Trump Questions the US Midterms. Why Do Early Votes Show Up in Big Batches?

Why is Trump questioning the US midterm results?

On election night in Tarrant County, Texas, the first report showed 559,784 early votes in the Senate race. The Election Day column showed zero. By the time the county ran its overnight report at 2:37 a.m., that column showed 136,201 votes. Anyone following the race online saw days of early voting arrive in one update, followed by votes cast in person on November 3.

Six days after the election, President Donald Trump questioned whether the midterms had been honest. In a November 9 Truth Social post, he pointed to polls, large batches of early votes and voting machines. He had told The New York Times on election night that he was willing to accept the results. Democrats won control of both the House and Senate, according to the Associated Press’s race calls. Trump’s post gave no specific machine or tally discrepancy to check. Tarrant County, which he did not name, offers a clear example of how the numbers reach the public.

How can half a million early votes appear at once?

Start with someone who voted a week before Election Day. Their ballot is already in the county’s custody when the polls close. Add all the other people who voted early and a county can have a substantial tally ready for its first public update. In Tarrant County, the first report also listed 18,665 absentee votes in the Senate race, in a separate column. The zero in the Election Day column meant that those votes hadn’t appeared in that report.

This is where a live results screen can give the wrong impression. The time next to an update tells you when a batch of numbers became visible. The people in that batch may have voted days earlier. Reporting schedules vary by state and county, so the order you see in Texas can look different elsewhere.

Why can a candidate’s share fall even as their vote total rises?

Take Democrat James Talarico. Tarrant’s first report gave him 345,834 votes, or 59.79% of the Senate votes reported so far. At 2:37 a.m., he had 427,879 votes, but his share was 58.72%. He had picked up more than 82,000 votes. The pool of votes had grown faster than his share of it because the later additions split differently. These were two unofficial snapshots of one county in a statewide race.

A poll tells you what a sample of people says it will do. A partial tally tells you how the voters included in that update actually voted. Neither tells you on its own how the remaining ballots will split.

Has Trump questioned the count like this before?

In 2016, after losing Iowa’s Republican caucuses to Ted Cruz, Trump said Cruz had stolen the contest and argued that this explained why “all of the polls were so wrong.” Later that year, Trump won the Electoral College but lost the popular vote. He claimed without evidence that millions had voted illegally.

In 2018, Trump urged Florida to stick to its election-night numbers in tight races for governor and Senate. He alleged that ballots had appeared out of nowhere and that others were missing or forged. Officials were still counting mail and provisional ballots and conducting mandatory recounts. Republicans Ron DeSantis and Rick Scott ultimately won those races.

In 2020, Trump claimed the presidential election was being stolen as mail ballots were counted. Election-security officials said they had found no evidence that voting systems had deleted, lost or changed votes. In Georgia, a statewide hand tally confirmed Joe Biden had won the state. In 2024, Trump alleged cheating in Philadelphia before the result was known. As returns showed him winning the presidency, he dropped the allegations and celebrated the result, AP reported. The cases span his own races and races in which he was backing other Republicans.

What if a voting machine really does have a problem?

Arizona offers a real example. During the 2022 midterms, some Maricopa County printers produced ballots too faint for on-site scanners to read. Trump seized on the disruption as he claimed the midterms were being stolen from Republicans. A later independent review found that some printers couldn’t consistently maintain the heat needed to print clearly. The county said ballots that could not be read on site were taken to its elections department and counted there.

How do officials check what the machines counted?

The first check is basic accounting. How many voters were recorded? How many ballots were counted? How many are pending or were rejected, and why? Tarrant County publishes a reconciliation form that records those categories. The county’s separate hand-count audit results had yet to be posted as of November 9.

South Carolina shows what an audit can look like. Officials randomly select voting locations and contests. In public, auditors count the selected paper ballots before comparing their totals with the machine’s results tape. India uses a related check with a different voting system. Election officials count voter-verifiable paper slips from five randomly selected polling stations per Assembly constituency or Assembly segment and compare them with the electronic count.

The results screen tells you how many votes have been reported so far. If someone says a machine changed votes, the paper record, where available, gives officials a way to check the tally.