HOW WE REPORT
Editorial standards
An outsider perspective. All sides heard. None endorsed. We aim for accurate, unbiased reporting that lowers the temperature and helps readers understand competing arguments. Our commitment is to evidence, fair treatment and clear explanation. These standards apply to our articles, headlines, summaries, profiles and issue coverage.
Accuracy before persuasion
We offer an outsider perspective on politics. Our aim is to be an unbiased, accurate and less inflammatory source of news, data and explanation: genuinely engaging with all substantive sides while endorsing none. We approach partisan camps with independence and curiosity. This is an editorial stance, not a claim that we are free from all bias or that our writers are foreign to the countries they cover. We apply the same standards to every party, ideology and public figure. We distinguish verified facts, attributed claims, analysis, forecasts and opinion. We do not advocate for a party or candidate.
Fair attention to competing arguments
We give substantive left-wing and right-wing arguments comparable space, depth, prominence and scrutiny when both are relevant. We explain their strongest arguments, supporting evidence, tradeoffs and credible objections. We include independent, minority and affected-community perspectives and disagreements within each camp. We disclose missing responses rather than inventing them. Equal attention does not require equal factual credibility: conclusions follow the evidence.
Evidence, attribution and uncertainty
We prioritize original documents, official datasets, research and court records, with dates and context. Contested claims are attributed and checked independently. Allegations remain distinct from findings; legal outcomes and responses are stated accurately. Numbers need denominators, time periods and relevant uncertainty. Missing evidence, data limitations and material corrections should be explicit. Market prices are contract prices, not polls or guarantees.
Public opinion and online reactions
Claims about what most people think require credible representative polling or surveys. We identify the original source, field dates, population, sample size, question wording, methodology, sponsor and uncertainty, and explain material differences between polls. We distinguish adults, registered voters and likely voters. Selected posts, likes, comments and platform polls do not establish mass opinion or consensus. Without representative evidence, we describe selected online reactions and disclose that limitation.
Social posts in context
Public X, Bluesky, Threads and Reddit posts are valid qualitative sources of expressed public sentiment and can illustrate contrasting viewpoints. When describing sentiment within a country, we prioritize domestic voices and distinguish foreign commentary, diaspora perspectives and organized advocacy where relevant. Domestic context requires credible public self-identification or documented affiliation; names, language, ethnicity and location badges alone do not establish residence or nationality. We disclose uncertainty about location. We verify original links, authors, dates, quotations and thread context, and explain selection methods and platform limitations. A post establishes that an account expressed a view; it does not establish that the claim is true. We do not invent posts, engagement counts or sentiment percentages. We avoid unnecessary identifying details about ordinary users. X posts can be embedded; Bluesky, Threads and Reddit posts currently appear as attributed excerpts or paraphrases with original links. Context must remain understandable when an embed is unavailable.
Rumors and unresolved claims
A rumor is not false merely because it has not been verified. When a claim is materially relevant and its circulation is documented across independent examples, we may describe it as a widely circulating, unverified rumor whose truth or falsity cannot currently be established from available evidence. We identify the platforms, country context and observed time period, explain what is known and unknown, and include verification efforts and relevant responses. Popularity is not proof. One post does not establish widespread circulation. Reliable confirming or contradictory evidence takes precedence over an unresolved label. We avoid unnecessary amplification of private or defamatory allegations and update the status as evidence changes.
Reasoning and logical fallacies
We examine every side by the same standard. When an argument contains a clear logical fallacy, we identify the sourced argument and explain the inference error. A disputed premise, factual error or value judgment is not automatically a logical fallacy. Labels are not insults, and a flawed argument alone does not disprove its conclusion. We interpret arguments charitably and acknowledge plausible alternative readings.
Clear writing with college-level analysis
Our reporting should offer college-level analytical depth while explaining specialized terms and using clear sentences. We remove formulaic openings, empty transitions, clichés, vague attributions and inflated adjectives. Phrases such as “delve into,” “a testament to,” “game-changer” and “only time will tell” should give way to specific facts and direct explanations. We preserve the wording of accurately attributed quotations. A cliché is a reason to review prose, not proof of AI authorship.
Useful articles that are easy to share
Headlines, summaries and openings should state the issue and stakes accurately. Descriptive subheadings and concise, sourced takeaways help readers understand and share the reporting. Qualifications must survive in summaries and social copy. We avoid outrage bait, sensationalism, unsupported certainty and quotations stripped of context. Article imagery must be relevant and properly credited.
Connected coverage and original sources
We link to relevant earlier reporting, contrasting analysis and background articles throughout each story, with descriptive link text. Profiles, race pages, issue guides and methodology can supply further context. We verify destinations and avoid unrelated links, repetition and keyword stuffing. Articles and politician profiles include a section for reviewed alternate sources and perspectives at the bottom of the page. These links identify the publisher, explain relevance and state the review date and research limitations; pages awaiting review say so explicitly. We seek independent reporting and primary evidence, not multiple copies of the same syndicated story. Internal links complement primary-source citations rather than replacing them.
Review and corrections
Before publication, we review claims and citations, competing perspectives, polling limitations, reasoning and language. When evidence is missing, we narrow the claim or withhold publication. Updates should identify material corrections and their effect on the reporting. These standards guide editorial work; they do not constitute an automated guarantee of factual accuracy.
For details about our data, read our market methodology. Explore the newsroom and our issue guides.