Part 5: She Won. Or Did She? Election Truth Alliance — A Deeper Look

Just what are the anomalies and how significant are they?

Author’s Note: As part of this continuing series examining claims that Kamala Harris was the rightful winner of the 2024 election, today I am providing a report on one of the more organized and persistent actors in this space: the Election Truth Alliance (ETA). While not a household name, ETA has been gaining traction among those concerned about election integrity, offering technical, statistical, and legal arguments for why they believe the election outcome may not reflect the true will of the voters.

This post is not an endorsement of their conclusions. Rather it is a serious attempt to drill down and understand what ETA is claiming, assess the evidence they’ve presented, and answer a question many of you are quietly asking: If they’re right, why hasn’t the system responded? Let's explore.

Who Is the Election Truth Alliance?

Election Truth Alliance is a nonprofit, self-described “data-driven coalition” of election analysts, attorneys, and concerned citizens. Their website and public filings describe a mission of “restoring electoral transparency through statistical research and grassroots accountability.”

They’ve partnered with university researchers (including political scientist Walter Mebane of the University of Michigan), filed Freedom of Information Act requests to obtain cast vote records (CVRs), and released multi-state reports that claim to identify statistical anomalies and manipulation patterns.

ETA’s work has focused heavily on:

• Pennsylvania (with Mebane’s precinct-level red flag analysis)

• Nevada (particularly Clark County)

• Cross-state comparisons of “drop-off” rates and voting patterns

Their reports are technical and somewhat hard to parse. Below I try to break down and explain their main claims and theories arising out of those claims.

The Mebane Analysis in Pennsylvania

Dr. Walter R. Mebane Jr. is widely viewed as a founding figure in election forensics—a discipline he helped pioneer—blending political science and statistics to detect anomalies suggestive of manipulation He co-developed the Election Forensics Toolkit at the University of Michigan (funded by USAID), which applies multiple statistical tests—including second-digit Benford-style methods—to election data worldwide

That said, no leading expert is without critics. Mebane himself has been part of academic debates—such as over how best to apply Benford’s law tests—which led to a scholarly exchange published in Political Analysis. On some occasions, he has argued against overly simplistic uses of second-digit tests—acknowledging the nuance required in interpreting anomalies

In his latest paper, commissioned in part by ETA, Mebane examined Pennsylvania’s 2024 results and flagged approximately 25,000 votes in several counties that showed “statistical irregularities” inconsistent with normal voting behavior.

What Did He Find?

Mebane did not allege fraud. Instead, he developed a classification model trained on historical precinct-level voting data, designed to flag results that deviate significantly from expectations. Among the findings:

In Pennsylvania, 7,040,360 presidential votes were cast in 2024: 3,543,308 for Trump and 3,423,042 for Harris

Here’s what emerged:

• Depending on model variations, between 25,374 and 225,440 votes were flagged as potentially "distorted" by statistical irregularities—crucially, 25,374 is enough to exceed Biden–Harris’s statewide margin in Pennsylvania

• Of these, 25,374 votes are estimated with high confidence to reflect malevolent manipulation—not just natural voter behavior—according to the most conservative version of his model

He also identified:

• Precincts where Kamala Harris received disproportionately low vote totals relative to down-ballot Democrats.

• Outlier precincts where vote totals broke sharply from historical turnout patterns—even after adjusting for demographics.

• Clusters of zero votes for Harris in areas where affidavits indicate she was selected by multiple voters.

They are red flags—indicators that something may warrant further investigation, particularly if backed by ballot-level scrutiny.

Mebane’s conclusion? He stops short of alleging fraud but argues that “certain precincts show voting outcomes inconsistent with known probabilistic models”—and that further forensic examination of ballots in those areas is justified.

Clark County, Nevada: The Other Hotspot

ETA has also raised alarms about voting patterns in Clark County, Nevada, particularly the presence of vote spikes and drop-off discrepancies.

Their claims include:

• Sudden late-night vote surges in favor of Trump that they say resemble a “tailing pattern” more typical of digital manipulation than organic turnout.

• Tabulator-level anomalies: Specific voting machines allegedly show skewed results, inconsistent with nearby machines and precincts.

• Absentee ballot concerns: ETA claims unusually high rejection rates for mail-in ballots favoring Harris, and discrepancies between CVRs and official tallies.

As with Pennsylvania, the argument is not that fraud is proven—but that there are significant anomalies warranting further investigation to explain the anomalies.

So Why Hasn’t the System Responded?

The question that begs to be answered is this: If these red flags are genuinely concerning, why no recounts? Why aren’t Democratic officials or election boards stepping in?

Two theories emerge:

1. Institutional Inertia and Political Fear

ETA and its allies argue that officials fear the optics of challenging results—especially after the trauma of 2020. Election boards are risk-averse. Democrats don’t want to “sound like Trump.” And many jurisdictions lack the infrastructure or legal mandate to investigate based on statistical red flags alone.

They also point to limited access: Without legal standing or a court order, it’s difficult to obtain full ballot images or logs—especially when equipment is leased under contracts with private vendors.

2. The Bar Hasn’t Been Cleared

The counterargument is that ETA’s findings, while provocative, don’t yet meet the threshold for triggering official action. Critics point out:

• Mebane’s model is statistical, not forensic. It identifies possible issues but doesn’t demonstrate specific acts of tampering.

• No hard evidence (missing ballots, logs of intrusion, altered software, whistleblower declarations) has been presented.

• State audits in both Pennsylvania and Nevada have occurred—and no official body has found fraud or corroborated the core ETA claims.

In other words: the system isn’t ignoring ETA. It’s asking for more.

What Would Move the Needle?

For ETA’s work to drive real change, they likely need to:

• Gain physical or legal access to ballots, logs, or CVRs in the flagged precincts. This is not easy and they are working on it.

• Bring whistleblowers forward from inside election offices or vendor networks. If fraud exists, then there are fraudsters who know what happened.

• Secure legal standing to compel a recount or forensic review. This too is not easy, but they appear to be working on it.

• Replicate findings across states to establish systemic patterns that help distinguish localized noise from national signal,

Final Thoughts

Final Thoughts

After a deeper dive into the work of Election Truth Alliance, I’m left with the view that they are performing a valuable and necessary public service: applying technical rigor and open-source transparency to examine the integrity of our most fundamental democratic process. They haven’t proven that the 2024 election was stolen—but they have raised serious, data-driven questions that merit attention.

What sets them apart is their approach: no theatrics, no overt partisanship, and a willingness to share their data, explain their methods, and invite scrutiny. That’s rare—and it’s welcome.

It’s important they keep going: continue publishing, refine their models, invite constructive critique, and stress-test their conclusions against alternative explanations. If they succeed in proving fraud, it will be a landmark achievement. If not, their work may still help strengthen public trust by exposing vulnerabilities and pushing for reforms.