Democratic seats, meaning roughly 78% of simulated plans produce more Democratic seats than either reference plan. Both plans are Republican-favorable relative to the neutral baseline.
• Both plans are at the bottom for county splits (near 0th percentile), meaning they split far fewer counties than typical simulated plans. This reflects the intentional preservation of many, but not all, county boundaries (as is required). It is worth noting that the simulation does NOT take this into account.
• Both plans are at the 100th percentile for dissimilarity, meaning they produce more racially segregated districts than virtually every simulated plan. This is consistent with the packing pattern visible in District 2's BVAP.