How we rank prop firms: the 2027 methodology
The ranking scores each prop firm on 25 public criteria, grouped into five weighted pillars. Here are the entry conditions, the scale for each criterion, its sources and how the score is calculated.
2027 edition · updated October 29, 2026 · next updates planned for December 2026 and January 2027 · Ranking methodology
Methodology version 1.0, 2027 edition.
Principles
- A score comes from a scale, never from an opinion. Every criterion has written levels: two people rating the same firm reach the same score.
- Every score carries its evidence. The value found, the source and the date it was read are published on each firm's page; quotes are copied word for word.
- Not applicable (n/a). A criterion that does not apply is left out of its pillar.
- Missing data is never guessed. It is “not verified” and counts as 1, the lowest score, for as long as it cannot be found. An unpublished rule is never treated as neutral.
- A price shown by a selector is read as displayed. When a site shows the price of a $100,000 account only once that size is chosen, it is read on the official page after that choice, at list price without a promo code, with its quote and its address.
- We exclude no one, we inform. A regulator warning lowers the score and is shown on the firm's page with its source; it does not exclude the firm.
- One methodology per edition. It is frozen with the data; every correction is dated.
- What never enters into a score: the country, a partnership with Tradoshi, an affiliate link, a promo code, any payment.
The five entry conditions
A prop firm that does not meet all five is not ranked, and it is not named.
- A funding programme open to the public: a paid evaluation or instant funding, with a price shown. Proprietary trading companies hiring employees are not prop firms in the sense of the ranking.
- Its own active website. A website that refuses our crawler but serves a visitor is active: it is checked in a browser. No protection is ever bypassed.
- No report of the website as dangerous (phishing, malware).
- At least six months of activity at the data freeze date (first public archive of the website).
- No payout freeze in progress at the freeze date, documented by the firm, a regulator, a court or at least two media outlets.
Every firm that meets the conditions is rated and has its own page, with its international rank and its rank in each market it offers. The ranking publishes the top 50 internationally and in each market.
The five pillars and the score
| Pillar | Weight | Criteria | Weight per criterion |
|---|---|---|---|
| Trust | 30 % | 1, 2, 3, 4, 5, 6, 7 | 4.3 % |
| Payouts | 25 % | 8, 9, 10, 11, 12 | 5 % |
| Trader reviews | 20 % | 13, 14, 15, 16 | 6.7 % |
| Trading conditions | 15 % | 17, 18, 19, 20, 21, 22 | 2.5 % |
| Visibility | 10 % | 23, 24, 25, 26 | 2.5 % |
Overall score out of 100 = sum of the pillar scores multiplied by their weights, × 20. Ties are broken by the Trust pillar, then by Payouts.
Example: FTMO, ranked #1. (3.86 × 30 + 3.00 × 25 + 5.00 × 20 + 3.83 × 15 + 4.00 × 10) / 100 × 20 = 77.6 out of 100. The trust score shown on each page is the Trust pillar × 20.
Each firm also has a rank in each market it offers (futures, forex and CFDs, crypto, stocks), with the same score: the market rank sorts the same scores and recalculates nothing. A score is never relative to the other firms' average, so a firm's score does not change when another firm's does.
The 26 criteria and their scales
Each criterion is scored from 5 (the best level) to 1. Criterion 16 is reserved: 25 criteria are rated in 2027.
Trust · 30 %
1. Regulator warnings
| 5 | Absent from every list consulted |
|---|---|
| 2 | No licence in a given country, or a general list of dubious projects with no finding specific to the firm |
| 1 | Warning for fraud |
Source: AMF, IOSCO I-SCAN (FCA, CySEC, CONSOB, FINMA, SEC, CVM, FSCA…), ASIC, FINMA, CNMV, the CFTC RED List, FSMA. Matched on the exact domain, never on the name; every listing is checked at the source.
2. Registered company and age
| 5 | In the register for 6 years or more |
|---|---|
| 4 | 4 to 5 years |
| 3 | 2 to 3 years |
| 2 | 1 year |
| 1 | Under a year, or no company named on the website |
Source: The country's register when it offers an open search (United Kingdom: Companies House; France: SIRENE; Czech Republic: ARES; Estonia: business register; Brazil: CNPJ), otherwise GLEIF (LEI). n/a when the country's register cannot be searched freely: the country never enters into a score.
3. Identity and contact
| 5 | All 5 points |
|---|---|
| 4 | 4 points |
| 3 | 3 points |
| 2 | 2 points |
| 1 | 0 or 1 |
Source: The firm's website. Points: legal name of the entity selling the programme, postal address, registration number, direct contact (email or phone), a named executive.
4. Regulated execution partner
| 5 | Named, licensed broker or FCM, relationship confirmed by the partner or a register |
|---|---|
| 4 | Named, licensed broker or FCM, relationship stated by the firm |
| 3 | Named broker, unlicensed or licensed only offshore |
| 2 | Only the platform is named |
| 1 | Nothing |
Source: The firm's website, then regulators' registers (FCA, NFA BASIC, CySEC, ASIC…).
5. Honest marketing
| 5 | Visible risk warning and simulated accounts disclosed, no income promise |
|---|---|
| 4 | Neither promise nor warning |
| 3 | Earnings testimonials without a warning |
| 2 | Implied income promise |
| 1 | Quantified or guaranteed earnings, or simulated accounts presented as real |
Source: The firm's website, exact quote and date.
6. Secure website
| 5 | No alert, certificate valid for more than 14 days |
|---|---|
| 3 | Certificate expiring within 14 days |
| 1 | Invalid certificate |
Source: Technical check; public phishing and malware lists.
7. Protected domain (SPF, DMARC, headers)
| 5 | 5 points out of 5 |
|---|---|
| 4 | 4 |
| 3 | 3 |
| 2 | 2 |
| 1 | 0 or 1 |
Source: DNS and headers: SPF, DMARC present, DMARC set to quarantine or reject, HSTS, at least two other security headers.
Payouts · 25 %
8. Public payout proof
| 5 | Payouts published one by one in the last 3 months and verifiable by a third party (blockchain, named provider, audited report) |
|---|---|
| 4 | Published one by one in the last 3 months, not verifiable by a third party |
| 3 | Cumulative total only, or payouts older than 3 months |
| 2 | Trader testimonials only |
| 1 | No proof |
Source: The firm's payout page, blockchain explorers, payment provider.
9. Published processing time
| 5 | 24 h or less, in writing |
|---|---|
| 4 | 1 to 3 business days |
| 3 | 4 to 7 days |
| 2 | More than 7 days |
| 1 | Not published |
Source: Withdrawal rules, help centre, terms and conditions.
10. Payout frequency and first payout
| 5 | On demand or daily |
|---|---|
| 4 | Weekly |
| 3 | Every two weeks |
| 2 | Monthly |
| 1 | More than a month, or not published |
Source: Same pages. The longer of the cycle between payouts and the wait before the first one is used.
11. Withdrawal terms and methods
| 5 | All 5 points |
|---|---|
| 4 | 4 points |
| 3 | 3 points |
| 2 | 2 points |
| 1 | 0 or 1 |
Source: Points: no cap per payout; minimum of at most 1% of the account; no more than 5 trading days between payouts; no buffer above the starting balance; at least two payment methods, one of them a bank transfer or a named provider.
12. Payout history and incidents
| 5 | Payouts proven for 3 years or more, no incident |
|---|---|
| 4 | 1 to 3 years, no incident |
| 3 | Under a year, no incident |
| 2 | Documented incident, resolved within the last 3 years |
| 1 | Unresolved incident: payouts or sales suspended, collective refusals, regulator or court proceedings |
Source: The firm's announcements, regulator and court decisions, press. An incident is documented by the firm, a regulator, a court or at least two media outlets, never by a forum post alone.
Trader reviews · 20 %
13. Trader rating (Trustpilot and Google)
| 5 | 4.6 and above |
|---|---|
| 4 | 4.3 to 4.5 |
| 3 | 4.0 to 4.2 |
| 2 | 3.5 to 3.9 |
| 1 | Below 3.5, or rating withheld by Trustpilot |
Source: Trustpilot and Google average weighted by the number of reviews, then pulled towards the average of all firms when reviews are few (Bayesian average, 20 reviews). n/a below 5 reviews.
14. Number of reviews
| 5 | 5,000 and above |
|---|---|
| 4 | 1,000 to 4,999 |
| 3 | 200 to 999 |
| 2 | 50 to 199 |
| 1 | Fewer than 50 |
Source: Trustpilot and Google combined.
15. Suspicious review warning
| 5 | No Trustpilot warning |
|---|---|
| 3 | Trustpilot warning, rating still shown |
| 1 | Profile restricted for breaching Trustpilot's rules, or rating withheld |
Source: Trustpilot profile, warning text quoted on the firm's page.
16. 12-month trend
Source: Reserved: not rated in 2027, as review dates are not available.
Trading conditions · 15 %
17. Rule clarity
| 5 | 14 fields out of 14 |
|---|---|
| 4 | 12 or 13 |
| 3 | 9 to 11 |
| 2 | 6 to 8 |
| 1 | 5 or fewer |
Source: The 14 fields: profit target, maximum daily loss, maximum loss, drawdown calculation, maximum duration, minimum days, consistency rule, leverage, news trading, EAs and trade copying, weekend and overnight, profit split, fees and refunds, withdrawal terms.
18. Cost of a $100,000 account
| 5 | 1st quintile of the market |
|---|---|
| 4 | 2nd quintile |
| 3 | 3rd quintile |
| 2 | 4th quintile |
| 1 | 5th quintile, or not verified |
Source: Pricing page: list price without a promo code or temporary discount, activation fees included, one month for a subscription. Nearest size scaled to $100,000 pro rata, only from $50,000 to $200,000. Thresholds specific to each market, frozen when the data is frozen.
| Market | 5 | 4 | 3 | 2 | 1 |
|---|---|---|---|---|---|
| Forex and CFDs | ≤ $460 | ≤ $549 | ≤ $623 | ≤ $789 | > $789 |
| Futures | ≤ $209 | ≤ $265 | ≤ $316 | ≤ $378 | > $378 |
19. Profit target relative to maximum loss
| 5 | 1.0 or less, or instant funding |
|---|---|
| 4 | 1.01 to 1.5 |
| 3 | 1.51 to 2.0 |
| 2 | 2.01 to 2.5 |
| 1 | Above 2.5, or not published |
Source: Sum of the targets of all phases divided by the maximum loss allowed.
20. Drawdown type
| 5 | Static |
|---|---|
| 4 | End-of-day trailing, locked at the starting balance |
| 3 | End-of-day trailing, never locked |
| 2 | Intraday trailing, locked |
| 1 | Intraday trailing never locked, or not written |
Source: Drawdown rules, help centre.
21. Rules restricting how you trade
| 5 | None |
|---|---|
| 4 | 1 or 2 |
| 3 | 3 or 4 |
| 2 | 5 or 6 |
| 1 | 7 or more, or rules too poorly documented to count |
Source: Rules counted: consistency during the evaluation and on the funded account, news trading banned, weekend or overnight banned, EAs banned, maximum duration, more than 5 minimum days, mandatory stop loss, risk cap per trade.
22. What the trader keeps: split and fees
| 5 | All 5 points |
|---|---|
| 4 | 4 points |
| 3 | 3 points |
| 2 | 2 points |
| 1 | 0 or 1 |
Source: Points: split of at least 80% from the first payout; of at least 90%; evaluation fee refunded; no activation fee; no subscription or data fee after passing.
Visibility · 10 %
23. Track record
| 5 | 6 years and more |
|---|---|
| 4 | 4 to 5 years |
| 3 | 2 to 3 years |
| 2 | 1 year |
| 1 | Under a year |
Source: Domain register (RDAP), otherwise the first Wayback Machine archive. The more recent of the two dates, domain or company, when both exist.
24. Websites linking to it
| 5 | Among the 10,000 most-linked domains on the web |
|---|---|
| 4 | Top 100,000 |
| 3 | Top million |
| 2 | Top 10 million |
| 1 | Beyond, or absent |
Source: Common Crawl's public domain graph (harmonic centrality). Never a commercial score.
25. Press mentions, Wikipedia page
| 5 | Wikipedia page and 20 or more mentions over 12 months |
|---|---|
| 4 | 20 or more mentions |
| 3 | 5 to 19 |
| 2 | 1 to 4 |
| 1 | None |
Source: GDELT (world press), Wikipedia and Wikidata. An ambiguous name is checked before it enters into the score.
26. Audience
| 5 | 500,000 subscribers or more |
|---|---|
| 4 | 100,000 to 499,999 |
| 3 | 20,000 to 99,999 |
| 2 | 2,000 to 19,999 |
| 1 | Fewer than 2,000 |
Source: Public counters read through YouTube's official API, on the channel the firm itself links to.
What this ranking does not measure
- The payout delay actually observed, not just published: it is reserved for a future edition, based on payouts that traders record themselves, with their consent.
- Execution quality (slippage, spreads, the platform during the session) and customer support quality.
- The share of traders who pass the evaluation or get paid: no firm publishes it in a verifiable way.
- Anything behind an account login: only public pages are read.
Re-evaluation and independence
A prop firm can request the re-evaluation of a criterion, with a fact and evidence, at contact@tradoshi.com; every correction is dated.
The ranking cannot be bought: no payment, partnership or promo code enters into a score, and these pages carry no affiliate link. Tradoshi connects the accounts of many prop firms to its trading journal and offers a plan to prop firms: none of this enters into the ranking.