How we check firms

The same data feeds our annual industry report, The State of Prop Trading 2026, which publishes every figure with its denominator.

Where the data comes from

Prices and rules come from each firm's own website and help center. Each firm page links its source and shows the check date. The latest check was 2026-11-12.

What we do when we can't confirm something

The value stays empty and is marked. We don't estimate it and we don't copy it from other comparison sites.

Corrections

Every source line on the site carries a "Wrong? Report it" link, and every page has "Report an error" in the footer. A report reaches us the same day. We read the firm's page again, correct the figure within one business day, date the change on the page and list it on the updates page. Firms can also write to info@proptested.com. A figure that was true on the day we read it stays in the change log with its date; what we remove is the error, not the record.

What the labels mean

We read every plan on the firm's own pages and its plan picker. Unconfirmed values show as n/a. A few offer prices are worked out from the percentage a firm states, so confirm the final price at checkout. A firm only gets a Tested label after we've traded a real account there, and no firm carries it yet.

How we rank plans

A low price alone says little: a $14 plan that almost nobody passes can cost more than a $45 one. So we score every setup out of 100 in the same way, whichever firm sells it.

Cost until funded, 45%. We run each plan's own rules, 3,000 times, for a simulated trader with no edge: target, second phase, max loss and how it moves, daily limit, consistency rule and minimum days. That gives comparable pass odds. We then work out the typical spend until passing, including resets, activation fees and two months of fees on monthly plans. It's compared with the same market per $10K of account. Instant accounts are tested on reaching a cushion of one max loss. The odds compare rules; they aren't your chance of passing.

Firm record, 35%. The Trustpilot rating, weighted by the number of reviews so a handful can't match thousands, plus payouts we can see on the blockchain. A Trustpilot guideline warning caps this part. MyFundedPerps has no profile of its own and uses My Funded Futures, the team it names on its site.

Payout terms, 20%. Profit split, funded consistency rule, days until the first payout, and lifetime payout caps.

Firms below 45 on record are hidden from the plan list until you switch off Established firms. Setups with missing core rules can't be scored and sit at the bottom. Invite-only plans aren't listed. Other sorts use only the value you choose, and Lowest cost counts monthly plans as two months. Commissions never enter the score.

Published pass rates and the simulation

Ten firms publish a pass rate of their own, four of them in a footer disclosure that is refreshed every year. PropTested read them on 2026-11-05 and shows each one on the firm's page as a sourced fact. They are not the same measure as our odds, and they are not the same measure as each other. Topstep, Tradeify and Take Profit Trader count every attempt. Topstep and Tradeify also publish a per-person rate, which runs far higher because one person buys several evaluations, a median of three at Tradeify. E8 Markets and Hola Prime count customers, Earn2Trade counts new subscriptions, and My Funded Futures and TradeDay do not say what they count. The FTMO and Alpha Capital numbers come from interviews and press, not from a disclosure.

Our odds are the pass rate of one simulated trader with no edge who follows each plan's rules, so they compare rules for the same trader at every firm. They are not a forecast of anyone's chance. On 2026-11-05 we set the simulated median of each firm's evaluation plans next to the figure the firm publishes.

The rank correlation over the ten firms was -0.18, so the two orderings have nothing in common. That is not a surprise. Only three firms count the same way, too few to fit weights on, and a real pass rate also depends on who buys the plan, which no rule model sees. The simulated spread between firms is narrow, 22% to 35%, so the cost score is driven mostly by price, as intended. We have not changed the weights because of this check, and the odds stay labelled as simulated wherever they appear.

Every published figure we hold, with its period and basis. "Read" means the firm's live page on that day, "archived" a Wayback Machine copy, "published" the date of an article.

Matching

Match results only use your answers and the published numbers.

Money

We may earn a commission through links and codes. Firms can't pay for a better position.