Reading about how swing states are important for the election, I was wondering how safe the "safe states" actually are. So I plugged some numbers into a spreadsheet, and came to some interesting (?) results.
So first, the data. I used the 2020 election results, starting with Turnout_2020G_v1.2.csv (from https://election.lab.ufl.edu/voter-turnout/2020-general-election-turnout/) for number of people eligible to vote (columns D and E). Added the results from https://www.fec.gov/documents/4228/federalelections2020.xlsx (H, I, J, and K calculated from that), and the number of registered voters from https://ballotpedia.org/Partisan_affiliations_of_registered_voters#2021 (F and G). Non-voters L is eligible voters minus total votes (E - L). Democrats M and Republicans N is the bigger of registered and voters (F or H; G or I), to see if that makes a difference in swinginess. Columns P and Q are the results calculated from the table to make sure it works (Maine and Nebraska cancel each other out), row 54 is the sum of the column above.
The results: Columns R are the states where non-voters alone are the biggest group, S adds third party voters to that, resulting in 148 or 156 electors that could vote for anyone. Columns T and U are each of Democrats and Republicans plus non-voters, and here the non-voters could help each party win everywhere, except DC which is safe for Democrats. For funsies I added the last column V that calculates non-voters from the voting-age, not voting-eligible population, resulting in 287 electors for anyone.
Conclusion? Alabama, Arizona, Arkansas, Georgia, Hawaii, Indiana, Louisiana, Mississippi, Nevada, New Mexico, Oklahoma, Rhode Island, South Carolina, Tennessee, Texas, and West Virginia could be "rainbow states" that could send electors for any candidate. The rest except for DC could vote for either Democrats or Republicans, making all of them swing states. And maybe the fear of non-citizen voters (which I think is the majority of the difference between voting-age and voting-eligible people) determining the election results is valid. Or maybe that means that a significant amount of people living in the US and thus being affected by its government are not represented by said government.
The End: Of course that completely disregards non-voter demographics, even if they would vote they're not likely to all vote the same. Still, enough motivated ex-non-voters could turn basically any state into a swing state. One vote of someone who thinks their vote doesn't matter won't change much, but the votes of all who think that way certainly can.
I live in Texas and completely disagree with your point there. Sure things have been gerrymandered to hell, but they’ve done so based on voting trends and not registered voters. We could easily flip even the most gerrymandered districts if people got out. Also don’t forget that the Governor and the President are decided by results that don’t care about Gerrymandered districts.
Texas also allows early voting, and all you need is your drivers license. Will it get 100% of people: no, but enough to make a difference. It feels like you’re falling into the same trap as the OP response here, that unless it’s 100% perfect it won’t make a difference.