How we rank prop firms (and why it can't be bought)

Every score is computed by a public formula from the data shown on each firm page. Sponsors, affiliate commissions and featured listings have zero input. The code is in build.py → score().

Payout speed (25)

Daily / on-demand = full · 48h, 5 days, 7 days or weekly = 70% · bi-weekly, or several qualifying days first = 40%.

Profit split (25)

Best advertised split: 100% = full · 90% = 90% · 85% = 80% · 80% = 70% · lower = 60%.

Drawdown model (20)

Any EOD or static option = full · trailing/dynamic only = 55% · intraday only = 35%.

Consistency rule (15)

At least one plan with no consistency rule = full · rule on every plan = 50%.

Reputation (15)

Trustpilot rating ÷ 5, as displayed by the firm (to be replaced by direct Trustpilot reads).

Tie-breaks

Firms with equal scores are ordered by (1) data completeness: the share of challenge fields verified on the firm's own pages, higher first; then (2) the number of confirmed changelog entries, more first (a firm whose changes we can track publicly is easier to hold to account); then (3) name, A–Z.

What "payout speed" measures

We score the published payout eligibility rule: how soon after trading you're allowed to request money. Daily, on-demand or anytime requests score full marks. Fixed 48-hour, 5-day, 7-day or weekly cycles score 70%. Bi-weekly cycles, or rules that need several qualifying winning days first, score 40%. Fast processing after you request doesn't change this score.

Missing data is never scored as zero

If we can't verify a component, it shows "Insufficient data" and is left out: the score is the weighted average of the components we could verify. A firm is ranked only when at least 4 of the 5 components are verified and at least 50% of its challenge fields are filled. Otherwise it's listed as Unranked: insufficient data, below ranked firms, until we can verify more.

Data rules

Every fact links to the page it came from and the date we checked it.
If we can't verify a field it shows "Unknown". Unknowns are excluded from scoring, not counted as zero.
We never label firms as scams. Documented disputes are linked neutrally to their sources.
Changes are detected by snapshotting each firm's rules pages and diffing the text; a human confirms before logging.