A strategy and evidence review asks whether the conclusions being made about a trading method are supported by the evidence available. The review begins with the rules themselves, then examines the data, testing method, costs, risk and performance sample used to judge those rules.
A profitable backtest can be useful evidence. It is not, by itself, proof that a strategy has a durable edge or that future trading will reproduce the same result.
Define the strategy before judging it
Evidence is difficult to interpret when the strategy can change after every outcome. The first step is therefore to define what the method is supposed to do: market universe, setup conditions, entry, invalidation, position sizing, exit logic, time horizon and any discretionary decisions.
The Trading Systems framework separates a trading idea from a repeatable system. If the rules are not stable enough to reproduce, the review should identify that before focusing on performance metrics.
Evidence can come from different stages
A strategy may have manual chart studies, historical backtests, simulated forward tests, demo results or live trade records. Each source answers different questions and has different limitations.
Historical testing can show how defined rules would have behaved on available data. Forward testing can show whether the process survives new observations and real-time decision constraints. Live records add actual execution, costs and behaviour, but may still be too small a sample to support broad conclusions.
Data and costs matter
A review should ask where the data came from, whether prices and timestamps are appropriate for the strategy, whether missing observations or corporate actions matter, and whether the test accidentally used information that would not have been available at the decision time.
Transaction costs also need to match the market and trading frequency. Spread, commission, slippage, financing and market impact can change a result materially, especially when the raw edge per trade is small.
The educational Backtesting a Trading Strategy page explains why realistic assumptions belong inside the test rather than being added only after an attractive result appears.
Validation and overfitting
The more choices made while building a strategy, the easier it is to select a rule set that looks good on the same data used to design it. A review should therefore examine separation between development and evaluation data, parameter sensitivity, repeated testing and whether the method depends on one narrow period.
Trading Strategy Validation and the research pages on time-aware validation explain methods such as out-of-sample testing and walk-forward analysis. No single technique removes model risk, but disciplined validation can make unsupported confidence harder to hide.
Risk and regime dependence
Average return is not enough to describe a strategy. The review should consider drawdown, loss concentration, exposure, leverage, distribution of outcomes and whether performance depends heavily on one market regime.
A strategy can also be profitable overall while carrying a loss profile that is unacceptable for the capital available. That is why evidence review connects directly to Risk Management and Trading Performance & Analytics.
What the review can identify
A useful review can separate findings into three categories: what the current evidence supports, what remains uncertain, and what should be tested next. Examples include a rule that needs clearer definition, a cost assumption that is too optimistic, a sample that is too narrow, an unstable parameter, a concentration risk or a claim that is stronger than the data justifies.
The outcome should make the next research step clearer rather than simply label the strategy good or bad.
Research should remain falsifiable
MFXG's broader Applied Financial Engineering & Research approach treats a model or strategy as a hypothesis that must survive attempts to disprove it. When new evidence contradicts the original thesis, the correct response may be to revise or reject the method rather than search for a new explanation that preserves it.
Boundaries of the review
A strategy review does not guarantee future performance, provide a stream of trade signals or manage client capital. It does not convert historical results into a promise of return.
The scope is research, analytics and education. Personalized regulated recommendations, custody, brokerage and trade execution remain outside the stated scope unless separately authorized and legally permitted.
Requesting a review
Use the contact page to describe the strategy, the evidence you already have and the question you want the review to answer. If the main problem is execution rather than evidence quality, a Trading Process Review may be the more useful starting point.