A Adam Arafat. WA-10 Congress
← Watchtower

Endorsements Should Survive Audit.

Endorsements are useful to voters only when the underlying record can be inspected. Watchtower's consolidated endorsement-accountability frame uses two fully sourced examples: the AFL-CIO/CLARITY labor record and the Sierra Club/League of Conservation Voters PFAS clean-water record. The point is not to punish endorsers. The point is to make campaign endorsement claims auditable without overclaiming motive.

Last build: 2026-07-10 Sources cited: 8 Graded claims: 5
info@arafatforcongress.org About
Short Endorsements Should Survive Audit. 05:51
Listen options
Transcript Download
TLDR

The safe accountability test is simple: public endorsement, stated mission, public position, official vote.

Empty community forum room with microphone stand and chairs after a rainy public meeting

Endorsements are not decorations. They are claims about trust. When a campaign cites an endorsement, voters should be able to inspect the record behind it: the organization, its stated public-interest mission, the relevant public position, the roll call, and the incumbent's vote.

Watchtower is using a narrow public-record test. First, did the organization endorse the incumbent? Second, what mission or public position is the endorsement meant to signal to voters? Third, how did the incumbent vote on the official record when that mission was implicated? If those records conflict, the story is not motive. The story is accountability. Other endorsement lanes should wait until they have the same complete packet.

The labor example uses the Washington State Labor Council, AFL-CIO 2026 endorsement list, the AFL-CIO's public opposition to H.R. 3633, and the official House vote record for Roll Call 199 on July 17, 2025. Those records are public. They are checkable. They can be read without trusting any campaign's interpretation.

Source-linked endorsement accountability panel
The endorsement-accountability frame keeps the claim narrow: public endorsement, stated mission, official vote, and source links.

The environmental example belongs in the same umbrella. The Strickland campaign endorsement page lists Sierra Club and League of Conservation Voters Action Fund among the organizations supporting the campaign. The PFAS source packet documents the clean-water record voters should be able to compare against those environmental signals: JBLM-area PFAS contamination, cleanup photo ops, the December 10, 2025 FY2026 NDAA vote, outside environmental criticism of the bill's PFAS posture, and a post-vote statement that emphasized JBLM projects without mentioning PFAS.

That does not mean Sierra Club or any environmental organization acted improperly. It does not prove motive, coordination, corruption, or bad faith. It asks the same narrow question as the labor example: when a campaign uses an endorsement to signal trust on clean water or environmental protection, should voters also see the source-linked PFAS record before the endorsement becomes a shortcut?

The answer should be yes. Endorsements can still matter. Organizations can still make strategic choices. Incumbents can still disagree with endorsers. But voters deserve to know when mission, position, vote, and public explanation do not line up cleanly.

That is the Watchtower standard for endorsement accountability: mission, position, vote, source, receipt. It is firm enough to inform voters and restrained enough not to turn every endorsement into an accusation.

Receipts

Methodology & Sources Open source record checks, assumptions, and ratings Open

Watchtower applies open-source forensic doctrine to every Ready story. Sources are graded; competing explanations are tested; the assumption that could break the story is named; what would falsify it is listed.

Estimative Probability

A source-linked endorsement audit can verify endorsement, organization position, roll-call result, and member vote without making motive claims.

Each element is backed by primary public records.

Analytic Confidence
HIGH

The core documents are public, official, and independently accessible. The story's claim is methodological and avoids unsupported inference.

Linchpin Assumption

The endorsement list, AFL-CIO position, and official roll-call record are the correct public records to compare.

Falsifier

A primary correction from WSLC, AFL-CIO, Congress.gov, or the House Clerk showing the endorsement, position, roll-call result, or member vote was recorded incorrectly.

Analysis of Competing Hypotheses
  • leading

    Endorsements can reflect broader strategy even when one vote diverges from an organization's position.

    This is plausible and is why the story avoids claiming the endorsement was improper.

  • rejected

    A vote divergence proves an improper endorsement exchange.

    The public record reviewed here does not prove an improper exchange or motive.

  • less supported

    Voters do not need endorsement audits because endorsing organizations already do internal vetting.

    Internal vetting may exist, but public endorsement claims still benefit from source-linked public receipts.

Premortem: What Could Kill This Story
  • WSLC updates its endorsement list and removes or corrects the 10th CD entry.
  • AFL-CIO updates its H.R. 3633 scorecard or legislative alert in a way that changes the position record.
  • Congress.gov or the House Clerk corrects Roll Call 199 or the member-vote entry.
  • A public explanation from the endorser or incumbent adds context that should be included alongside the receipt.
Source Claims (NATO STANAG 2511 Grades)
  • ENDORSEMENT-AUDIT-METHOD-001 A1 HIGH

    The endorsement-audit method relies on public records: endorsement list, organization mission or position, official roll call, and a source packet. Sources include Strickland campaign endorsements, WSLC, AFL-CIO, Congress.gov, House Clerk, and the Watchtower PFAS packet. Other endorsement categories are intentionally held unless a comparable packet is available.

  • WSLC-ENDORSEMENT-LIST-2026-001 A1 HIGH

    The Washington State Labor Council, AFL-CIO 2026 endorsement list includes 10th CD - Marilyn Strickland. Source: WSLC 2026 Election Endorsements PDF.

  • AFLCIO-HR3633-POSITION-001 A1 HIGH

    AFL-CIO opposed H.R. 3633 and linked the position to retirement and financial-stability risk. Source: AFL-CIO scorecard and July 14, 2025 legislative alert.

  • ROLL199-HR3633-OFFICIAL-001 A1 HIGH

    Congress.gov reports H.R. 3633 passed House Roll Call 199 on July 17, 2025 by 294 to 134, with Democratic members split 78 Yea to 134 Nay. Source: Congress.gov House Roll Call Vote 199.

  • SIERRA-PFAS-ENDORSEMENT-AUDIT-001 A2 HIGH

    The Strickland campaign endorsement page lists Sierra Club and League of Conservation Voters Action Fund. The Watchtower PFAS packet links the clean-water accountability frame to the FY2026 NDAA vote and JBLM-area PFAS record. Sources: Strickland campaign endorsement page, House Clerk Roll Call 320, and Watchtower PFAS story.

How This Was Researched

This story uses only primary public documents for the core finding: WSLC's endorsement list, AFL-CIO's public legislative position, Congress.gov's roll-call summary, and the House Clerk member-vote record.

The analysis deliberately excludes motive, quid-pro-quo, internal endorsement deliberations, and claims about whether the endorsement should have been granted. The publishable claim is narrower: endorsement records should be auditable against public positions and official votes. This version uses the two strongest sourced examples and does not add Stonewall or other community-endorsement lanes without a complete source packet.

How we check our work

Watchtower holds itself to the same standards U.S. intelligence agencies use for their own internal reports. Before any story goes public, it has to pass every check on this page. Here is exactly what each rating means and how we earn it.

How sure are we?

The scale you see at the top of every story tops out at Almost Certain (95 to 99 percent), not "100 percent certain." That is on purpose. Honest analysts do not say "certain" because real evidence has limits. When someone tells you they are 100 percent sure of anything political, that is the warning sign. Almost Certain is the firmest call a careful analyst makes.

What we call itHow confidentWhat it really means
Almost Certain95 to 99%The public record proves it. No serious alternative explanation survives.
Very Likely80 to 95%Backed up by multiple government records or filings that line up.
Likely55 to 80%The record points this way. Some honest uncertainty remains.
Roughly Even Chance45 to 55%Genuinely a toss-up. We tell you straight when we are not sure.
Unlikely20 to 45%Most of the evidence cuts the other way.
Very Unlikely5 to 20%The record strongly contradicts it.
Almost Certainly Not1 to 5%The public record contradicts it. We would not publish a story at this rating.

Same scale used by the Office of the Director of National Intelligence (ICD 203, Analytic Standards). Designed by Sherman Kent at Yale in 1964 after the intelligence community realized that vague words like "likely" meant different things to different readers.

How good are our sources?

Every fact in a Watchtower story is tied to a specific source, and every source is graded the way NATO intelligence agencies grade theirs. Sources get two scores. The first is how trustworthy the source itself is. The second is how solid the specific piece of information is.

Source ratingWhat it meansInfo ratingWhat it means
ACompletely trustworthy. Government filing, court document, official agency record.1Confirmed by independent sources.
BUsually trustworthy. Reputable institution, minor doubt.2Probably true based on what else we know.
CSometimes trustworthy. Use with care.3Possibly true. Cannot fully verify.
DOften unreliable.4Doubtful.
EUnreliable.5Improbable.
FCannot be judged.6Cannot be judged.

A claim graded A1 is the strongest a piece of evidence can be. Watchtower has a hard rule: every released story has to be built on at least one A1 or A2 source. Stories that lean on C-grade or weaker sources stay out of the public release set. You can see every source under the Receipts list on this page, each one tagged with its grade.

Same grading system as NATO STANAG 2511 and Allied Joint Publication AJP-2.1. Used by U.S. and allied military intelligence to evaluate every reported fact.

How confident are we in our own reasoning?

This is different from "how sure are we" above. The probability band rates the conclusion. This rates the analysis itself, the chain of reasoning that got us there. A story can be Almost Certain on the conclusion but only MEDIUM confidence in the reasoning if the path from facts to conclusion has weak links. We tell you both.

  • HIGH Multiple primary sources line up. The chain of reasoning is short and well-supported. Other explanations have been tested and ruled out.
  • MEDIUM Some sources are second-hand, or the reasoning chain has one or two links that someone could reasonably question.
  • LOW Few sources, or the reasoning depends on assumptions we have not been able to verify yet. We rarely publish at LOW.

What else could explain this?

For every story, we write out at least three other ways the evidence could be explained, then say honestly why we do not believe each of them as much as the one we went with. This is meant to keep us honest. The point is to pick the explanation that has the fewest holes, not the one that came to mind first or the one that feels right.

  • leading Our working explanation. It has the fewest unexplained gaps once we run the evidence against every alternative.
  • less supported A plausible alternative. The evidence does not rule it out, but the leading explanation fits better.
  • rejected Inconsistent with what the records show. We can rule this one out.

This technique is called Analysis of Competing Hypotheses. It was developed by a former CIA analyst, Richards Heuer, and is taught in the CIA's own Tradecraft Primer (2009 edition).

What would prove us wrong?

Before we publish a story, we write down exactly what evidence would force us to correct or take it back, and we publish that list right with the story. Look for the Premortem section above. If any of those scenarios later turns out to be true, the Corrections and Updates log on this page records what changed and when. We are betting on the public record being right; if the record changes, so do we.

This technique is called a premortem. It comes from Gary Klein, a decision-science researcher who wrote it up in the Harvard Business Review in 2007. Imagine the project failed; what would have killed it? Then list those things up front.

What is this story betting on?

Every Watchtower story rests on one assumption that, if it turned out to be wrong, would knock the whole thing down. We name it explicitly in the Linchpin Assumption section above, and we tell you what evidence would prove that assumption wrong. You do not have to take our word for any of it. The linchpin is there so you can stress-test the story yourself.

What if we get something wrong?

Spotted an error? Use the corrections form at the bottom of this page. Every correction we accept gets logged publicly with the date, what changed, and why. We do not silently edit. We treat our own track record as evidence too.

Corrections & Updates

No corrections or updates issued since publish. Spotted an error? Use the corrections form below.

Found an error? Have a tip?

Every Watchtower claim is tied to a primary source. If a claim, number, quote, or source link is wrong, tell us with the primary record that contradicts it. Corrections are published with the story; sources are protected.

Send a correction or tip
Embed this finding on another site

Drop this snippet into any HTML page to embed the trust strip + headline as an iframe widget:

Or cite this story directly via citation.txt (BibTeX / RIS / APA 7 / MLA 9 / plain).