Editorial Policy & Methodology

Funded Prediction Trader publishes research about prediction markets, event contracts and funded-trading mechanics. Some of it is about our own product. This page states how the work is produced, what we will not claim, and what to do when we get something wrong.

Articles are published under the FPT Editorial Team byline. That is a real description of how the work is made — research, drafting, fact-checking and editing pass through the same small desk — and it is deliberately not a fictional journalist. We do not invent author personas, we do not publish stock-photo headshots, and we do not attribute work to people who do not exist.

Where an article is about FPT's own product, it says so plainly and carries the same disclosure as every other page on this site.

Every quantitative, regulatory, product-capability, fee, eligibility or dated claim must be supported by a source a researcher actually opened. Sources are recorded with the publisher, the publication date, the date we accessed the page, and the exact claim they support. That record is what appears in the Sources & Method block at the foot of an article.

We prefer sources in this order:

1. Official first-party documentation and data — a venue's own fee schedule, rulebook, help centre or API docs.
2. Regulators and public records — filings, orders, dockets, court documents.
3. Primary datasets and original research — including peer-reviewed work.
4. Reputable financial, technology, sports or general-news reporting.
5. High-quality specialist publications, only where no primary source exists.

Affiliate blogs, SEO listicles and content farms are not used to establish a consequential fact, in any circumstance. Several claims that circulate widely about prediction-market venues turn out to be contradicted by the venues' own published documents; we check the document.

Claims about FPT Scale — prices, profit targets, drawdown behaviour, consistency rules, payout caps, position limits — are checked against the approved offer specification and the typed configuration the website itself reads from. Numbers are imported from that configuration rather than retyped, so an article and the product cannot quietly diverge.

Some things we will not say, regardless of how well they would convert:

• That trading prediction markets produces guaranteed profits, guaranteed payouts, or guaranteed funding. None of those are guaranteed, and the rules say so.
• That an account size is money paid to you or capital you can deploy. Account size is the benchmark that profit targets, maximum loss and drawdown are measured against.
• That FPT is affiliated with, endorsed by, or operated by any venue, league or promotion we write about. It is not.
• Trader counts, pass rates, payout totals, testimonials, ratings or awards. FPT is pre-launch and has none of these; publishing them would be fabrication.
• A launch date or a countdown, which the approved specification does not contain.

Before an article can be marked published, it passes an automated gate that runs in the repository and in continuous integration. The gate has two parts.

Blocking checks. A fabricated or malformed citation, an unsupported financial claim, a prohibited claim, a missing disclosure, a broken canonical URL, a dangling internal link, placeholder text, an invalid date, or a body that is not present in crawlable HTML — any one of these stops publication outright. No score overrides them.

A weighted score. Articles are scored across 9 dimensions and must reach 82 out of 100 to publish. The heaviest weights are on factual accuracy and source support, reader-intent satisfaction, and originality — in that order.

Articles carry a publication date, and an updated date when the piece has materially changed. Where the subject can move — venue fees, product availability, regulatory status — the article also carries a last reviewed date, which tells you when a human last checked the claims against their sources rather than when a word was edited.

We do not bump dates on a schedule to look fresh. A modification date that does not correspond to a real change is a false signal to readers and to search engines, and it makes the honest ones worthless.

If you find an error, tell us and we will correct it. Substantive corrections are made in the article, and the updated date moves with them.

We use automated tooling in research and drafting, and we are not coy about it. What that tooling is never permitted to do: invent a source, invent a statistic, invent a quotation, or assert a product capability that is not in the specification. Every consequential claim is mapped to a source record before an article can pass the gate, and a claim with no record blocks publication rather than shipping with a hedge.

Editorial judgement — what is worth writing, what the angle is, and whether a piece is honest — stays with people.

FPT sells evaluations for funded prediction-market accounts, and this blog exists partly to introduce people to that. We think the honest version of that arrangement is: publish work that is worth reading whether or not you ever buy anything, be explicit about where the product fits, and keep the sales pitch out of the parts of an article that are supposed to be teaching you something.

If a piece fails that standard, it is a defect. Read the rules if you want the product detail, and start here if you want the research.