A CP test is designed to test and measure your current fitness and should be performed regularly (every 4-6 weeks) as part of any training block.
If performed optimally, it provides an accurate, up-to-date critical power which then determines power zones for the next 4-6 week training period.
There are two protocols typically used for critical power tests:
- Two-part test
- 5k or 10k race
Two-Part CP Test
Common options include: 3/10 minute or 9/3 minute. The two parts include a short and medium-length segment (both between 3-30 minutes) with sufficient rest between.
9/3 minute and 6/3 lap are recommended by Stryd while 3/10 is recommended by Steve Palladino. The latter from research showing a 10 minute or greater duration is less prone to error/overestimating CP.
CP was considerably overestimated when only trials lasting less than 10 min were included, independent of the mathematical model used.Critical power: How different protocols and models affect its determination
For that reason, I defer to Steve’s recommendation of a 3/10 protocol for most runners and 3/12 or 3/15 for more experienced runners. For the best result, test the shorter duration first and then the longer duration (i.e. 3/10 versus 10/3).
5k or 10k Race
An alternative approach to estimating critical power is using a 5k or 10k race.
One benefit of this approach is that it’s easier to produce max-effort performances in a race context versus solo. It also provides a chance to practice racing (e.g. warmup, pacing, mental toughness, etc.).
However, registration fees make this a costlier option unless there’s a free parkrun near you.
Because a 10k is typically run at critical power, it will generally estimate CP better than a 5k. However, most people should opt for a 5k instead. There’s less stress on the body in a 5k which means recovery is faster—minimizing downtime and letting you resume training more seamlessly.
On the flip side, if you know you can recover quickly from a 10k then you might consider doing one to shift things up.
Which Test is Best?
There’s really not one clear winner. In fact, I would recommend doing both a 3/10 test and 5k race regularly, ensuring there’s at least one of each in the last 90 days.
The 5k race to practice racing and the 3/10 test to practice pacing and mental grit.
It’s not uncommon to go out too strong in a 3/10 test and burn out before the end. The more you do 3/10 tests, the better you’ll get at running a controlled, even pace over the entire duration. It’s also a great exercise in mental grit. When you’re running by yourself and things start hurting, you have two options: slow down because nobody is watching or hold on as long as you can.
A 3/10 test lets you practice grit which translates on race day. What you do in training is what you’ll do in a race.
Another benefit of doing both types of tests is better data for Stryd’s auto CP, WKO, Golden Cheetah, or other platform using a power-duration curve to estimate (model) critical power. These platforms look at your max effort runs at various durations (typically within the last 90 days) and estimate your critical power from that data.
The more data points at different durations, the more accurate those models. By doing both types of tests, those models get three different durations (a 3 minute, 10 minute, and 5k time) which dramatically increases accuracy.
And if you throw in a 10k once every 90 days (particularly for sub-20 minute 5k runners), that further refines your critical power estimate.