Hidden GOP trick threatens to make Dem dream a mirage

It’s Election Day and, if you believe the polls, both Democrats and democracy itself are heading for a very good night. The final generic ballot averages show Democrats about 9 points ahead, which is the widest margin by which either party has carried a midterm in years.

And while I hope those numbers hold up — or even turn out to be conservative — I’m haunted by the last week of October of 2024 when Kamala Harris had a solid 4-point lead over Trump in the national polls but still lost the election by an average of 2.7 points, a spread of almost 7 percent. Fully 60 percent of the 81 national polls taken in the final two weeks before the election had Harris winning big.

According to the pundits, there are a zillion “reasons” for the big spread between the polling and the final vote (she also lost the popular vote by about 1.5%), but the biggest variable has never been included in the polling and should be: voter purges by Republican secretaries of state in Red states with large Blue cities.

The pollsters blamed the usual suspects: bashful Republicans, people who don’t answer the phone, Hispanic voters were mis-modeled, people who skipped 2020 finally showed up for Trump.

But there’s a much simpler explanation for this “red shift” sitting in plain sight that Greg Palast reported here in January, 2025: “Trump Lost. Vote Suppression Won.”

Greg used the federal US Election Assistance Commission’s own official numbers and found that 4,776,706 voters were wrongly purged from the rolls in the months ahead of that election. More than 2.1 million mail-in ballots were challenged by Republican poll watchers claiming “irregularities” like mismatched signatures (people forgot to include their middle initial, signed in a hurry, etc.: the GOP trains their poll watchers to look for these mail-in ballot “errors” in Blue cities). And roughly 2.1 million people cast a “provisional” ballot thinking their votes would be counted but they never were.

Based on those official numbers from the US government and the states, he calculated a “suppression factor” of 2.3% of the total 2024 vote, and that hit was almost entirely concentrated among Black, Hispanic, college town, and urban voters; without that voter suppression, Harris would have easily carried Wisconsin, Michigan, Pennsylvania, and Georgia and today would be President of the United States.

That 2.3% spread that gave the election to Trump (and to Republicans in the House and Senate) is pretty much identical to what the polling industry is still baffled by today because they’ve never, ever bothered to include voter purges (sanctioned by five corrupt Republicans on the Supreme Court) in their math.

This is the new Republican strategy that has exploded since 2000: simply knock out Democratic voters’ ability to vote or have their votes counted while claiming you “must do it to prevent non-citizens from voting.”

The simple reality is that a poll is merely a survey of people who say they intend to vote; Pew ran a validation experiment that matched people in Philadelphia who’d told them they were registered and were going to vote against the actual voter rolls. They found that only 86 percent of the people who claimed to be registered were, and the situation was far worse among Democrats than Republicans.

That was before the modern Republican purge machine got up to speed; it really kicked in nationally after the 2000 election showed how Jeb Bush’s purges of Black Florida voters just before the election handed the presidency to his brother. (Heavily Black Leon County’s election supervisor later testified that over 50,000 legal voters were disenfranchised by Jeb Bush’s voter purges, and Bush “won” by 537 votes)

The Brennan Center tracks the trend of purging voters and found that states removed over 19 million voters just in the 2-year period between 2020 and 2022. That was a mind-boggling 21 percent jump in purges from the 2-years of 2014-2016, which, itself, was an even more amazing 33 percent jump in the number of voters purged during the 2-years of 2006 to 2008.

Republicans are now mostly doing this by what used to be called “caging,” a strategy that was illegal until 5 corrupt Republicans on the Supreme Court legalized it in theHusted v. A. Philip Randolph Institutecase. They mail out a postcard or letter — primarily into big cities and college towns — that looks like junk mail and if the voter doesn’t respond to it and misses a single election they get purged.

The federal Election Assistance Commission’s own data brief for 2024 reports that between the 2022 and 2024 elections Red states mailed out 38.8 million confirmation postcards and removed 21,298,175 voters from the rolls, roughly one out of every ten of the 211 million Americans who were registered. That’s tens of millions of postcards mailed, up from zero in 2000 (Jeb Bush had used the Texas felon list for his 2000 Florida voter purge).

While the polling companies haven’t figured this out yet, the Democrats on the Supreme Court are clear about what’s going on.In her dissent in the Husted case, Justice Sotomayor wrote:

“It is unsurprising in light of the history of such purge programs that numerous amici report that the Supplemental Process has disproportionately affected minority, low-income, disabled, and veteran voters.

“As one example, amici point to an investigation that revealed that in Hamilton County, ‘African-American-majority neighborhoods in downtown Cincinnati had 10% of their voters removed due to inactivity’ since 2012, as compared to only 4% of voters in a suburban, majority-white neighborhood.”

But the five Republicans on the Court prevailed, and Ohio’s purger-in-chief John Husted hailed his victory and this new way of purging Democratic-leaning voters as, in his words, a “model for other [Republican] states” to follow.

And follow it they have.

Most of the people purged never find out they’ve been removed from the polls. There’s no requirement to notify you, even though voting is supposed to be a “right” (it’s mentioned 5 times in the Constitution).

If a government official wants to take away your gun, they have to go to court and give you an opportunity to fight the process because owning a gun is, according to five corrupt Republicans on the Supreme Court, a right. But in their minds that doesn’t apply to your right to vote.

So, as Demos notes, most people don’t discover they’ve been purged until they try to vote, and then it’s too late to fix it. If the voter is persistent they’re given a “provisional ballot,” but in most Republican-controlled states those ballots are never even counted unless they’re “cured” by the voter showing up at the Secretary of State’s office in person with ID. (Stacey Abrams organized a massive effort to get voters who’d been purged and thus given provisional ballots into Georgia state offices to “cure” them, which is why Jon Osoff and Rev. Warnock are in the Senate today.)

When the voters aren’t counted, even though they told a pollster they thought they were registered, the pollsters don’t attribute it to the purge but, instead, to voter error or forgetfulness. It’s insidious.

— In 2016, Michigan’s then-Republican Secretary of State Ruth Johnson accepted a purge list of 449,922 voters that Kris Kobach’s Interstate Crosscheck “double voter” list said should be purged; the program was later found to have treated Michael Louis Brown and Michael James Brown as the same person. Following that purge, Trump carried the state with 10,704 votes and, again, the pollsters were baffled as to what had happened.

— Similarly, when Brian Kemp was Secretary of State of Georgia and controlled the voting list in that state he ran against Stacey Abrams for Governor in 2018. In 2017 alone, his office purged nearly 670,000 voters and froze another 53,000 voter registration applications that Abrams’ group had brought in: 70 percent were from Black Georgians. He won the election by 54,723 votes, again confounding the pollsters.

APM Reports tracked down people Kemp had purged and easily found at least 70,000 of them hadn’t moved or died; they were purged for no good reason other than to hand Kemp the governor’s office.

This year the purge machinery has been running full-tilt-boogie. Texas put 1.2 million voters on their “suspense list” this summer, 300,000 of them in heavily-Black Harris County (Houston) alone. Talarico won’t know what hit him, as neither the media nor the polling companies seem to have any interest in this new GOP strategy.

In Georgia, Republican Secretary of State Brad Raffensperger (pronounced “Raff-ens-purger”) purged 471,000 people from the rolls in the past 12 months, so you can expect the Georgia polls to be that far out of sync from tonights voter totals, too.

Trump’s Department of Homeland Security, run by former plumber Markwayne Mullin, has their new “SAVE” program that’s another massive purge list and it’s already been found by a federal judge to be wildly inaccurate and heavily biased against Democratic voters.

Rachel Maddow reported last night on MSNBC that this list was compiled with names stolen from federal databases by Elon Musk’s pimply-faced hackers under the rubric of DOGE making government “more efficient.” In actuality, they were compiling a federal list of voters acceptable to the GOP that they are now trying to force on the states.

That SAVE purge effort was paused for a while, but the 6 Republicans on the Supreme Court just gave it the green light over Justice Jackson’s loud objection, her pointing out that it would wrongly flag lawful voters. That, of course, was the whole point, and now they’ll be pointing their purge machine at the 2028 election if nobody raises a fuss.

The voter purge, now endorsed by corrupt Republicans on the Supreme Court, is the GOP’s single most effective way of “winning” elections, and if the purge numbers were factored into the polling statistics we’d have actually accurate estimates of who’s going to win or lose upcoming elections.

While purge numbers typically aren’t available until a year or two after an election, pollsters are statisticians: they should be able to easily see the trend lines here and apply them to their polling numbers. After all, they already weight their numbers for education, party registration, gender, race, and a dozen other variables. This isn’t rocket science.

When the election results come in tonight, many of the pollsters will, over the next month or so, “adjust” their polls to reflect what actually happened. But this week we can compare the actual voter tallies with the polling tallies and the spread between the two will be the result, more than anything else, of the purges Republicans have carried out over the past two years in the states they control (along with mail-in ballots’ signatures challenged by Republican poll-watchers in Blue states).

That’s Palast’s “suppression factor” and we voters deserve to know what it is.

Outlawing this crap, of course, will take a change in the Supreme Court or congressional legislation; the John Lewis Voting Rights Act would have gone a long way toward that in 2022 if Kirsten Sinema and Joe Manchin hadn’t stabbed us in the back by refusing to break the Republican filibuster.

In the meantime, until the polling industry starts reporting the purge gaps, we’ll be hearing about “surprise Republican over-performance” in the elections.

If the GOP’s purges over the past two years are between 2 and 3 percent like they were in 2024, they may not be able to overcome the unpopularity of Trump’s policies so Democrats will still win. But you can bet they are already getting into gear for 2028. with their SAVE list and other purge strategies.

It’ll be interesting to see how small or large the purge margin is between the polling and actual election outcomes tonight.