Prop Firms Compared: Scores, Rules, Costs and Payout Evidence (August 2026)

Two readings of the cheapest-challenge column are both honest, and they disagree. The lowest sticker price and the lowest cash out of pocket are different rows, because a pay-later route charges a small fee up front and an activation fee on passing that can exceed the price of buying an instant-funding account outright. A third reading, capital bought per dollar of fee, gives a third answer again.

Every firm on this page is scored against the same seven weighted pillars and the same evidence rules. The payout pillar outweighs the rest, and any number a firm publishes about itself is logged as a claim rather than a verified result. Scores move when the evidence moves, so this order is not fixed.

Prop trading firms sell access to a funded trading account and take a share of the profit made on it. The trader pays for an evaluation, trades to a profit target inside a set of loss limits, and on passing receives an account funded with the firm's capital. Profits are split, most commonly 80% to the trader. Losses stop at the loss limit, and the trader's downside is the fees paid, not the account balance.

Almost every retail funded account is a simulated environment. Orders execute against the firm's pricing feed, the balance is a number and the payout is money, and whether any given order reaches an outside market is a decision the firm makes and rarely shows. That is not by itself a mark against a firm. It is the reason the terms matter more than the platform, and the reason a firm that states plainly how it books trades scores above one that leaves it unstated.

Every firm in this category publishes a payout total. Not one of them publishes an audited one. CryptoSlate classes payout evidence three ways, and the class is what the score reacts to.

• CLAIMED — the firm states it on a surface it controls. A homepage counter, a rewards page, an internal rating on an internal scale, or a review-platform profile the firm pays for and solicits reviews into, because there the firm selects the sample.
• REPORTED — traders state it, on a platform the firm does not manage. Trustpilot, Reddit, Discord, or a first-hand account given to CryptoSlate. Volume raises confidence and never changes the class.
• VERIFIED — CryptoSlate can point to a record the firm does not control. A regulator's register, a completed corporate transaction, a review platform's own enforcement notice, or a transaction CryptoSlate executed and holds the receipt for.

No firm here has VERIFIED evidence that it pays traders. The VERIFIED items that do exist establish something adjacent, including ownership, longevity or a purchase. Others are enforcement actions against the firm's own reputation data. First-hand payouts described by a reviewer sit in REPORTED, because a recollection is not a receipt, and that line is the reason the first-party purchase below clears the bar while a first-party payout does not.

Every claim in the second column is the firm's own, read from its public surfaces in August 2026.

What the Fastest Payout Prop Firm Claims Are Worth

A firm quoting an average processing time is describing its own log, which nobody outside it can audit. A firm quoting a deadline with a stated penalty for missing it has written a term a trader can hold it to. A firm paying on demand with no cycle removes the question, because there is no queue to be slow in.

The number that matters more is the one before the stopwatch starts. A 24-hour guarantee attached to a 14-day first-payout window is a 15-day wait. A guarantee counted in business hours on a single city's clock pauses at weekends. A payout cycle that restarts every time an account scales resets the wait each time the trader succeeds. Read the cycle, the eligibility window and the clock the guarantee runs on before reading the processing time.

Regulatory Posture, and What a Warning Listing Does and Does Not Mean

Almost nothing in this category is financially regulated, and the firms say so in their own documents. The pattern repeats across the table: a marketing entity in a free zone, a service entity in a small offshore jurisdiction, a payments entity somewhere else, and a clause stating that the marketing entity carries out no regulated activity. Some are registered for education and training services instead of financial services, and describe a payout as a reward with eligibility conditions instead of an entitlement.

Crypto Fund Trader sits on FINMA's public warning list, added in August 2024 because its trading name is absent from the Swiss commercial register. The company behind that name was entered on the register five months before the listing, so what is unregistered is the name, not the company. FINMA's own position is that a listing does not make an activity unlawful, only that authorization is absent. That cuts both ways. It is weaker than a fraud finding and it has not been withdrawn. The full record is in the Crypto Fund Trader scoring breakdown.

What a Payout Is, Contractually, Before It Is Income

The documents that govern these payments do not describe a share of trading profit. They describe a reward, paid on eligibility conditions, by an entity that states it carries out no regulated activity, on capital the trader never owned and positions the trader never held. The evaluation fee is written the same way, as a one-time service charge and not a deposit, which is why it is non-refundable by default. How any of that is treated for tax turns on the country, on the entity making the payment and on how the trader receives it, so it is a question for a tax professional before the first payout and not after the tax year closes.

Account Forfeiture, and the Terms That Take Profit Back

The failure mode traders underestimate is not a breached drawdown. It is a terms clause that closes a profitable account. Automated account-linking detection can close an account and retain the balance when it associates it with another account, and one trader reports exactly that at The5ers, REPORTED-class, with the burden of disproving the match falling on the trader. Coordinated-trading clauses do the same when several accounts hold correlated positions, and refused payouts on that ground are the dominant complaint theme at Goat Funded Trader. The5ers also forfeits all profit and closes the account for publishing private correspondence, which covers support tickets and chat logs, so a trader who takes a dispute public loses the balance being disputed.

Breakout's disclosures run the other way. They state that a funded trade may be recorded as an internal book entry with a hypothetical result calculated against it, or routed to a market maker or exchange, at the firm's discretion, with no way for the trader to see which happened. The same firm discloses that it earns a fee every time an evaluation trader fails and rebuys. The most disclosure and the most uncomfortable disclosure come from the same place, which is what disclosure looks like when it is genuine. The record is in the Breakout scoring breakdown.

What the Rest of This Comparison Set Does Not Print

Comparison pages in this category are paid by the firms they rank, and the scoring scales show it. The bottom of a competing table still reads as an endorsement, which means the ordering carries no information about the firms the page declines to criticize. A tracker that publishes a payouts database has the same problem in a different shape, because a database of numbers the firms supplied is a restatement, not a confirmation.

CryptoSlate prints the low scores in the same table as the high ones, with the same columns filled in, and prints the payout evidence underneath. That is the only reason the ordering above means anything.

Evaluation fees are the primary revenue line. Pricing is built around the repeat attempt and not the single pass, because a failed evaluation converts into another purchase. Activation fees, reset fees and checkout add-ons sit on top, and several are priced as a percentage of the evaluation fee, so they scale with account size.

Trading revenue is the second line. Commissions and spreads are charged on the funded account, and where a firm books a trade internally instead of passing it to a market, the losing side of that trade is income. The profit split is the third line and the smallest, because it only pays when the trader does. A firm's own disclosures state which of these it leans on, and reading them is the fastest way to work out whose success it is priced for.

Every route to a funded account is one of four shapes, and the shape sets the difficulty far more than the profit target does.

Two-Step Evaluations

A challenge phase with a profit target near 8% to 10%, then a verification phase at roughly half that target, both inside the same loss limits. Two phases means two chances to breach, so the pass rate is lower than a single target implies. The compensation is that two-step programs carry the loosest rules on most price lists: static drawdown, no consistency rule, and the longest time allowance, often none at all.

One-Step Evaluations

A single phase to a higher target, usually 10%. It looks simpler and is frequently the harder product on the same price list. The tightening is not in the target. It is the daily loss limit dropping, the maximum loss switching from static to trailing, and a best-day or consistency rule appearing that the two-step does not carry. Compare the rule set, not the number of phases.

Instant Funding Prop Firms, Also Sold as No-Evaluation or No-Challenge Accounts

No challenge at all. Pay, receive a funded account, start trading. Instant funding prop firms are the most heavily governed product in this category, because the constraint moves from the evaluation into the account: the tightest drawdown floors, the lowest opening profit splits, live floating-loss limits that close the account on unrealized losses of 1% to 2%, weekend-holding bans, news-trading bans and the highest consistency percentages. Buying capital without an evaluation means paying for it in rules, and the sizes are usually capped well below the phased routes.

Three-Step and Staged Programs

Three phases with lower targets each, marketed as gentler. The extra phase adds another breach opportunity and lengthens the time to a first payout. Watch the plan name here, because on some programs the name describes the funded balance the account can reach after every step, not the balance it starts at. A program named for $20,000 that opens at $5,000 and steps through two stages is a $5,000 account until it is not.

Which Evaluation Type Suits Beginners

The route that suits a first funded account is the one with the fewest ways to lose it. That means a static floor, so a winning week cannot move the failure level closer. It means no consistency rule, so one good day does not block the payout. And it means no minimum trading days and no time limit, so nothing forces a trade that was not there. Two-step programs carry that combination most often, and the extra phase is what buys it. The route to avoid first is the cheapest instant account, because the tightest floors and the live floating-loss limits sit there, and a first-time funded trader is the one most likely to breach one by accident.

The drawdown floor is the single rule that decides whether a funded account survives a winning week. There are three behaviors and they produce completely different outcomes from identical trading.

How Prop Firm Drawdown Works: Static, Trailing and End-of-Day

Static. On a $50,000 account with a 5% static floor, the failure level is $47,500 for the life of the account. Trade it to $60,000 and the floor is still $47,500, so profit has bought $12,500 of room. Static is the only drawdown type where making money makes the account safer.

Trailing on live equity. On a $100,000 account with a 10% trailing floor, the floor opens at $90,000 and follows every new equity high, including unrealized ones. Run a position to $106,000 in equity and close it at $101,000. The floor has followed the peak to $95,400. The balance now reads $101,000 and the distance to failure has fallen from $10,000 to $5,600. Profit moved the floor closer. A variant of this trails only on closed-trade balance highs, so floating profit, breakeven closes and losing trades never move it, and on the same sequence that floor would sit at $90,900 instead of $95,400. Those two floors carry the same one-word label on most comparison tables.

End-of-day trailing. The floor recalculates once a day off the closing or midnight balance. On a $50,000 account with a $2,500 allowance, a day that runs to $53,000 intraday and closes at $52,000 sets tomorrow's floor at $49,500. The intraday high is ignored, which makes this gentler than live trailing and harsher than static.

Two extra behaviors sit on top of trailing floors and cost accounts regularly. Some floors rise permanently and reset only when a payout is withdrawn, and they reset to 90% of the initial capital instead of to the current balance, which makes withdrawing the only lever that ever moves the floor down. And some stop trailing at a set profit level and lock to the opening balance, so the account gets safer the moment it clears that threshold and is at its most fragile in the run-up to it.

Consistency Rules

A consistency rule caps how much of total profit any single day, or single position, may contribute. Under a 35% rule, a trader sitting on $8,000 of profit with a $3,000 best day is at 37.5% and outside the limit. Where the rule gates the withdrawal, the account survives and the payout is blocked until total profit reaches $8,572, at which point the same $3,000 day falls inside 35%. The rule does not punish losing. It punishes one good day, and it converts a fast profit into an obligation to keep trading.

The percentage is only half the rule. Check whether it applies per day or per position, whether it is measured in the evaluation or on the funded account or both, and whether breaching it gates the withdrawal or closes the account. Those three variables differ by model inside the same firm, which is why the consistency column in the table above is marked “Varies by model” on most rows.

Daily Loss Limits

A 3% daily loss limit on a $100,000 account is $3,000, measured from the balance at a fixed reset time. The failure most traders do not see coming is the definition mismatch: the reset-time balance usually excludes open positions, while the live equity check includes them. Carry a position $3,500 underwater through the reset and the account breaches the instant the clock rolls, on a trade that was inside the limit a minute earlier.

Minimum trading days carry a similar trap. On some programs a day only counts toward the minimum if it closes in profit, or in profit above a set percentage, so a week of flat sessions advances nothing.

What Forex Prop Firms Restrict That Futures Programs Do Not

A forex or CFD program prices its own account structure as an upgrade. A swap-free account is sold as a surcharge, commonly a percentage of the evaluation fee, and the same upgrade often cuts the leverage available on the account, so avoiding overnight interest is paid for twice. Weekend holding and trading around scheduled news releases are barred on the tightest models, which rules out a strategy built around a release before the drawdown ever comes into it. Futures programs replace all of that with a hard daily close a few minutes before the session settles. Every open position closes with it, so nothing is carried overnight and nothing is charged for carrying it.

Why a Firm's Published Rules May Not Describe the Model You Bought

A firm's public rules page usually describes one drawdown behavior with one worked example, and that example is often true of only the models it was written for. Cheap evaluations are frequently governed by the tighter rule and the rules page is frequently written around the flagship one. A trader can read a fixed 8% floor with a clean $100,000 to $92,000 example, buy the low-priced model on the same site, and be trading a 10% trailing floor instead.

Check the drawdown against the specific model name on the checkout, not against the rules page and not against the marketing. Every drawdown cell in the comparison table above marked “Varies by model” is flagged for exactly this reason, and each firm's scoring breakdown lists the models one by one.

The evaluation fee is the number every comparison table prints and it is rarely the total. A $10,000 evaluation sits low on most price lists, and the charges outside it can exceed the fee itself. At larger sizes the fee becomes one of several charges and the add-ons scale with it, because they are commonly priced as a percentage of the evaluation.

What the Arithmetic Looks Like

The pattern is easier to see on round numbers than on any live price list. Take an evaluation advertised at $1,000. Attach the two add-ons that appear at checkout on many programs, one shortening the payout cycle and one raising the split, each priced at 20% of the evaluation fee. The purchase is $1,400, and neither add-on appeared on the selector where the $1,000 was shown. Now apply a 2% processing charge on withdrawals. The 90% split bought with one of those add-ons delivers 88.2%, permanently. No single firm above charges every one of these, and the columns show which firm charges what.

A pay-later route charges a token amount up front and an activation fee on passing. That is deferred and conditional, not lower. It reduces what you risk before you know whether you can pass, and not what you pay once you have.

Ladder Pricing Is Not Linear

Divide the fee by the capital it buys before choosing a size. Price lists in this category step up in uneven jumps, and adjacent sizes can sit close enough in price that the larger account costs less per dollar of capital than the smaller one. Firms price the ladder to move traders up it. Run the division across the account-sizes and cheapest-challenge columns above, then run it again at the size you actually want, because the entry rung is rarely the best value per dollar on a list.

What You Pay When You Fail

This is where the pricing model does its work. A firm that sells resets at a stage-based discount makes a second attempt cost less than a fresh purchase. A firm that sells no resets at all makes a breach a full repurchase at list price, add-ons included, which on a large account is a four-figure event. Neither approach is wrong, and the difference between them decides more of the final cost than the sticker price does.

Refunds That Are Not Refunds

A refundable fee is a genuine feature, but its conditions determine when and how the refund is delivered. Three patterns recur. The refund arrives on the fourth successful payout instead of on passing, which makes it contingent on trading well for months. The refund applies to some models on a price list while the refundable label sits on all of them. Or the refund is paid in credits instead of money, some usable only against further purchases and some added to the funded account as trading equity, and neither can be withdrawn or transferred.

There is no single best prop firm, because the rule that decides the outcome changes with what you trade and how long you hold it.

• Beginners. Breakout's static-floor plans carry no consistency rule, no minimum trading days and no time limit, and there is no reset to buy, so a breach means starting over at list price.
• Crypto. Crypto Fund Trader's instrument range covers 715 crypto pairs with a route that reaches a live exchange order book, and its score in the table above is the one to read before buying. Whether a firm gives genuine exposure or lists crypto as one CFD asset class at cut leverage is set out across the firms quoting crypto markets.
• Forex. FundingPips program rules keep static floors on the phased models and refund the evaluation fee after the fourth reward on selected phased models, and the firm does not accept US or Canadian traders. Eligibility varies more than any other rule here, and it is compared firm by firm for currency traders based in the US.
• US traders. Breakout accepts US residents directly, and FTMO's US arrangement with OANDA runs through a CFTC-regulated partner on separate pricing and a separate account structure.
• Cheapest. The5ers on sticker price, with what Goat Funded Trader charges on passing as the counter-example, because its pay-later route is cheaper to start and more expensive to finish.
• Scaling over years. The5ers scaling programs step a funded account toward $4,000,000, on a payout cycle that restarts at every step up.

Scores run 0 to 10 across weighted pillars: payout reliability and evidence, rules fairness and transparency, cost and value, trading conditions, profit split and scaling, firm transparency and stability, and support and reputation. Payout evidence carries the heaviest weight, which is why the classing above moves scores. Every firm name in the comparison table links to the full breakdown, with each pillar scored one by one. The weightings, the pillar definitions and the evidence classes are set out in CryptoSlate's ranking methodology.

CryptoSlate earns affiliate commission on some links in the comparison table, on the terms set out in how CryptoSlate makes and spends money. A commercial arrangement does not move a score. The low scores are printed in the same table as the high ones, with the same columns filled in, because a comparison that only shows firms worth recommending is not a comparison.