Most recorded breaches are correct. The rule was hit, the risk engine flagged it, and a trader looking at the same data would agree. Disputes still arise in every contractual relationship, including ones where both sides act in good faith, and the outcome turns almost entirely on records you kept before you knew you needed them. What follows is that preparation, and the escalation path if you ever use it. It is not legal advice, it assumes nothing about any firm's conduct, and every legal point below is a question to put to a qualified lawyer in your own jurisdiction, not a conclusion to act on.
In a prop firm dispute the trader is arguing against the counterparty's own system of record, because the firm supplies the platform, the risk engine, the dashboard, the trade record and the rule text. The only counterweight is an independent record created before the dispute existed: scheduled trade history exports, a dated copy of the rule set, UTC timestamps of your own, and an end of day screenshot of the dashboard state. Escalate in writing by asking for the specific calculation the firm applied rather than asserting that the decision was wrong.
Why the evidence problem comes first
You are arguing against your counterparty's own system of record. The firm supplies the trading platform, the risk engine that computes drawdown, the dashboard that displays your state, the trade record itself, and the rule text that defines a breach. It can also amend that rule text. None of this implies misconduct. It is the shape of the contract, and it means one party holds every artifact and the other holds a memory.
No forum repairs that asymmetry for you. The CFTC's guidance on choosing a reparations proceeding states flatly that "the Administrative Judge will not investigate your case for you" and that you must collect and submit the evidence supporting your claim. Arbitration panels, chargeback analysts and small claims judges work the same way. Every route below rewards the trader who arrives with a file and frustrates the trader who arrives with a narrative.
Preparation also protects you from yourself. Many disputes end within an hour when the trader pulls their own record and finds the firm was right: an overnight position that crossed a reset, a size pattern that tripped a consistency threshold, a payout schedule they misread. Our breakdown of why prop firm payouts get denied covers the causes that most often turn out to be real.
What should you document, and when?
Five artifacts, captured on a schedule while the account is still active. The schedule matters more than the thoroughness.
| Artifact | Capture cadence | What it establishes |
|---|---|---|
| Full rule set, terms and payout policy, saved as PDF or full page screenshots | On the day you pay, and again on every change notice | Which version of the rules you agreed to, and exactly what changed and when |
| Platform trade history export | Weekly, plus after any unusually large day | Fills, order times, prices and commissions in a file you hold, not a page you can be logged out of |
| Your own trade log, timestamped in UTC | At the time of the trade | A clock independent of the platform's display settings and the broker's time zone |
| Dashboard screenshot of balance, equity, drawdown floor and daily loss state | At the close of each trading day | The firm's own displayed state on that date, before any later recalculation or rule revision |
| Every support ticket, ticket number, chat transcript and email | Continuously, exported outside the firm's helpdesk | What was asked, what was promised, and by whom |
Trace one fill and the reason for the schedule becomes obvious. Your order fills at the broker. The fill is written to the firm's database. The dashboard reads from that database, the history export reads from that database, and the risk engine reads from it too. Three doors, one room, and every door sits behind your login. When the account closes, whether by a breach flag, by your own decision or because the program ended, all three doors close together. The export you saved last Sunday is then the only copy on your side of the gate.
Continued access to your historical record is not something you control. That is prudence, not suspicion. The same logic applies to open risk during an account freeze, which we cover in what happens to open positions when a funded account is suspended. Export while you can, not on the day something goes wrong.
Why do UTC timestamps decide boundary disputes?
Because a boundary dispute is arithmetic on timestamps, and the clock in your platform screenshot is probably not the clock the rule is written in. MetaTrader is the clearest case. MQL5 documentation states that TimeCurrent() returns "the last known server time" and that "the time value is formed on a trade server and does not depend on the time settings on your computer." Brokers set that server clock to their own chosen offset, and they do not all follow the same daylight saving rulebook.
So an MT4 or MT5 history screenshot without the server's UTC offset for that specific date is an ambiguous record. cTrader is the contrast case: its Open API deal model carries createTimestamp, executionTimestamp and utcLastUpdateTimestamp, each documented as Unix time in milliseconds, which is UTC anchored by definition and needs no interpretation.
Here is the money attached to that ambiguity. Take an illustrative program whose daily loss limit resets at 17:00 America/Chicago, aligned to the CME trade date roll (confirm current hours and holiday schedules with CME Group directly, since they change). Your MT5 broker's base offset is UTC+2 and it follows the EU summer time schedule. Your History tab shows a fill at 00:30 server time. What trade date does it belong to?
| Date of the fill | Server offset | Fill in UTC | Fill in Chicago | Verdict against a 17:00 CT reset |
|---|---|---|---|---|
| Mid March 2026 (US on DST, EU not yet) | UTC+2 | 22:30 previous day | 17:30 previous day | 30 minutes after the boundary, so the new trade date |
| Late March 2026 (both regions on summer time) | UTC+3 | 21:30 previous day | 16:30 previous day | 30 minutes before the boundary, so the old trade date |
Identical clock reading on the screen, opposite sides of the reset, three weeks apart, with nothing changed except which continent had already moved its clocks. US daylight saving runs from 2 a.m. on the second Sunday of March to 2 a.m. on the first Sunday of November under 15 U.S.C. 260a. EU summer time runs from the last Sunday in March to the last Sunday in October, at 1 a.m. GMT, under Directive 2000/84/EC. In 2026 that puts the spring mismatch at 8 March to 29 March (21 days) and the autumn mismatch at 25 October to 1 November (7 days), 28 days in total. During those windows a Chicago clock and a UTC+2 base server sit 7 hours apart instead of the usual 8, one hour closer than the rest of the year. The spring gap changes length each year and the EU has repeatedly debated ending seasonal clock changes, so recompute for the year you are reconstructing. A trader who logged "00:30" cannot reconstruct the trade date. A trader who logged "21:30 UTC" can.
Ask for the calculation, not the conclusion
The most productive move in a written dispute is requesting the specific arithmetic the engine performed. Drawdown rules are where identical trading produces opposite verdicts on a definition buried in one clause.
Illustrative worked example. An account starts at $50,000 with a $2,000 trailing drawdown, so the opening floor is $48,000. You buy 2 E-mini S&P 500 contracts, which are worth $50 per index point. The position peaks at 6.00 index points unrealized: 2 contracts times 6.00 points times $50 equals $600, so peak equity is $50,600. You close at 1.00 point: 2 times 1.00 times $50 equals $100, so closed balance is $50,100.
| High water mark method | Peak used | Resulting floor | Verdict when equity later touches $48,400 |
|---|---|---|---|
| A: floor trails intraday unrealized equity high | $50,600 | $48,600 | Below the floor, so a breach |
| B: floor trails closed balance high only | $50,100 | $48,100 | Above the floor, so no breach |
Two floors $500 apart from identical trading. That is why "you breached me wrongly" is the weakest opening available: unfalsifiable, it invites a defensive reply, and nobody can act on it. The strong version has numbers in it. Which high water mark method was applied to this account? What value did the floor hold at 14:32:07 UTC when the flag was raised? Was that value derived from unrealized equity or from closed balance? Those questions are answerable and auditable, and they force a look at the record instead of at your tone. The same definitional ambiguity runs through prop firm consistency rules, where the measurement window and the denominator decide everything.
A question with a number in it gets answered. An accusation gets a template.
What a copier log proves, and what it does not
A server side trade copier keeps its own execution log on its own clock, independent of any single firm's dashboard. That log establishes what was instructed, at what moment, and to which accounts it was routed. It does not establish what the firm's risk engine computed, it does not override the firm's system of record, and where the dispute is about a drawdown formula rather than timing, it adds close to nothing. Decide which axis your dispute sits on before you offer it as evidence.
Where it genuinely matters is the fan out. One instruction leaves the master account and arrives at several funded accounts at several firms. Those fills do not land together: each broker acknowledges at its own speed, so the per account fill timestamps can spread across a wide interval. If the rule at issue is a daily reset, a news window or a session close, the relevant timestamp is the fill on the account being disputed, not the instruction on the master. Picture one instruction emitted at 21:59:58 UTC, three routed copies, fills confirmed at 21:59:59, 22:00:01 and 22:00:04 UTC, and a 22:00 UTC reset line drawn across all three lanes. Two of those fills sit on the new trade date and one sits on the old one, from a single click. A trader holding only the master log has documented the wrong event.
The instruction is short: record both the instruction time and the per account fill time, and keep them in UTC.
The escalation ladder, rung by rung
Rung 1 is first line support, in writing, factual and unemotional. Quote the exact rule text by clause number, give the timestamps in UTC, attach your export, state the specific outcome you are requesting, and keep it under a page. A support agent almost never has authority to reverse a risk engine decision, but does have discretion over whether your ticket gets escalated with context or closed with a template. An agent who feels accused becomes an obstacle rather than a router. Write for the person who will read the ticket after the agent.
Rung 2 is a formal written escalation to a supervisor or a compliance function. Restate the facts without editorializing, cite the date and reference of the rung 1 ticket, ask for the engine values above, give a reasonable deadline for a substantive reply, and say what you will do next without threatening. Most disputes are resolved or lost here, because that is where decision authority actually sits.
Whether arbitration is mandatory, which jurisdiction's law governs, where any hearing physically happens and whether class action is waived are all fixed by the clause you accept at signup. It silently caps every option at rung 3.
What external avenues actually exist?
Rung 3 is gated by three facts you cannot change afterwards: registration status, the dispute resolution clause, and elapsed time. Check all three before spending a day on it, and treat this section as questions for a qualified lawyer in your jurisdiction, not as answers to rely on.
Registration first, because the main US futures forums are registration gated. NFA's BASIC database publishes registration status and disciplinary history for NFA Members, and the lookup is free. CFTC reparations then applies its own test: eligibility requires finding out whether the individual or firm in your complaint "was registered with the CFTC at the time of the alleged wrongdoing or at the time the complaint is filed." The underlying statute, 7 U.S.C. 18(a), runs against a person "registered under this chapter" and allows a complaint "at any time within two years after the cause of action accrues." A pure evaluation program contract with an entity holding no registration generally falls outside that channel, and a registered affiliate does not automatically pull the evaluation contract inside it. The regulatory classification of retail evaluation programs is unsettled and varies by jurisdiction, so treat your counterparty's status as a fact to verify. NFA arbitration is a separate forum with its own membership gate and its own filing deadline; confirm both directly with NFA before you rely on either.
Reparations also requires that you are not involved in other actions on the same claim, which the CFTC describes as including NFA arbitration or a civil court action. So you pick one forum at a time instead of filing everywhere, though the CFTC does note that a complaint withdrawn before it is sent to the respondents may be refiled with the CFTC or in another forum if the two year deadline has not expired. Published tiers at time of writing ran from a $50 voluntary proceeding with no claim limit, through a $125 summary proceeding for claims of $30,000 or less, to a $250 formal proceeding for claims above $30,000. Confirm the current schedule and procedure before filing.
Then do the expected value arithmetic, because filing is a project. Say you clear the eligibility gate, the disputed payout is $2,400, the forum charges a $125 filing fee, and you honestly assess a 25% chance of full recovery. Expected gross recovery is 0.25 times $2,400, which is $600. Subtract the fee and expected net is $475. Price 12 hours of preparation at $50 an hour, which is $600 of your time, and expected net becomes negative $125. Filing only makes sense if you value that time below about $39.58 an hour ($475 divided by 12), or if you assess the probability materially higher. The probability is your own estimate, not a published figure, and fees and forum availability must be checked against your own agreement.
Civil and small claims options exist, but the arbitration clause usually decides them first. 9 U.S.C. 2 makes a written arbitration provision in a contract involving commerce "valid, irrevocable, and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract or as otherwise provided in chapter 4." The chapter 4 carve out was added in 2022 and concerns sexual assault and harassment claims, so it does not reach a trading dispute. That leaves only general contract defenses such as fraud or unconscionability, and whether any of them touch your agreement is a question for a lawyer who has read it. In practice the paragraph you skipped at signup determines your forum, your governing law and often the city you would have to travel to.
The payment side and its clocks
A chargeback addresses the purchase, never the trading outcome. Under 15 U.S.C. 1666i a cardholder can assert claims and defenses arising from a transaction against the card issuer, but the statute conditions that on a good faith attempt to resolve with the merchant, on the initial transaction exceeding $50, and on the transaction having occurred in the same state as the mailing address you previously gave the issuer or within 100 miles of it. Those amount and location conditions fall away where the seller is the issuer, is under common control with it, is a franchised dealer, or obtained the order through an issuer sponsored mail solicitation. Card network dispute rules are contractual and run separately, so an issuer may process a dispute the statute alone would not compel. Claims are capped at the credit outstanding on that transaction when you first notify the issuer, so the ceiling is the fee you paid, not the payout you are disputing.
The clock usually kills it anyway. Regulation Z requires a billing error notice, sent to the address the creditor discloses for billing inquiries, to be received "no later than 60 days after the creditor transmitted the first periodic statement that reflects the alleged billing error." Say you pay an evaluation fee on 3 March and it lands on the statement transmitted 12 March. Counting 60 days from 12 March gives 19 days left in March, plus 30 in April makes 49, plus 11 in May makes 60, so the deadline is 11 May. Your account is then closed on a disputed breach on 20 June. From 12 March to 20 June is 19 plus 30 plus 31 plus 20, which is 100 days. The statutory window closed 40 days before the dispute existed. Card network chargeback windows run on their own timers, vary by network and reason code, and are not set by Regulation Z, so confirm the applicable one with your issuer.
Two consequences worth acting on now. Pay with a credit card rather than a debit card, because the statutory routes differ: Regulation Z gives credit card billing error rights, while Regulation E is built around unauthorized electronic fund transfers, and a fee you knowingly authorized is not unauthorized. Debit cards may still carry network level dispute rights, so ask your bank, but the statutory floor is thinner. And understand that filing a chargeback may itself breach the program agreement, terminate the relationship and forfeit any remaining claim. It ends the relationship. It is not a negotiating tactic.
Is this documentation worth your time?
The maximal version is not. An elaborate manual journal with per trade annotations and a written rationale for every entry costs hours a week to insure against an event that may never happen, and most of it duplicates data the export already holds. Recurring cost, contingent payoff.
Automate the cheap parts and skip the rest. A scheduled weekly trade history export plus one end of day dashboard screenshot is roughly two minutes a week and captures most of the evidentiary value. Two notes on the export. In NinjaTrader 8, the Trade Performance window takes a From and To date plus an Accounts filter; take the Executions display rather than the aggregated Trades display, because NinjaTrader documents executions as all historical executions while a trade is a completed round turn with scaled entries and exits counted separately. In MetaTrader, save the history report from the terminal's History tab and record the server's UTC offset for the dates covered, because MQL5 documents those records as carrying trade server time. An export without that offset is half a record.
For one kind of trader even that is more than warranted: a single small account, rules they trade nowhere near the edge of, and an agreement mandating arbitration in a jurisdiction they will never travel to. No amount of documentation creates a forum that was never available, so the two minute floor is the correct and final answer there. Scale the effort to the size of the account and the reachability of the remedy.
Documentation does not promise a reversal. Most disputes are resolved or lost at rungs 1 and 2, external escalation is slow and frequently unavailable against an unregistered counterparty, and none of the above is legal advice. What it does buy is a closed evidence gap: if you are right, you can show it, and if you are wrong, you find out in an hour instead of a month.
Frequently asked questions
What evidence do I need to dispute a prop firm breach?
You need your own trade history export, a dated copy of the rule set as it read on the day you paid, UTC timestamps for the trades in question, and an end of day dashboard screenshot showing balance and drawdown state. Support correspondence with ticket numbers completes the file. Everything else is interpretation of those artifacts, and none of them can be recreated after account access ends.
Can I still get my trade history after a funded account is closed?
Not reliably, because dashboard and export access is a function of an active account rather than something you control. Whether the account closed on a breach flag, at your own request, or because the program ended, the dashboard and the export usually go behind the same locked login. Export on a schedule while the account is live rather than on the day something goes wrong.
Why do UTC timestamps matter so much in a prop firm dispute?
Because most boundary disputes are arithmetic on timestamps, and platform clocks are not UTC. MetaTrader stamps records in broker server time, and MQL5 documentation confirms that value is formed on the trade server independently of your computer's settings. If a US anchored daily reset is compared against an EU anchored server clock, the offset between them shifts by an hour twice a year, so the same displayed reading can land on either side of the reset.
How should I word a dispute ticket so it gets taken seriously?
Write it factually, quote the rule by clause number, give timestamps in UTC, attach your export, and ask for the specific calculation rather than asserting the decision was wrong. A question such as "which high water mark method was applied and what value did the floor hold at 14:32:07 UTC" is answerable and auditable. The agent rarely has authority to reverse a risk engine decision, but does decide whether your ticket is escalated with context or closed with a template.
Can I report a prop firm to the CFTC or NFA?
Only where the counterparty is actually registered, which you should check in NFA's BASIC database first. CFTC reparations eligibility turns on whether the firm was registered with the CFTC at the time of the alleged wrongdoing or at the time of filing, and the Commodity Exchange Act at 7 U.S.C. 18(a) allows a complaint within two years after the cause of action accrues. NFA arbitration is a separate forum with its own membership gate and filing deadline, which you should confirm with NFA directly, and the whole question is one to put to a qualified lawyer rather than a forum thread.
Can I charge back a prop firm evaluation fee?
Sometimes, but a chargeback reaches the purchase and never the trading outcome, so the ceiling is the fee you paid rather than the payout you are disputing. Regulation Z requires the billing error notice within 60 days of the statement that first showed the charge, a window that often closes before a trading dispute even arises. Card network windows run on separate timers set by network rules, so ask your issuer which applies, and treat a chargeback as a final step since filing one may itself end the relationship.
Is a credit card better than a debit card for evaluation fees?
Yes, structurally, because a different statute applies to each. Credit card purchases carry billing error rights under Regulation Z and claims and defenses under 15 U.S.C. 1666i, subject to that statute's conditions. Regulation E is built around unauthorized electronic fund transfers, and a fee you knowingly authorized is not unauthorized, so the statutory floor for debit is thinner, though network level dispute rights may still exist. Ask your bank rather than assuming either way.
Can I take a prop firm to small claims court?
That depends entirely on the dispute resolution clause you accepted at signup, which frequently mandates arbitration in a named jurisdiction. Under 9 U.S.C. 2 a written arbitration provision in a contract involving commerce is valid, irrevocable and enforceable except on general contract grounds such as fraud or unconscionability. Read that clause before you pay and ask a lawyer what it actually permits, because it caps every external option and cannot be renegotiated afterwards.
Does a trade copier log help in a dispute?
It helps on timing and intent only. A server side copier log establishes what was instructed, when, and to which accounts it was routed, on its own clock rather than any single firm's dashboard. It does not establish what the firm's risk engine computed and adds close to nothing where the dispute is about a drawdown formula. Record per account fill times as well as the master instruction time, since the fills are what the rules measure.
How much documentation is actually worth doing?
A scheduled weekly trade history export plus one end of day dashboard screenshot, which is roughly two minutes a week. That captures most of the evidentiary value at a fraction of the effort of a manual per trade journal, which costs hours weekly for marginal extra protection. Scale up only if the account is large or you routinely trade near a rule boundary.