When to test, when to skip, Yousuli Help

The tradeoff

Tests cost: a 20-minute FTP test takes 2 to 4 days of full recovery; a CP test (3 min + 12 min) takes 3 to 5 days; a metabolic test takes a full day and partial recovery. In an 8 to 12 week build, that recovery cost adds up.

The decision: is the data worth more than the lost training time?

When to test

Definitely test

Before an A race, 4 to 8 weeks out, if the current zones are 90+ days old. The race readiness check will surface stale thresholds.
At the start of a new training relationship. Without baseline zones, every prescription is a guess.
After a major training change (new periodization, new block focus, new sport added). The old zones may not represent the new state.
After a major life change (4+ weeks off, illness, significant weight change, fit change). Anything that could plausibly move the threshold by more than 5 percent.

Probably test

Every 6 to 8 weeks during a structured build, for any threshold the plan depends on.
After a peak race (the race itself often functions as a test).

Probably skip

Mid-block when zones were tested less than 4 to 6 weeks ago and training has been consistent.
In recovery weeks. The test would be misleading.
Within 2 weeks of an A race. The recovery cost is too high; trust the zones you have.

Definitely skip

During the taper for an A race. You go into the race with the zones you have.
When the athlete is sick or injured. Either you get an artificially low value or you make the situation worse.
As a motivational tool. Athletes who are motivation-shopping for a test are usually looking for a number to validate a feeling. The number does not solve the feeling.

Alternatives to a formal test

Often you can update a threshold without a structured test:

Race results: a 10K race functions as a run threshold test. A 70.3 race functions as a long-course CP test. Use the race FIT to extract a fresh number.
Best 20-minute power from a hard training ride. Not as clean as a structured test but better than 3-month-old data.
Best 100 m TT within a swim set if the athlete was fresh going into it.

The Hub's best efforts ladder and power curve make these extractions easy to spot.

What to do after a test

Whether the test was structured or extracted from training data:

Log the test as an Assessment (see Logging an assessment or the dedicated assessment articles).
Confirm the calculator-derived threshold goes into MetricHistory as the new canonical value.
The Hub auto-flags the new value as needs-review if it differs by more than 20 percent from the previous value (see Activity quality flags).
Accept or decline the new value from the metric review page.

Common testing mistakes

Testing too often because "more data is better". Recovery cost erodes the training adaptation.
Testing under stress (work week, travel week). The number is artificially low and you train the next block off a bad baseline.
Skipping the structured protocol ("I'll just do a hard ride"). The number is not comparable to the baseline.

When to test, when to skip