| Research thesis Modern prop trading did not simply “give retail traders more capital.” In many of the largest programs, the trader never controls real market capital at all. The trader risks an evaluation fee and time; the prop firm assumes a contingent payout obligation; and a broker or liquidity provider may bear little or no trade exposure unless the firm chooses to replicate selected trades. The industry is therefore less like one traditional proprietary desk and more like a spectrum of evaluation, data-selection, payout and capital-allocation businesses. |
Key findings
| Finding | Why it matters |
| A “$100,000 funded account” may contain no $100,000 of trader-accessible capital. | FTMO, FundingPips and FundedNext explicitly describe core CFD funded stages as simulated environments with real rewards tied to simulated results. |
| Challenge fees are economically central. | FTMO’s founders said 8%–10% pass the Challenge and that the vast majority of FTMO income comes from evaluation/course fees. |
| Simulated trading removes market P&L from the demo account, not payout risk from the firm. | FTMO’s 2025 revenue rose to CZK 8.8B while normalized EBITDA fell to CZK 1.19B; Forbes reported higher-than-expected client payouts as a major cause. |
| Broker exposure is model-dependent. | Pure simulation may use brokers or liquidity providers only for quotes. Selective replication adds live-market risk. Broker-native allocation programs such as Axi Select place real capital behind trader activity. |
| Regulation remains uneven. | European regulators have warned that many evaluation-style prop firms sit outside ordinary investment-firm authorization when they do not provide regulated execution, while the U.S. legal boundary remains unsettled. |
Figure 1. “Funded” accounts can shift risk very differently depending on whether trading remains simulated, is selectively replicated, or is placed into a live broker-funded allocation.
The word ‘funded’ hides the most important fact
Retail FX used to have a relatively intuitive risk chain. A trader deposited cash with a broker, placed leveraged trades, and either the broker internalized the exposure or hedged it with external liquidity. The trader’s account balance was real, the losses were real, and the broker’s dealing model determined where the opposite side of the exposure ultimately sat.
Modern prop trading changed that architecture. In the most common challenge model, the trader first buys an evaluation. If the performance and drawdown rules are met, the trader receives what marketing often calls a funded account. But leading firms now state explicitly that many of those accounts are still simulated. FTMO says its CFD traders continue on demo accounts with fictitious capital, while FundingPips says all of its current account models run on simulated capital. FundedNext’s May 2026 terms likewise define the funded stage as simulated trading rather than brokerage or live order execution.
That distinction changes the meaning of risk. The trader is not losing a $100,000 balance when a $100,000 simulated account breaches its drawdown rule. The trader is losing the fee paid for access, the account status, and the time spent building the performance record. The firm, meanwhile, has promised a real monetary reward if the simulated performance satisfies its contractual rules. The apparent trading account is therefore partly a measurement device and partly a contingent compensation contract.
Challenge fees are not just a screening cost
The economics of the challenge funnel are unusually important because most participants never reach the payout stage. In a 2026 Forbes interview republished by FTMO, co-founder Otakar Šuffner said only 8%–10% of people pass the FTMO Challenge. He also said the vast majority of FTMO’s income comes from fees for its educational courses and evaluations. FTMO says its two-step fee is refunded only after a successful trader reaches the first reward withdrawal; its one-step fee is not refunded after passing.
That does not mean every failed challenge is pure margin. Prop firms pay for platforms, market data, technology, customer support, affiliate acquisition, payment processing, compliance, fraud controls and—most importantly—rewards to successful traders. But the funnel creates a very different unit-economics profile from a broker that earns primarily when a customer trades. The prop firm monetizes access before it knows whether the customer will ever generate a payout liability.
The incentives also explain why maximum loss, daily loss, consistency and payout-frequency rules matter to the firm as well as to the trader. These rules are presented as discipline filters, and they do serve that function. Economically, however, they also cap the size, timing and concentration of the firm’s contingent reward obligations. A rule that prevents one outsized trading day from dominating a payout, for example, is simultaneously a behavior filter and a liability-management tool.
Figure 2. FTMO’s founders disclosed an 8%–10% Challenge pass rate. This is a firm-specific figure, not an industry-wide pass rate, and passing a challenge is not the same as receiving a payout.
| Illustrative fee-funnel math — not a profit estimate Using FTMO’s standard €540 price shown for a €100,000 two-step challenge, 100 purchases would produce €54,000 of gross evaluation fees. At an 8%–10% pass rate, only 8–10 traders would pass the Challenge. Even if every one of those traders later became eligible for the first reward and therefore received a full fee refund, the fee refunds would total only €4,320–€5,400. This says nothing about reward payouts, marketing, staff, technology or taxes; it simply shows why the evaluation funnel can generate substantial gross cash inflow before payout economics are considered. |
Who pays the winners when the trades are simulated?
A common misconception is that a prop firm must earn a trader’s payout from the same trades that generated the simulated profit. That is not how the largest evaluation-style models necessarily work. FTMO says a trader receives the contractual reward whether or not FTMO copied that trader’s positions into its own live trading. The simulated account and the firm’s proprietary live book are separate systems.
In economic terms, the reward can be funded from the company’s general resources: challenge-fee revenue, retained earnings, and any gains generated by the firm’s own trading or other businesses. FundedNext’s terms are unusually explicit on this point: a performance reward is described as a conditional contractual benefit based on simulated performance, not as profit generated in the real market on the customer’s behalf.
This creates a liability that behaves differently from ordinary broker exposure. If a large cohort of simulated traders becomes profitable at the same time, the firm may face a surge in cash payouts even if it did not replicate those trades. If it did replicate them, live gains can offset that liability—but only to the extent the selection, sizing and execution of copied positions actually match the winning simulated flow.
FTMO’s 2025 results show that payout risk is real
The clearest evidence comes from the industry’s largest disclosed operator. Forbes reported in September 2026 that FTMO’s 2025 revenue rose about one-third to CZK 8.8 billion, yet normalized EBITDA fell to CZK 1.19 billion from CZK 2.4 billion in 2024. The publication said one of the main reasons was that amounts paid to clients were higher than management had expected, alongside difficult market conditions.
On those figures, FTMO’s normalized EBITDA margin compressed from roughly 35.3% in 2024 to 13.5% in 2025 even while revenue grew about 29%. This is exactly the risk that can be invisible when the industry is described only through simulated accounts: the firm may not be taking the trader’s demo loss, but it is underwriting the payout distribution. When trader rewards run above assumptions, the economics can deteriorate sharply.
Figure 3. FTMO’s latest disclosed annual figures show that payout pressure can hit the prop firm’s own economics even when revenue grows. Sources: Forbes Czech / FTMO-republished Forbes interview.
Broker exposure can range from almost zero to genuine market risk
The phrase “prop firm broker” can obscure another distinction. In a pure simulated model, a broker or liquidity provider may provide price data without executing the trader’s order. FTMO states that its CFD demo infrastructure uses real quotes from liquidity providers, but those liquidity providers do not provide trading liquidity to FTMO’s clients because client trades are not sent to live markets. The broker-side market risk is therefore minimal at the account level.
The picture changes when the prop firm uses successful trader data to place its own trades. FTMO says it continuously analyzes simulated trading and may execute selected strategies on its own account. At that point, the firm—not the trader—takes live market risk, while its broker and liquidity providers execute the real positions. The trader’s payout does not depend on whether this happens, so the firm must manage the correlation between payout liabilities and whatever live exposure it chooses to take.
A third model is closer to traditional capital allocation. Axi Select requires a trader to use a live Axi account, maintain a minimum balance, and build an Edge Score. A separate Axi-funded Allocation Account then mirrors the trader’s activity with a multiplier. Axi states that the allocation is its capital, cannot be withdrawn by the trader, and can scale to as much as $1 million. Here the provider’s capital is genuinely exposed to live market P&L, and the trader may also have personal capital at risk in the linked live account.
FTMO’s September 2026 launch of FTMO Futures adds yet another hybrid: evaluation and Sim-Funded stages remain simulated, but a small number of consistent traders may be invited into a Live Funded Account with real capital. The same brand can therefore operate multiple risk architectures at once.
Four models that retail traders often call “funded”
| Model | Trader cash at risk | Is trader order live? | Primary provider risk | Example |
| Simulation / reward | Evaluation fee + time | No | Reward liability, fraud/abuse, operations | FTMO CFDs; FundingPips; FundedNext CFDs |
| Simulation + selective copy | Evaluation fee + time | Trader order no; firm may copy separately | Reward liability + selected live-market P&L | FTMO proprietary trading using selected client data |
| Sim-funded to live invitation | Evaluation fee + time | Not initially; live after invitation | Reward liability, then real funded-account P&L | FTMO Futures |
| Broker-funded allocation | Trader’s own live balance where required | Yes | Real allocated-capital P&L + brokerage risk | Axi Select |
Why risk rules are also balance-sheet controls
Modern prop firms often frame risk limits as a test of discipline: a 5% daily loss cap, 10% maximum loss, minimum trading days, consistency rules, prohibited arbitrage, news restrictions or maximum risk per trade. From the firm’s perspective, these constraints do something more mechanical. They shape the distribution of payouts.
A static maximum-loss rule prevents a trader from creating an unlimited negative path in the simulated environment, but the more important economic effect is selection. Strategies that need wide drawdowns, long holding periods or concentrated event exposure are less likely to survive. That filters the pool toward short-duration, tightly risk-managed strategies whose payout behavior is easier to model. Consistency rules similarly reduce the chance that one extreme event creates a very large reward claim.
Firms also police strategies that exploit differences between simulation and real execution. FundedNext’s 2026 terms prohibit latency arbitrage, off-market pricing exploitation and strategies designed to profit from simulation mechanics. FundingPips applies risk-per-trade and consistency rules on some account models. These restrictions are not merely house rules; they protect the gap between a simulated P&L and the P&L that could realistically be replicated in a live market.
The broker-prop relationship is becoming more vertically integrated
The industry’s next phase may reduce some of the separation between simulation and brokerage. FTMO completed its acquisition of OANDA in December 2025 after receiving approvals from five regulators. The companies say OANDA remains a standalone regulated broker while FTMO focuses on modern prop trading, but the group now owns both a large evaluation business and a globally regulated brokerage infrastructure.
FundedNext has taken a different route. Its prop business remains a simulated program, while clients can choose to send approved rewards directly to FNmarkets, a separately regulated brokerage in the same broader FN ecosystem. The firms state that the relationship is optional and that live trading at FNmarkets is a separate contract. The commercial logic is still obvious: a successful simulated trader can be converted into a live brokerage customer without leaving the ecosystem.
Axi Select starts at the broker itself and removes the challenge fee entirely. That avoids the most controversial incentive—earning upfront revenue from repeated challenge attempts—but it introduces a different trade-off: the trader must fund and use a live account, so personal market risk exists from day one. None of these models is inherently superior for every trader; they simply allocate risk differently.
Regulation is still trying to define where the product begins and ends
The regulatory difficulty is that a simulated evaluation may look like a financial product without technically being one. Belgium’s FSMA warned in 2024 that many prop-trading firms do not hold investment-services authorization and described the model as a form of shadow investing in which consumers pay for challenges and may never place a real trade. Italy’s Consob issued a similar funded-trading warning in July 2024, and a June 2026 Consob fintech paper said the regulatory issue lies in models that formally avoid acting as brokers and therefore may fall outside ordinary MiFID II obligations.
In the United States, the My Forex Funds case illustrates why the boundary remains unresolved. The CFTC alleged that Traders Global acted as a counterparty to retail forex customers even though most activity was simulated. A federal court dismissed the case with prejudice in May 2025 after finding serious misconduct by the CFTC in how the litigation was prosecuted, and the agency was sanctioned. Crucially, that procedural outcome did not produce a merits ruling that broadly legalized or prohibited the modern prop model.
The current U.S. FTMO/OANDA structure also highlights the distinction. FTMO’s U.S. materials state that its evaluation and rewards services are simulated and are not subject to financial-regulator oversight, even though some infrastructure is provided within the broader OANDA relationship. For a retail trader, the presence of a regulated broker in the corporate ecosystem should therefore not be confused with regulatory protection for the prop challenge itself.
| The central conflict is economic, not necessarily fraudulent In a pure simulation model, a losing trader normally creates no reward liability, while a winning trader does. That means the provider has an economic interest in controlling payout frequency and size. The existence of that conflict does not prove misconduct: reputable firms can manage it through transparent rules, independent review, strong reserves and selective hedging. But traders should understand that the provider is their contractual counterparty for the reward, not merely a neutral venue matching their orders to a market. |
What traders should ask before buying a challenge
| Question | Why it matters |
| Is the funded stage simulated or live? | Do not infer this from words such as “capital,” “funded,” or “account size.” Read the legal terms. |
| Where does the reward legally come from? | Is it a contractual reward from the firm, profit share from a real account, or compensation from a signal-provider arrangement? |
| When is the evaluation fee refundable? | Passing alone may not trigger a refund; some models require the first approved payout, while others never refund one-step fees. |
| Who decides whether a payout is valid? | Look for discretionary review clauses, prohibited-strategy definitions, consistency requirements and post-payout audits. |
| Does the firm copy trades into a live book? | If yes, ask whether copying is automatic or selective. A payout that is independent of copying creates basis risk for the firm. |
| Which legal entity owes the payout? | A regulated broker elsewhere in the group may not be the entity that owes the prop reward. |
| What happens if the firm shuts down? | Evaluation fees and unpaid rewards generally do not receive the same client-money protections as deposits at a regulated broker unless the contract says otherwise. |
| How much counterparty exposure are you comfortable with? | A profitable simulated account can still be economically worthless if the firm cannot or will not honor the reward contract. |
The real answer: risk did not disappear — it moved
The modern prop-trading boom changed retail FX because it separated apparent buying power from actual balance-sheet capital. A trader can control a $100,000 simulated account while risking only a few hundred dollars in fees. That is attractive precisely because the traditional market-loss risk has been removed from the trader’s side of the screen.
But somebody still carries economic risk. In a pure simulation model, the trader bears the fee and opportunity cost while the prop firm bears the reward obligation and business-model risk. If the firm selectively replicates trades, it adds genuine market risk to its own balance sheet. In a broker-funded allocation program, real market P&L moves even closer to the traditional proprietary-trading model. The broker’s exposure ranges from almost nothing to substantial depending on which architecture is used.
For investors and traders, this is the most useful way to analyze the industry. The question is not whether a prop firm calls an account funded. The question is what asset actually exists, which entity owes the money, what event creates the payout, and whose balance sheet absorbs the loss when the trader is right.
Methodology & Sources
This article distinguishes firm-reported facts, regulator statements and analytical calculations. “Funded” is used as an industry term only; where the underlying account is simulated, the article says so explicitly. The FTMO fee-funnel example is illustrative and is not a profit estimate. Financial calculations use published figures and simple arithmetic; they exclude taxes, acquisition accounting, non-recurring items and undisclosed internal hedging. Research current through October 5, 2026.
1. Forbes: The FTMO Cover Story — FTMO / Forbes Czech. Pass rate, fee-income model, 2024 financials, cumulative payouts and selective use of simulated-trader data. Source link
2. FTMO 2025 revenues rise to CZK 8.8B — Forbes Czech. Latest disclosed 2025 revenue, normalized EBITDA and payout-pressure commentary. Source link
3. How FTMO works — FTMO. Simulated CFD account structure and separation between client simulation and FTMO proprietary live trading. Source link
4. FTMO Account technical infrastructure — FTMO. Liquidity providers supply quotes to simulated accounts; client orders are not live. Source link
5. FTMO Trading Objectives — FTMO. Profit targets, maximum daily loss, maximum loss and best-day rules. Source link
6. Why is there a fee? — FTMO. Fee purpose and refund mechanics. Source link
7. FTMO Futures: How it works — FTMO. Sim-Funded stage and invitation-only Live Funded Account. Source link
8. FTMO completes OANDA acquisition — OANDA. December 2025 acquisition and regulated-broker separation. Source link
9. Axi Select funded trader program — Axi. Live-account entry, Axi capital allocation and profit-share structure. Source link
10. How Axi Select withdrawals work — Axi Help Center. Separation between trader-owned live account and Axi-funded Allocation Account. Source link
11. FundedNext Terms of Service — FundedNext. Definition of simulated funded accounts and contractual performance rewards. Source link
12. FundedNext Risk Disclosure — FundedNext. Historical stage-advance disclosure and simulation risk warning. Source link
13. FundingPips account models — FundingPips. All current FundingPips accounts described as simulated capital with real rewards. Source link
14. FundedNext / FNmarkets relationship — FundedNext. Separation between simulated prop service and optional live brokerage relationship. Source link
15. FSMA warning on prop trading firms — Belgian FSMA. Regulatory warning on challenge fees, simulated accounts and authorization. Source link
16. Consob warning on funded trading — Consob / CNMV-hosted copy. July 2024 funded-trading warning. Source link
17. Judge sanctions CFTC in My Forex Funds case — Reuters. May 2025 dismissal and sanctions; case ended on litigation conduct rather than a broad merits resolution. Source link
18. FTMO US / OANDA fee and oversight disclosure — FTMO x OANDA. U.S. simulated-service structure and statement that prop services are not subject to financial-regulator oversight. Source link
| Editorial note This article is educational analysis, not investment, legal or tax advice. Program terms change frequently and may differ by jurisdiction. Traders should verify the current terms and legal entity before purchasing any evaluation or opening any live brokerage account. |
Johan Shamshad is a financial markets writer at Dave Finances covering cryptocurrencies, trading platforms, brokers, fintech, financial regulation, and developments across global markets. He previously worked at Gulf News, adding newsroom experience to his coverage of fast-moving financial and digital-asset markets.
His work focuses on identifying market-moving events, company developments, regulatory changes, product launches, and shifts in trading and financial infrastructure.
Johan contributes news and analysis designed to help readers understand not only what happened, but why a development matters and how it may affect the wider financial landscape.