Training Analysis Workflow

Training analysis is useful only when it improves a real decision. This workflow prevents a dashboard full of metrics from becoming a collection of unrelated scores.

Before you begin

• Choose one decision and time range, such as whether to keep tomorrow's key workout.
• Let relevant activities reach Ready status.
• Confirm thresholds, time zone, sport filter, and load-source priority.
• Link completed activities to planned workouts when the plan is part of the question.
• Record subjective context such as session RPE or current state when available.

1. Define the question

Start with a decision, not a metric:

• Should the next key workout remain as planned?
• Did this week deliver the intended training stress?
• Is a repeated execution problem caused by schedule, load, or workout design?

Set the relevant sport and date range. Do not mix a seven-day execution question with an all-time chart.

2. Compare plan and actual

Before reading any score, answer:

• What was planned?
• What was completed, skipped, moved, or left unlinked?
• Which key sessions differed in duration, intensity, structure, or load?
• Was the difference intentional?

If this step is skipped, load metrics lose their training purpose.

3. Validate the evidence

Open the activities behind the totals and check:

• Processing status and source provider.
• Sensor availability and obvious data errors.
• Threshold and zone context.
• Which source produced activity load.
• Terrain, weather, equipment, or other execution context that changes interpretation.

An explicit missing signal is better than filling the gap with a precise-looking guess.

4. Read load and response together

Use the PMC signals as a set:

• CTL describes the longer load trend.
• ATL describes recent training pressure.
• TSB shows the difference between longer and recent modeled load.

Then compare them with execution quality, subjective fatigue, and a relevant response trend. Load describes applied pressure; it does not prove adaptation by itself.

Trainingload.ai

Training load in decision context

Use one time range to connect CTL, ATL, and TSB with the activities and active-plan period that produced the trend.

Plan

4 week block

Load

steady build

Activity

matched

AI review

ready

5. Draft one conservative change

Change one primary variable first:

• Keep frequency and reduce one session's intensity.
• Keep the workout purpose and shorten duration.
• Move one key session by 24–48 hours.
• Leave the plan unchanged and collect the next key-session result.

State what should remain unchanged so the adjustment does not silently rewrite the entire training block.

6. Confirm and record the rationale

Before applying a plan change, record:

• Trigger: the plan/execution or load pattern that prompted review.
• Decision: the exact workout and field that changes.
• Protected intent: what the plan should still accomplish.
• Expected result: what should improve or stabilize.
• Review point: the next activity or date that will test the decision.

When AI Coach prepares the adjustment, inspect the proposed workout and confirm it only after these five items are clear.

Verify the decision

• Every conclusion can be traced to a plan, activity, setting, or trend visible in Trainingload.ai.
• Observed data is separated from interpretation and uncertainty.
• Only the intended future workout changes.
• Completed and skipped history remains intact.
• A specific next review point is defined.

Troubleshooting

• The metrics disagree: compare date range, sport, thresholds, and load sources before choosing one signal as “correct.”
• The chart changed after a settings edit: identify the changed threshold or source priority and reinterpret history consistently.
• Plan and actual cannot be compared: link the correct activity and confirm the planned workout has enough structure for expected load.
• AI suggests a large rewrite: ask for one conservative change with protected constraints and require confirmation.
• There is not enough evidence: keep the plan unchanged or make the safest reversible adjustment, then collect the next relevant session.

Next steps

• Manage an Active Plan
• Activity Detail
• AI Coach