Direct answer
Strategy Tagging is a structured method for assigning labels (tags) to trading activity based on predefined, repeatable rules. The core “rules” are not about forecasting; they are about making your labeling consistent enough that another person can apply the same criteria and reproduce the same tags for the same decisions.
To explain Strategy Tagging accurately, treat it as a classification and documentation system. You define: (1) what decision moment you tag, (2) which inputs are allowed, (3) which tag categories exist, (4) the exact conditions under which each tag applies, and (5) how you compare outcomes afterward. If any of these parts are vague, the tags become subjective and lose verification value.
Mechanism or definition
A practical rule set for Strategy Tagging has five building blocks.
1) Define the “unit of tagging”
Choose the smallest decision you will label. For example, you might tag the setup at entry, the decision that changed your position size, or the reason you held versus exited. The rule should be explicit: the same kind of event must always be tagged the same way.
Assumption: you can identify the decision moment from your journal or logs (timestamped notes, execution records, or both). If the decision moment is ambiguous, tagging rules cannot be applied reliably.
2) Separate stable mechanics from variable conditions
Tags should primarily represent stable characteristics of your decision process (for instance, “risk was defined before entry” or “the plan included a specific exit condition”). Market conditions, spreads, slippage, and execution quality are often variable, so they should either be captured as separate labels/fields or treated as contextual variables when you analyze performance.
Rule: do not mix “what you did” with “what happened.” If a tag changes because the market moved differently, your tagging criteria may be measuring outcomes rather than decision traits.
3) Specify allowed inputs
Strategy Tagging rules must state what information can be used to assign tags.
Examples of inputs (conceptual, not platform-specific):
- Journal text fields you explicitly record before or at the decision moment.
- Pre-trade plan details such as whether an exit level was written down.
- Execution facts you observe in your trading records.
Limitation by design: if you allow hindsight information that you only notice after the trade, tagging can become outcome-driven.
4) Write “if-then” conditions for each tag
Each tag needs a clear application rule. A good rule is testable: someone can check whether the conditions are met.
Example structure (generic):
- Tag A applies if the plan contained a predefined exit rule AND you recorded it at or before entry.
- Tag B applies if the plan lacked a predefined exit rule.
- Tag C applies only when both conditions are simultaneously true.
If a journal entry contradicts itself, you need a conflict rule, such as “use the earlier record” or “mark as inconsistent.” That conflict rule is part of the tagging system.
5) Decide how many tags apply
Your system should define whether you use:
- Single-label tagging (choose the best matching category), or
- Multi-label tagging (apply multiple tags that reflect different traits).
Rule: mixing single-label and multi-label logic without stating it leads to comparisons that are not apples-to-apples.
Evidence or example
A testable micro rule set can look like this (illustrative, because it does not depend on live prices).
Assume you are tagging each journal entry at the time you place a trade. You define three tags:
- “Exit-defined”
- “Exit-not-defined”
- “Conflicting-plan-noted”
Assumptions:
- Your journal includes a written exit statement (or an explicit “no exit plan” note).
- You can tell whether the note exists at decision time.
Rules:
- If the journal contains an exit rule recorded at decision time, apply “Exit-defined.”
- If there is a decision-time note stating no exit rule, apply “Exit-not-defined.”
- If the journal includes both an exit rule and a later contradictory note for the same decision moment, apply “Conflicting-plan-noted.”
Verification step (the key): take a past set of journal entries and have another person apply the rules. If they label the same entries differently, your rules are not yet precise enough.
Next, you can analyze outcomes later, but the tagging rules themselves should not assume any particular outcome. A limitation is that outcomes can vary due to costs and execution, which are not controlled by tagging. So even if tags are applied consistently, the results you observe are still uncertain.
Limitations and risks
Strategy Tagging has material limitations and failure modes.
Data inconsistency and ambiguity
If your journal notes are missing, vague, or written after the fact, you can still tag, but the tags may reflect uncertainty rather than the decision trait. A common failure mode is “filling gaps” with assumptions. That breaks verifiability.
Rule you can use to reduce this risk: include an explicit “insufficient evidence” label when decision-time inputs are missing.
Outcome leakage (hindsight influence)
If tagging criteria allow information that becomes known after the trade outcome, you risk creating labels that correlate with what happened, not what was decided. Even subtle leakage—such as adjusting tags after seeing results—can invalidate comparisons.
Confusing tags with signals
Strategy Tagging is often confused with a “trading signal” because people may start to interpret tags as predictive. A rule for keeping the method grounded: treat tags as descriptors of process, not as standalone instructions to trade.
Market, cost, and execution variability
Outcomes depend on market conditions, costs, execution quality, and jurisdiction-specific constraints. Historical relationships do not establish future results. Therefore, you should avoid statements that imply predictive accuracy or stable profitability from tagging alone.
Overfitting to your own style
If your tag categories are tailored only to your current routine, tags may become less useful when your process changes. A rule set should include periodic review: when the decision process evolves, the tagging rules may need to be updated, or you should treat changes as a separate contextual variable.
Verification or next question
To independently verify Strategy Tagging facts, focus on reproducibility and testability.
- Re-run the tagging on the same historical journal entries using only the allowed inputs.
- Check inter-rater consistency: can another person apply your if-then rules and reach the same labels?
- Confirm that your tagging rules do not reference outcome knowledge.
- Document your assumptions explicitly (what inputs exist, what counts as decision time, and how conflicts are handled).