Prop Firm EA Setfile: Inputs, Testing and Safe Sharing

A setfile saves input values for an Expert Advisor. It can change only behaviour that the EA already implements and exposes through those inputs. It cannot add a news filter, a drawdown calculation or a trading strategy that is absent from the program. This guide concerns downloadable EA presets, not the separately hosted JPTC Algo.

Match the preset to the exact EA

Record the EA name, platform, executable version or hash, preset version, symbol and timeframe. Compare the loaded values with the developer's input documentation. Similar names do not prove compatibility between different EAs, versions or MT4/MT5 programs.

The official MetaTrader guide explains saving and loading input presets. Load is not a compliance or profitability check. Verify the actual values after loading; a file opening successfully does not prove that every intended setting was applied.

Check units before changing risk settings

The name RiskPerTrade does not define its units. A developer might interpret 0.01 as a fraction representing 1%, or as 0.01 percentage points. Likewise, a stop input might use points, pips, price or money. Read the actual schema and test the resulting order calculation.

MaxLots is a volume limit in lots when the program defines it that way; it is not a percentage of account equity. Fixed volume and percentage-based sizing both require checks of stop distance, contract size, tick value, volume steps, currency conversion, margin and concurrent positions. Neither mode is automatically safer.

Separate platform support from account permission

Record the firm, purchased programme, phase, account size, platform, add-ons and applicable rule date. Check third-party software, strategy, access and copying restrictions separately. A preset named FTMO or FundedNext is not provider approval. A margin closeout, a trading-rule breach and an unmet Reward condition are different events.

For a dated example, FundedNext's CFD EA policy checked on 16 October 2026 permits eligible MT4/MT5 accounts below $50,000 only with the paid EA add-on. From $50,000 it requires fully manual trading, including SL/TP and lot tools; cTrader and Match-Trader do not permit EAs. These conditions do not establish permission for another product or strategy.

Test the whole configuration and its failure handling

Save a reproducible test package: exact executable and preset, input meanings, historical data and modelling mode, symbol/account specifications, costs, execution delays, period and report. Test daily reset time, open and closed P&L, commissions, swaps, other EAs and manual positions against the actual account formula. A spread threshold does not guarantee a slippage limit.

Check initialization, rejected orders, reconnects, restarts, missing data and attempted stop/close actions in an authorised test environment. Gaps, delayed fills, outages and software defects can still cross a limit. A preset cannot guarantee that an account stays inside its daily or total loss allowance.

Worked examples: costs and a separate loss budget

Fictional cost illustration, not a broker quote: assume 50 trades, an additional cost of 1 pip per trade and $10 per pip for the stated position. The total is 50 × 1 × $10 = $500, not $50. Actual costs depend on instrument, contract, volume, currency and execution.

Fictional budget illustration, not a recommended setting: a $25,000 reference amount and a separately chosen 2.5% internal budget give $625. If each loss were exactly $250, two would total $500 and three $750, exceeding $625. Include existing exposure and costs before interpreting any remaining room; an internal stop is not a provider guarantee.

Optimise without treating test scores as certification

Keep parameter selection separate from genuinely unseen evaluation data. A walk-forward procedure uses chronological selection and evaluation windows; record their lengths and every reselection. A strong historical result or a smooth curve alone proves neither robustness nor overfitting.

Profit factor is gross profit divided by the magnitude of gross loss when the denominator is non-zero. Sharpe calculations require a stated return series, reference rate, sampling and annualisation method. No universal PF, Sharpe, win-rate, 99% modelling label or fixed number of years certifies a preset. Review changes to rules, software or measured behaviour; record and retest changes instead of imposing a monthly retuning or fixed risk-reduction recipe.

Share a documented configuration, not an implied permission

Check the actual software licence, preset distribution terms and provider restrictions before sharing or selling. Being a text configuration does not itself grant redistribution rights or permission to run identical trades on other accounts. Remove account identifiers, credentials and other private values.

Include the exact EA/preset versions, units, tested environment, periods, costs, known limitations and change log. State which results are simulations and which are actual records. Community feedback can help find errors but does not replace your own permission and configuration checks.

What to check in the JPTC EA Hub

The EA Hub supplies downloadable software and relevant licence access. Check the actual file-specific documentation, available inputs, included presets and test report; no article-wide claim establishes that every file has five years of tests or permission at every firm. The separate hosted Algo requires no customer installation or VPS.

The results page publishes monthly figures from that service's own broker statement. It is not an independent audit, a pass-rate study or evidence for another EA/preset. Check current product, checkout and affiliate terms directly; no fixed referral payment, customer result or successful evaluation is promised here.

Official documentation: EA inputs and presets · Strategy Tester · Test report metrics · Sharpe methodology · FundedNext CFD EA policy.

JPTC EA Hub · JPTC Algo · Affiliate

Hosted JPTC Algo broker-statement images

These service images do not prove another EA’s results or provider permission.

@jptradingcapital on Instagram →

FAQ