MQL5 Quantitative Research & EA Optimization Project

I have an existing MT5 Expert Advisor (EA) written in MQL5. The source code ( .mq5 ) will be provided.

The underlying strategy is a trend-following system that performs well during trending market conditions, but requires professional quantitative research to improve its ability to identify and avoid highly sideways/range-bound market conditions.

The objective of this project is not to build a new EA from scratch.

The objective is to professionally research, refine, validate, and improve the existing system, with particular attention to:

• Dynamic range/sideways-market filtering

• Trading logic refinement (if necessary)

• Statistical robustness

• Parameter optimization

• Walk-Forward Analysis

• Out-of-Sample testing

• Monte Carlo stress testing

• Multi-asset and multi-timeframe research

• Robust .SET generation for live deployment

I operate using a single-instance / single-account isolation model, deploying separate .SET configurations for each asset and timeframe in order to manage risk and operational exposure.

I am looking for a top-tier Quantitative Developer / Quantitative Analyst / Algorithmic Trading Researcher, rather than someone who simply runs standard MT5 optimizations.

Total Project Budget: $2,000 USD

The project will be divided into 7 sequential milestones of $280 USD each. The lastest is 320 dollars.

Milestone 1 ($280 USD) will be awarded immediately upon selection.

The remaining milestones will be unlocked sequentially based on the quality of the previous research and successful delivery.

This structure is intentional: I want to establish a successful working relationship and validate the research methodology before expanding the project.

Important

The number of .SET files is not the primary measure of success.

I would rather receive fewer configurations that demonstrate strong statistical robustness than a large number of over-optimized configurations.

If a symbol/timeframe does not pass the agreed robustness criteria, it should be rejected rather than artificially optimized simply to produce another .SET file.

Milestone 1 — Quantitative Audit, Range Filter & First Validated .SET

Budget: $280

Scope

1. Perform a quantitative and technical audit of the existing EA.

2. Investigate the strategy's behavior during:

   • trending markets;

   • sideways/range-bound markets;

   • changing volatility regimes.

3. Identify and implement an appropriate mathematical/volatility-based mechanism to filter unfavorable range-bound conditions.

4. Refine or improve the trading logic where the research indicates that doing so could improve robustness.

5. Perform an initial parameter optimization.

6. Conduct appropriate Walk-Forward Analysis.

7. Conduct Monte Carlo stress testing.

8. Perform Out-of-Sample validation.

9. Analyze the results for signs of curve fitting and parameter instability.

Deliverables

• Updated .mq5 source code, if code modifications are made

• Quantitative diagnostic report

• WFA results

• Monte Carlo results

• Out-of-Sample results

• Explanation of the research methodology and conclusions

• 1 validated Master .SET configuration for the strongest initial symbol/timeframe candidate

The objective is to identify a robust configuration, not simply the configuration with the highest historical Profit Factor.

Milestone 2 — Core System Refinement

Budget: $280

Continue the quantitative research based on the findings from Milestone 1.

Possible work includes:

• further refinement of the range filter;

• parameter stability analysis;

• trading-rule refinement;

• execution logic improvements;

• additional WFA cycles;

• additional Out-of-Sample testing;

• Monte Carlo robustness analysis;

• elimination of unstable or overfit parameters.

Deliverables

• Improved .mq5 , if applicable

• Updated research/validation report

• Up to 2–3 additional validated .SET configurations

Only configurations that pass the robustness criteria should be delivered.

Milestone 3 — Conservative FX Majors & Metals Portfolio

Budget: $280

Apply the validated research methodology to major FX pairs and metals.

Potential candidates include:

• EURUSD

• GBPUSD

• USDJPY

• AUDUSD

• XAUUSD

The actual symbols should be determined by research results rather than predetermined assumptions.

Deliverables

Up to 4–6 validated .SET files, each associated with its corresponding symbol and timeframe.

Each configuration should be supported by appropriate validation.

A target Profit Factor of approximately 1.4 or higher may be considered desirable, but this is not a mandatory requirement by itself.

Robustness, stability, drawdown and Out-of-Sample behavior are more important than maximizing historical Profit Factor.

Milestone 4 — Equity Indices Research

Budget: $280

Research the applicability of the strategy to major equity indices.

Potential candidates include:

• S&P 500

• NASDAQ

• Dow Jones

• B3 WIN

• B3 WDO

The freelancer should determine which instruments and timeframes demonstrate sufficient statistical robustness.

Deliverables

Up to 3–5 validated .SET files, with corresponding symbols and timeframes.

Configurations that fail the validation criteria should not be delivered simply to meet a numerical target.

Milestone 5 — 24/7 Crypto & Weekend Markets

Budget: $280

Research the behavior of the EA in continuous 24/7 markets.

Potential candidates include:

• BTCUSD

• ETHUSD

• SOLUSD

Special attention should be given to:

• changing volatility;

• spread behavior;

• weekend market conditions;

• different liquidity regimes;

• parameter stability across different market periods.

Deliverables

Up to 3–5 validated .SET files, only for configurations that demonstrate acceptable robustness.

Milestone 6 — Aggressive / High-Yield Configurations

Budget: $280

Research a more aggressive parameter profile intended for higher return potential while remaining within predefined risk constraints.

The research may explore:

• higher risk/reward configurations;

• increased exposure;

• more aggressive position-management parameters;

• higher-return parameter regions.

A historical Profit Factor in the approximate 1.8–2.2 range may be considered a desirable research target, but it is not the primary acceptance criterion.

The freelancer should avoid sacrificing statistical robustness simply to reach a specific historical Profit Factor.

Deliverables

A package of validated aggressive .SET configurations for the strongest symbols/timeframes identified during research.

Each configuration should include the relevant performance and robustness statistics.

Milestone 7 — Prop Firm / Low-Drawdown Profile

Budget: $320

Research configurations specifically designed for environments with strict risk limitations, such as funded/prop trading accounts.

The research should focus on:

• low drawdown;

• controlled daily risk;

• exposure management;

• consistency;

• appropriate position sizing;

• avoiding configurations that are statistically profitable but operationally unsuitable for strict drawdown rules.

A daily drawdown target of approximately 3% or below may be used as a reference depending on the specific testing methodology.

Deliverables

A package of validated low-drawdown / prop-firm-oriented .SET configurations, together with the relevant statistical and risk analysis.

The central objective of this project is robustness rather than curve fitting.

The freelancer is explicitly authorized to recommend and implement improvements to the EA's source code when quantitative research indicates that they may improve the system.

This may include, where justified:

• range/volatility filters;

• entry conditions;

• exit conditions;

• trade management;

• execution logic;

• parameter architecture;

• risk controls;

• robustness improvements.

However, changes should be based on research and testing rather than arbitrary modifications.

The objective is to improve the underlying system itself, not merely optimize its parameters.

I am not looking for someone who simply presses "Start Optimization" in MT5 and selects the highest Profit Factor.

I am looking for someone capable of:

• forming quantitative hypotheses;

• testing those hypotheses;

• interpreting statistical results;

• identifying overfitting;

• rejecting weak configurations;

• improving the underlying code when justified;

• determining which symbols/timeframes are actually suitable;

• producing robust .SET configurations for live deployment.

Ideal Candidate

Strong experience in:

• MQL5 / MetaTrader 5

• Quantitative trading research

• Algorithmic trading

• Statistical analysis

• Walk-Forward Analysis

• Out-of-Sample testing

• Monte Carlo analysis

• Robust optimization

• Parameter stability analysis

• Risk management

Clean, modular and well-documented MQL5 code is highly desirable.

This is intended to be a long-term quantitative research project.

If the first milestones demonstrate that the EA has a viable and robust statistical edge, I intend to continue working with the same professional on additional research, optimization and future re-validation.

The milestone structure allows both sides to evaluate the quality of the collaboration before committing to the full project.

Quality is much more important than speed.

I would rather spend more time researching and reject weak configurations than receive a large number of .SET files that only work because of historical curve fitting.

To confirm that you have read the entire project description, please begin your application with:

QUANT2026

Please also briefly describe:

1. Your experience with quantitative EA research.

2. Your experience with Walk-Forward Analysis.

3. Your experience with Monte Carlo testing.

4. How you approach avoiding overfitting.

5. Examples of similar MQL5/algorithmic trading projects you have completed, if available.

6. Your preferred methodology for determining whether a .SET configuration is robust enough for live testing.

Initial Job Budget: $280 USD

Total Project Budget: $2,000 USD across 7 sequential milestones.