What is mistake tracking?
Mistake tracking is a structured method for recording errors you make during trading-related decisions and later reviewing them to understand why they happened. “Mistake” here usually means a preventable or correctable problem in your process, such as using a plan you did not follow, violating a rule you wrote for yourself, misunderstanding available information, or reacting to information in a way you did not intend.
The purpose is learning about your behavior and decision workflow, not forecasting. Mistake tracking works best when it is treated as documentation of what you did, what you observed, and how you responded—so you can test whether your process is improving.
How does mistake tracking work?
A practical approach is to capture each “mistake event” with the same minimum set of fields. Common fields include:
- Time and date (when the decision was made)
- Situation description (what you believed was happening at the time)
- Your planned process (what you intended to do)
- The actual action (what you did)
- Why it happened (your best explanation)
- Outcome measure (what happened next, using a consistent rule)
Define your terms before you analyze. For example, decide what qualifies as a “process violation” versus a “market condition mismatch.” Decide what “outcome” means in your tracking. If you use an example, state assumptions: for instance, “Assumption: I measure outcome as whether my decision complied with my written rule, not whether the price later moved in my favor.” That separation helps prevent a common misunderstanding: labeling a mistake based on hindsight price movement rather than on the decision error itself.
Evidence and example (why structure matters)
Consider two logged events:
- You entered based on a rule you later realized you had not checked.
- You entered based on your rule, but execution details differed from what you expected.
If your log records only the final outcome, you may mistakenly conclude that the rule “doesn’t work.” If your log distinguishes rule-check failures from execution reality, your review can be more accurate: you can target “rule compliance” versus “execution assumptions.” This is a factual, process-level correction: the lesson is about what was different between planned and actual behavior.
A simple review method is to summarize mistakes by category and then read the raw entries behind the counts. Counts alone can be misleading; for example, one category may appear frequently because it is easy to recognize. The raw context shows whether the “why” you wrote matches the actual timeline of events.
Limitations and risks
Mistake tracking has several material limitations:
- Correlation versus causation: A repeated error may be linked to worse outcomes without being the cause. The “gevolg” (consequence) can look like “preventie” (a fix) even when the real driver is elsewhere.
- Changing conditions: Market and personal conditions can change. Historical patterns do not guarantee future results.
- Incomplete or inconsistent data: If you record decisions after the fact, miss relevant context, or change your definitions mid-way, your conclusions may be internally inconsistent.
- Execution and cost effects: If your “outcome” ignores costs and execution differences, your review may treat differences caused by fees or fills as decision-quality problems.
A key failure mode is hindsight labeling: you mark mistakes only when the trade later performs poorly, even if the decision process was correct at the time. This is a feasible error because it feels intuitive, but it prevents a truthful test of your process.
Verification and next questions
Independent verification means checking that your logs match reality. For example, verify that timestamps, decision notes, and execution records align with what actually happened in your activity history. Also check whether your “mistake” definition is applied consistently: if you review ten events and find that the same type of error is sometimes called a mistake and sometimes not, your tracking is unreliable.
To improve clarity, you can ask questions like: Which “mistake categories” are actually actionable? Which parts of my log are based on assumptions I cannot verify? Where do my entries confuse decision errors with outcome randomness?
If you want, you can also explore the detailed limitations and associated risks in separate guides (mistake tracking) and compare your own definitions against those discussions via the pages about limitations, risks, and advanced considerations.