Picking the metric where half of them have a `-` blank value seems a bit flawed.

VEP = only counts people who can actually vote.

VAP = counts everyone of voting age including people on visas who aren't allowed to vote.

We moved to VEP because the % of total population who aren't citizens started to grow rapidly. eg. see 2020 vs 2024. The eligible VAP count increased by 12M but actual VEP count only increased by 2M.

edit: looks like there is VEP data for older years though I am guessing its not very accurate: https://election.lab.ufl.edu/voter-turnout/#:~:text=National...

Looks like the 1800s had much much higher turnout! I guess my statement is only valid for the last 100 years.