How to Build a Trading Plan: Rules, Risk Limits and Review
A trading plan is a written set of decision rules, constraints and review procedures that connects a market strategy with how it is actually implemented. It identifies eligible instruments, signals, exits, capital exposure, trading costs and conditions under which no trade should occur. It is a process document, not a promise of income or a substitute for understanding the risks of the product being used.
This guide builds a professional plan step by step, with clear examples of sizing, execution, risk oversight and post-trade review. It also explains where even a carefully documented plan can fail because prices gap, liquidity disappears or the underlying strategy loses its edge.
What is a trading plan and what should it achieve?
A trading plan is a written framework connecting an investor’s objectives, eligible markets, research methods, trade criteria, operational processes and risk limits. Its practical value is consistency: it makes clear what information must be known before exposure is taken, what risks can be accepted and when the plan should be reviewed. A plan does not guarantee profit or eliminate uncertainty.
A vague statement such as “trade only good opportunities” is not sufficiently actionable. Good conditions need definitions that can be assessed before the outcome is known. A plan also distinguishes market research from execution: a persuasive narrative is not enough if the instrument is illiquid, unregulated for the intended customer or otherwise incompatible with the capital available.
Start with objectives and constraints rather than an entry signal
Objectives can include long-term capital growth, risk-controlled exposure or systematic research, but they should not be confused with promised trading income. State the capital available, investment horizon, practical time available for monitoring, legal market access and the maximum loss that could be tolerated without affecting essential obligations.
The plan must reflect real constraints. A person unable to observe markets during a trading session should not assume a strategy needs continuous intraday supervision but will somehow operate automatically. Equally, a longer-horizon strategy exposed to overnight gaps requires different safeguards from a short-dated contract that may expire before planned review.
Define eligible markets and instruments
List the instruments that the plan permits, and the products it excludes. Equities, funds, spot FX, bonds, futures, options, CFDs and cryptoassets have very different legal, custody, financing and market-access characteristics. A plan should not treat a cash share and a highly leveraged synthetic exposure as interchangeable simply because both reference the same underlying index.
Record currency, exchange, settlement arrangements, contract multiplier and, where applicable, maturity or expiry. For UK retail customers, FCA product restrictions and specific CFD leverage protections are relevant. Explicit exclusions can be as important as permitted trades, particularly for instruments with poorly understood loss or margin characteristics.
Describe the economic hypothesis
A strategy rule is strongest when it follows from a plausible and testable economic explanation. A trend-following approach might hypothesise that prices adjust gradually to changing information, while a mean-reversion method hypothesises that certain deviations are temporary. These mechanisms differ and may work in different conditions.
Write why the proposed pattern should exist and what would cause it to disappear. If the strategy claims to exploit an interest-rate surprise, specify whether the new data are interpreted relative to previous expectations and which asset is affected. An explanatory story added only after a profitable backtest is weak evidence.
Define precise entry criteria
An entry rule should specify the signal, data source, observation timestamp, permissible instrument and order placement assumption. For example, “act after a completed daily close above the highest high of the previous 20 completed sessions” is more testable than “buy a breakout.” The precise lookback is an illustration, not a recommended optimal value.
Do not use closing information to assume an order was filled at the same close when the decision could only occur afterward. State whether the order is next-session market, limit, auction or another permitted type. Timing conventions can materially change performance results.
Define an invalidation condition
An invalidation condition is the evidence that would challenge the original market hypothesis. It is conceptually distinct from an arbitrary monetary stop. A fundamental thesis may be invalidated by a change in creditworthiness, while a technical breakout thesis may fail when price returns and remains below a specified range under the chosen rules.
It is possible for a strategy to suffer a price loss without its thesis being conclusively disproved, and vice versa. Nevertheless, a plan must state how exposure is managed while uncertainty remains. A process that always widens an invalidation level after adverse movement is not operating under a fixed risk framework.
Risk limits at trade and portfolio level
The plan should set risk constraints appropriate to capital, market exposure and the possibility of total loss. These can include a capital allocation budget, exposure limits, concentration constraints and scenarios for overnight gaps or stressed liquidity. A risk amount derived from the distance to a stop is an estimate, not a guaranteed maximum loss.
Portfolio risk is not simply the sum of ticker counts. Several long positions in technology shares, a Nasdaq index contract and a growth fund may respond to the same discount-rate shock. Correlated exposures should be aggregated when evaluating the plan’s true downside.
Worked risk-budgeted size example
Suppose a fictional £20,000 account sets an illustrative intended trade-risk budget of 0.5%, or £100. A proposed unleveraged share entry is £50 and a planned stop trigger is £48, giving £2 nominal distance. Ignoring fees and gaps, the nominal position-size calculation is £100 ÷ £2 = 50 shares. Gross purchase value is £2,500.
This is only a calculation under assumptions. If negative news causes the share to open at £45, an executed exit at that price produces £5 loss per share, or £250, before fees—more than double the intended budget. The plan should account for this possibility and should not claim the stop makes the risk exactly £100.
Choose and document exit types
A plan can distinguish thesis invalidation, time-based review, profit-taking, trailing adjustments and risk-reduction exits. Each needs a precise description. A target based on an arbitrary multiple of stop distance may not correspond to the economics of the market hypothesis, and price targets should not be presented as achievable fill guarantees.
A stop order can become a market order after its trigger is reached, with execution depending on available liquidity. A stop-limit constrains the price but may remain unfilled. The SEC’s 2026 updated investor bulletin specifically warns about this distinction. See Stop-Loss Orders Explained.
Specify order routing and execution expectations
Execution rules should identify how orders are submitted, what spread or depth conditions make a trade unacceptable and how partial fills are managed. A strategy with a small expected gain can become negative after realistic commissions, bid–ask spreads, slippage and financing. Record actual execution rather than relying only on the signal price.
A pre-trade checklist can compare the intended quantity with available depth and ensure that the correct security identifier or contract month is selected. For markets without continuous cash trading, special opening and closing auction behaviour should be considered.
Decide how leverage and margin are handled
Margin posted is not the same as the economic amount at risk. A £2,000 collateral deposit might support much larger notional exposure under a derivative’s rules, making small underlying price changes financially significant. Additional margin demands can arise even when the original directional hypothesis remains unchanged.
Under UK FCA requirements, retail CFD leverage limits vary by underlying and covered providers must implement specified margin close-out and negative-balance protections. These do not guarantee that a CFD account will preserve capital or that losses stay small. Different products have different obligations; product-specific terms must be checked.
Build an economic-event calendar
Scheduled company earnings, inflation data, employment releases, central-bank decisions and contract expiries can produce gaps or sudden changes in liquidity. A strategy may choose to allow, reduce or avoid exposure around specified events, but the choice needs to be defined before the event and evaluated for its costs and consequences.
News-event rules should compare expected with actual releases and consider that markets may anticipate outcomes. Avoid using the description “high-impact” as if it guarantees a particular price direction. For commodity futures, delivery notices and expiry requirements may be just as important as macroeconomic announcements.
Operational safeguards and error procedures
A useful plan contains procedures for platform failure, connectivity loss, mistaken orders, stale prices and account access problems. It should identify which orders may remain live if a device disconnects and how to verify their actual status through the broker’s systems. Cancellation requests do not automatically prove that an order was never filled.
Operational mistakes can create losses unrelated to the quality of market analysis. Incorrect contract multipliers, expired instruments, mismatched currencies or unintended short exposures are examples. A practical pre-trade workflow can reduce some errors but cannot eliminate every technical or human failure.
Set explicit no-trade conditions
A robust plan should describe situations in which exposure is not permitted: insufficient liquidity, unavailable data, excessive concentration, unclear product terms, unreliable quotes or market events that create unacceptable uncertainty. This prevents the trading process from becoming an obligation to act whenever markets are open.
No-trade rules should be economically motivated and reviewable. A rule that is revised retrospectively to exclude losing historical trades creates biased performance evidence. If a filter is added, subsequent evaluation should include out-of-sample periods or forward observation rather than treating the modified backtest as fresh proof.
Trading journal: what to record
Record the decision timestamp, thesis, source data, planned exposure, entry and exit rules, execution prices, fees, financing, market conditions and reasons for subsequent changes. A journal should include rejected and cancelled signals where practical, because recording only profitable or executed examples can distort the strategy’s assessed quality.
Separating process errors from losses due to unavoidable uncertainty is important. An adverse market move need not imply a poor decision; an impulsive trade that happened to earn money may still represent a failure to follow the plan. Review evidence, not just the final profit figure.
Performance metrics that belong in the review
A review should consider net return, average win and loss, win rate, expectancy, profit factor, maximum drawdown, volatility and exposure concentration. Each statistic answers a different question and can be misleading in a small sample. A 90% win rate may conceal rare catastrophic losses.
Report the sample length, number of trades, eligible markets, fees and position-sizing rules. Changes in trade frequency can alter cost estimates. Backtested results are hypothetical and should be differentiated clearly from independently verified live outcomes. See Risk–Reward and Expectancy.
Separate strategy research from optimisation
Research starts with an economic rationale and interpretable rules. Optimisation adjusts parameters within a predefined process; searching thousands of combinations to discover the best past result can select noise. The more alternative ideas tried, the less reliable an unadjusted apparent success becomes.
Use chronological training and genuinely untouched evaluation periods, and assess whether results are robust to modest parameter changes. A strategy profitable only when a moving average is exactly 47 days rather than 45 or 50 is more fragile than one with broadly similar behaviour across sensible values.
Stress-test losses, not just average performance
Historical average returns conceal the possibility of clustered losses, sharp gaps and changes in correlation. Stress tests can examine a severe volatility spike, unexpected policy announcement, exchange outage, depegging event or liquidity freeze. These scenarios are not predictions; they expose dependencies that may not appear in normal-period backtests.
For a multi-asset plan, examine simultaneous losses across correlated exposures and whether financing or margin requirements increase. Capital adequacy depends on the path of returns and ability to meet obligations, not solely on a positive long-run arithmetic mean.
When and how to revise a trading plan
A plan should have a review process rather than be frozen forever. New regulatory requirements, market structure, costs or compelling evidence can justify revision. However, changes should be documented with a date and reason so subsequent performance can be attributed to the appropriate version of the rules.
Do not interpret every losing sequence as proof that a strategy must be redesigned, nor every winning sequence as confirmation that it should be enlarged. Assess whether outcomes are within plausible statistical variability or whether the underlying economic mechanism has changed.
Behavioural discipline and decision hygiene
A trading plan helps manage overconfidence, anchoring and fear of missing out by making decisions explicit before outcomes are known. It cannot remove emotion, and a rule followed mechanically can still be harmful if the premise is wrong. Periodic review should test both adherence and the quality of the model.
The plan should discourage retrospective justifications. If a trader widens a stop, adds exposure or changes time horizon, record whether that action was permitted and the evidence supporting it. Outcomes alone are not sufficient to judge the decision.
A practical pre-trade review sequence
First confirm market, instrument, correct contract and account. Verify the strategy signal using data available at the decision time, then note the competing interpretation. Check upcoming events, the exact order type, current quotes and sufficient depth. Calculate intended position size, notional exposure, potential gap loss and aggregate portfolio concentration.
Finally, document the intended exit and what would invalidate the idea. A rational conclusion may be not to trade because the expected benefit is too small after cost or the risk cannot be measured credibly. The plan should make that outcome easy to choose.
Apply the Atravix six pillars to the plan
The Atravix® framework provides six research lenses: Trade Idea Generation, Fundamental Analysis, Sentiment Analysis, Technical Analysis, Risk Management and Trading Psychology. The trading plan converts those lenses into permitted activities and documentation standards rather than pretending they are independent buy or sell signals.
For a particular market idea, identify the catalyst, fundamental basis, current market expectations and technical execution conditions. Then assess downside, concentration and behavioural bias. A disciplined final judgement can be a rejection when evidence is weak or the instrument’s operational risks exceed the plan’s constraints.
Trading plan checklist
| Plan component | Requirement | Verification question |
|---|---|---|
| Market and instrument | Exact security, contract and venue | Is the exposure legally available and understood? |
| Hypothesis and signal | Dated rule and information set | Could the signal have been known before the trade? |
| Risk budget | Capital, position and stress limits | What happens after gaps or correlated losses? |
| Execution | Order type, spreads and funding | Could the simulated fill have occurred? |
| Review | Journal, metrics and version control | Is the process repeatable and falsifiable? |
A trading plan is a documented process rather than a guarantee of performance.
Frequently asked questions
What are the essential parts of a trading plan?
Objectives, eligible markets, market hypothesis, entry and exit rules, risk budgets, costs, order procedures and review.
How much money should a trader risk per trade?
There is no universally suitable percentage. It depends on financial circumstances, total exposure, product terms and losses beyond intended stop levels.
Is a stop price sufficient risk management?
No. Gaps, execution failures, leverage and correlated portfolio exposures can create losses different from a stop-distance estimate.
Should a plan be changed after a losing trade?
Not automatically. Review whether the outcome is consistent with expected uncertainty or shows a meaningful rule, execution or model failure.
Can a plan tell someone when not to trade?
Yes. Clear exclusions and no-trade conditions are essential parts of a disciplined process.
Related learning
Understand trading strategy design, position sizing, exit orders and backtesting limitations.
Sources and further reading
- SEC Investor.gov — Order Types
- FCA — Retail CFD Requirements
- CFA Institute — Active Investing and Strategy Pitfalls
Important notice: General education only, not a personalised trading plan or financial advice. Sample percentages, position sizes and prices are hypothetical and do not ensure an actual loss limit. Financial products can lose value, and leveraged positions can cause rapid losses.