How Discipline OS prioritizes a correction

Discipline OS is TDLab’s behavioral correction engine. It runs deterministic detectors over execution data and reviewed fields, consolidates findings that involve largely the same trades and chooses the operational correction that deserves attention first.

From detectors to patterns

Detectors inspect objective execution data and, where applicable, fields you declared during review. Examples include post-loss activity, size changes, weak time blocks, weekday combinations, setup underperformance, plan adherence and execution-quality patterns.

The detector count shows how many checks could run for the selected accounts and range. A detector can be excluded when required fields or sufficient observations are unavailable. The help control in the page lists every detector and explains exclusions.

Discipline priority

The priority pipeline considers:

• Robust recovery: the positive historical change produced by the matching correction.
• Recurrence: whether the pattern repeats across independent sessions, episodes or weeks.
• Confidence: whether enough evidence supports a stable comparison.
• Overlap: whether multiple findings largely describe the same affected trades.
• Intervention breadth: whether a targeted cooldown can solve the issue before excluding an entire day or market window.

One operational correction

When findings overlap, TDLab keeps one recommended rule and marks the related patterns as covered by that correction. This prevents several cards from telling you to solve the same behavior in different ways. Rules with zero or negative simulated recovery are not recommended.

Verify, adopt and measure

Open the recommended rule in the Simulator to inspect the exact before / after and affected trades. If the intervention makes sense, promote it to the Playbook. New trades can then be evaluated as respected, violated or not applicable, while Behavior Score tracks adherence to decisions made in advance.