Two thirds of prop firms publish no dated evidence that they have ever paid a trader

"Do they actually pay?" is the first question a trader asks about a prop firm and the last one the industry answers in writing. So we asked a narrower version that has a yes or no: has this firm published, itself, a payout with a date on it?

For 23 of the 71 firms we track, yes. For the other 48 — 68% — we could not find one. Six of those 48 refused to let us read their site at all, which leaves 42 firms whose disclosure we actually saw.

The interesting part is not the count. It is what sits in the space where a dated payout would go, because it is not nothing. It is usually something quite large, quite gold, and quite carefully undated.

The undated counter

18 firms print an aggregate amount paid and attach no date to it. Tradeify's hero reads "$250M+ Verified Payouts". OANDA Prop Trader publishes "$650M+ Paid in rewards worldwide". Top One Futures says "Over $28 million verified payouts to date" — a sentence that uses the phrase "to date" while supplying no date, which is the whole pattern in five words.

A lifetime total with no as-of stamp cannot expire. It was true when it was written and it stays true forever, because there is no moment it is a statement about. Put a date on the same figure and it becomes checkable, ageable, and eventually embarrassing — which is the entire difference between a disclosure and a decoration. Nothing about an undated figure implies it is false. It implies nobody can ever tell.

Two firms show what that costs. Goat Funded Futures prints "$25M+ In payouts across the brand" in one section of its homepage and "$23M PAID OUT ACROSS THE BRAND" in another. Both are on the same page. Neither is dated, and because neither is dated there is no way to say which is the newer number, or whether one of them was simply never updated. Crypto Fund Trader goes the other way and publishes "+$ 21270544 PAID TO CFT TRADERS", precise to the dollar and anchored to no day at all.

The one broken counter in the set is worth naming because it is honest by accident. Lark Funding's total-rewards figure renders as "Approaching $0+" — a template placeholder that appears never to have been wired to a real number.

The undated certificate wall

8 firms skip the aggregate and show you people instead. Phidias is the clearest case. Its homepage renders a strip of gold "Certificate of Payout" cards — $37,500 to Lemuel H., $28,690 to Maximilian H., $28,000 to Alvaro S., about a dozen more. We pulled the markup behind them in a rendering browser, because the site returns 403 to anything scripted. There is no date on any card, no link behind any card, and no certificate page for a card to lead to. The word "certificate" is doing work that no certificate is doing.

The variants are all the same shape. Funded Futures Network runs a rotating toast reading "Recent Payout — Bro***** received $685" with no date anywhere in the markup. Lux Trading Firm publishes "payout proofs" as video thumbnails, which is the hardest evidence form to check and the easiest to leave undated. The5ers, Legends Trading and Funding Traders each show named individual amounts and no dates. Falcon Funded and Tradeify pair a certificate strip with the undated counter above it — the two patterns are not exclusive, and a firm running both has built something that looks exactly like a ledger and works exactly like a poster.

The policy page that proves nothing

13 firms publish a payout policy: the split, the schedule, the processing time, occasionally a claim about denials. Lucid Trading advertises a "15-minute average payout time" and "no payout windows". 10FOUR publishes a 24-hour payout guarantee. P1 Futures publishes a "6hrs Avg. Payout". Ment Funding publishes "Not a single refused payout since 2021".

These are descriptions of a process. A process description is not evidence the process ran, and the denial claim is the sharpest example: a firm reporting that it has never refused a payout is reporting on its own conduct, from its own records, with no amount, no volume and no date. It is not checkable in either direction.

FTUK is the near miss of the whole study. It publishes a page at ftuk.com/payouts, and the page is a legal document — "Payouts / Issue Date: January 11, 2025" followed by the Funded Account Disclaimer. A date, at last, and it is the date the disclaimer was issued. The only figure on the site is an undated homepage counter reading "+$12,000,000". UProfit comes closer still with "13,445 Approved payout requests since 2019": a count and a start date, missing only the as-of that would let anyone age it.

What good disclosure looks like

The bar is low and 23 firms clear it, in four different ways. There is no one format a firm has to adopt. There is only a date.

One firm we expected to name here did not survive checking. Finotive Funding publishes a payouts page our record cites, and on 2026-09-06 it carried individual named amounts and the line "total payouts updated daily" — a claim about how often a figure refreshes, not a date on any payout. A cadence is not a timestamp, so it is not listed above. The claim we cut is exactly the kind this study exists to be careful about.

One page dated forward

Apex Trader Funding does publish an itemised payout table, and it is the reason we hold dated evidence for the firm. On 2026-09-06 we opened it in a real browser and read every row. All 16 of them were stamped Sep 7, 2026 — one day after the date the reading browser held, on rows already marked Approved.

We are reporting that we saw it and nothing else. We do not know why the table prints that date. A timezone offset, a settlement date shown in place of a request date, and a template default would all produce this, and the page does not say which. It is here because anyone auditing that table would see the same thing, and because a dated table being odd is still strictly more useful than an undated counter that cannot be odd about anything.

A check we also failed

Our own repository runs npm run check-payout-proof over the 25 pieces of evidence recorded for those 23 firms. Today it reports 24 findings on 25 records: 21 have no stored snapshot, so no reader can check what the page said when we read it; two pages no longer carry dated payout amounts at all; and Traders Launch's ledger, which is properly itemised, has a newest row 375 days old.

That last one is the argument for dates, made against us. We can tell you Traders Launch's ledger has gone stale only because it is dated. Not one of the 26 undated figures on this page can be described as stale, fresh, or anything else — and neither can our own uncaptured evidence.

Limitations

• An empty field is our search, not the firm's history. Every count here is a count of what we found. It moves when we do research, not only when a firm changes what it publishes: this field held evidence for 2 firms on the morning of 5 September and 23 by the evening, and no firm published anything new in between. Anyone quoting this as "48 prop firms cannot prove they pay" is quoting our search log.
• JavaScript-rendered evidence is invisible to a plain fetch. Fintokei's payout figures read $0 in our captures and settle on a real number only after the page has run and scrolled. Dominion Funding's certificates render only in a real browser. Both were confirmed by hand. Any firm whose evidence works that way and that we did not open in a browser would land in the 48 wrongly.
• A page existing is not evidence. Several of these sites are single-page apps with a catch-all route, so /payouts, /payout-proof and /transparency all return HTTP 200 carrying the homepage. A checker that trusted status codes would have scored those firms as publishing a payout page. We treated a 200 as nothing until the content on it said otherwise.
• Six firms we could not read. Funded Trading Plus, FundingPips and TickTickTrader returned HTTP 403; Blueberry Funded served a bot interstitial; FundingTicks and Sure Leverage Funding returned no readable body. They are counted in the 48 because we hold no dated evidence for them, and nothing is claimed about what they publish. Being unreadable to us is a fact about our request.
• We read the firm's own site, not the whole internet. A firm posting dated payouts only to Discord, X or a review platform would not clear this bar, and that is deliberate — evidence a firm cannot be bothered to publish where its customers buy is evidence a customer cannot be expected to find. It does mean the bar is about publication, not about truth.
• One day, one read each. Sites change. A firm that posts a dated report next week moves out of this list, and we would like several of them to.
• The sample is our list. 71 firms we track, chosen for coverage of the market we write about, not drawn at random from the sector.

Method

For every firm FirmFax tracks we looked for one thing: dated, firm-published evidence that a payout happened. A ledger of payouts carrying dates, a dated payout report, or a dated case study naming an amount all count. A total with no as-of date does not; nor does a certificate with no date; nor does a policy page explaining how withdrawals work.

Where a firm's record already held such evidence, it is listed as proven and the evidence URL is published in the dataset. Where it did not, we read the firm's own site — homepage, sitemap, and the /payouts, /payout, /payout-report, /certificates, /payout-proof, /transparency and /leaderboard paths — and recorded what is published instead, quoting the firm's own words. Firms whose sites resist scripted requests were opened in a real browser. Every one of the 71 firms is named below and in the file.

Download the full dataset, all 71 firms (JSON)

Every firm, and what it publishes

The 48 firms we hold no dated evidence for, grouped by what sits in that space instead. Quotes are the firm's own words as printed on the date shown.