The Suunto Ambit3 Manual 6: Recovery Testing

With firmware 2.0 (released end of June 2015), Suunto gave its Ambit3 line two new features powered by Firstbeat Technologies: recovery testing and running performance.

Time to update my Ambit – or now, really Ambit3 – manual

Recovery Time

An estimation of the recovery time needed after a training session has been a part of the Ambit3 line from the beginning.

In fact, to be exact, post-exercise recovery time was already displayed by the Ambit2 and even the original Ambit. The Ambit3 line, however, added an active recovery estimation/tracking which takes into account not just the stress of an exercise, but also the amount of activity done afterwards.

(This feature is still often mistaken for an activity tracking á FitBit, and it could be implemented like that, but this is not how it is now and how it was meant to be: the data recorded is not ‘translated’ into steps and not stored anywhere it would be accessible once it has been cleared from the Ambit3’s memory. Rather, it is used to tell if you got your rest and recovery after a training session or kept on running around and should take longer to fully recover.)

Recovery Status / Tests

The new recovery tests are something else again:

They analyze heart rate variability, which is a measure of recovery state and stress, to give a physiologically based measure of how well rested one is.

This can be done in one of two ways:

Sleep Test

The “Sleep test” requires one to wear the HR strap for the whole night, which can of course be uncomfortable, depending on the person, but is less influenced by factors such as feelings of (mental) stress. It also runs over a longer analysis time, and should therefore be more of an in-depth analytical tool:

Quick Test

The “Quick test” is much faster, taking only 3 minutes, but can be influenced rather more strongly by how relaxed or stressed-out one is. This is why it is recommended that one prepare everything and then do the quick test right after waking up, before even having got out of bed.

At the very least, this is a test that should always be done under the same conditions, not whenever the fancy strikes, or it may end up measuring the effects of an earlier exercise, caffeine intake, or work stress more than one’s actual recovery state – as happened when I turned it on at the end of a day, just to show how it works:

Calibration Tests

For both tests, the first three tests done of each are used as calibration tests.

Ambit3 Calibration Test

For those, when you start the test, your Ambit3 will initially also display “calibration test”) to adjust the technology to individual status.

Recovery Test Results

Even so, the results of the recovery tests can be rather more surprising than seriously enlightening.

Case in point: Me.

During a few weeks of testing, I have hardly ever seen my sleep recovery giving a value above 50%, marking me as constantly “recovering,” even after several days without much physical exercise and with what I felt was decent sleep.

For the same days, the quick tests were only rarely below 50%…

By the way, what the recovery is meant to indicate is the following:

  • 0%-20%: “Not recovered” – Rest is recommended, not training.
  • 21%–50%: “Recovering” – Easy training is okay, but more not recommended.
  • 51%-80%: “Recovered” – Training up to hard intensity / HR zone 3 is okay.
  • 81%-100%: “Fully recovered” – All training should be okay.

In this final video, you can see the results from an(other) morning, with an overnight sleep test followed by a quick test. The combination of the two is hardly necessary and also gave the above-mentioned divergent results, but this quick test, at least, was done as it should be and gives a decent result:

With the different results from two tests following on the heels of a each other, it’s a case of “Go figure.”

The two measurements did, however, at least tend to track each other’s ups and downs relatively well, so the trends are probably interesting and useful.

Also, even with the differences, I prefer a measurement done when I want to do it (as with the Ambit3) to the implementation of the recovery check on Garmin’s fenix3, which comes after 6 minutes of an exercise, when I’m already out and doing my training. (And the fenix3 said that my recovery state was only “fair” rather than “good” only once during all my testing, on a second day of mountain training, when I could tell that I was sore anyways. All other days, it was never anything but “good.”)

Interview with Mikko Seppänen, Firstbeat

Seeing these idiosyncrasies, I asked Mikko Seppänen, physiologist at Firstbeat Technologies, to explain further…

Sleep quality testing is being explained as also being a kind of recovery testing, but is much longer-term and gets different results. So, which test is recommended for what user / use case?

Firstbeat: Sleep recovery test always gives a broader insight to recovery state because there are less external interfering factors that influence results AND the measurement period is much longer. That’s why it gives a more reliable result.

Quick recovery test is easy and quick to perform and therefore gives extra value for the cases when night measurement is not the best option:

You may sometimes feel unwilling to sleep with the heart rate belt / you have simply forgot to start measurement in the evening / something unexpected has happened during the night (baby crying) etc…

Sleep quality testing requires that one wears the HR belt while sleeping, which some people will find a disturbance itself… (Why) Is it worth doing so?

Firstbeat: Sleep measurement is definitely worth performing because it gives more reliable results (see the arguments above).

The instructions for the quick recovery test say that it should be done in the same conditions and that it also needs a baseline of tests. Does that mean there is no good use for the QRT to just quickly see how exhausting a (just-finished) training session was or if recovery level is good? (If one compares to Garmin’s fenix3, there one gets a recovery level indication after 6 minutes of a training session using the HR belt – totally different time/use.)

Firstbeat: Using the quick recovery test after an exercise makes no sense because the test is designed to measure the recovery status always in the same condition. This is to help the user to estimate his/her readiness for a coming training session.

Post-exercise condition is always different and the results are associated to fatigue à the harder the exercise, the lower the recovery score. This is because the sympathetic tone is dominating after exercise and consequently, heart rate is high and heart rate variability diminished.

The feature that Garmin has is slightly different: It evaluates “readiness” based on current performance level (measured at the beginning of the exercise)  and recent training history (how much and how hard the training in the recent days was). Both Sleep recovery and Quick recovery tests allow the user to assess his/her “readiness” and consequently plan coming workouts already beforehand.

Related to the last, why is there a need for a baseline when other HRV testing seems to be used to give an indication of stress levels from single measurements… What’s the difference in analysis?

Firstbeat: Three “calibration” tests are recommended for both sleep and quick tests and the recommendation is to perform them in a well-recovered state. This allows to  scan an “individual recovery scale”, i.e. to find what is the maximum intensity of recovery for that person. After this personal scaling, the future results are more reliable.

What would explain a (rather large) difference between sleep and quick recovery test results?

Firstbeat: One explanation for the differences between sleep and quick tests is that the sleep test is always performed in a calm state whereas interfering factors may appear before or during a quick test.  Both test results tell you how recovered you are compared with your best result. So, if you have not done any quick test in a really calm state then all results that you get are slightly overestimated.

But anyway, the quick test is more like a screening tool to, for example, observe trends in recovery state while the sleep test is more like in-depth analysis. AND it is likely that there are differences between the two test results to some degree.

Low sleep score may also be related to heart rate dynamics, e.g. when you have a high heart rate and a training effect (as measured by HRV analysis during the exercise, shown via the PTE) that accumulates quickly. So, you may have individual physiological characteristics related to a high sympathetic tone of your autonomic nervous system. If you usually train with a high intensity and have lots of work stress in addition, that may intensify sympathetic tone and decrease recovery score. Ilnesses and medications also decrease the recovery score.

For users who also use Firstbeat’s Athlete software: How could one – or Could one – use Firstbeat Athlete and the data from the Ambit3 (especially recovery and running performance) in parallel the best?

Firstbeat: Neither of the recovery tests is available in Firstbeat Athlete software. Athlete’s Training coach feature provides added value, for example.

So, if I do a recovery test and it tells me I must be completely stressed out (my average sleep quality value seems to be around 30%…), would you recommend going with that result and not doing any intense training, going with the recommendation given by Firstbeat Athlete’s Coach feature, or just going by feeling? This is where I’m wondering how to best approach all this modern tech ;)

Firstbeat: As you know, Training Coach only measures how you have trained and prescribes coming training sessions based on that. So recovery score does not influence Coach’s training prescription. Training Coach prescribes training sessions safely so that there are more easy/maintaining training sessions than improving (or harder) sessions, which usually allows you to recover adequately, thus allowing your fitness to improve. So I suggest that you could try to follow the coach for some time and see how your fitness (and sleep score) progress!

12 thoughts on “The Suunto Ambit3 Manual 6: Recovery Testing

  1. What happens if the three first calibration sessions were not done on the most ideal moment? Does it get more intelligent afterwards, i.e. does it keep on updating when you perform a test in fully recovered condition afterwards? Or can you recalibrate? Thanks !

    1. If you do a “force firmware update” for your watch (or Suunto releases a new firmware), one of the things that gets lost is the data from that calibration. So, that would be (the) one way to do a clean sweep.

      Frankly, though, I don’t know how much this calibration matters. I’ve never seen more than “40% recovered” from that – but that might also be just me; my HR data tends to be quite all over the place. (Yay, I’m not average :-p )

  2. Just wanted to say thanks for all of your work with the Suunto Ambit3 Peak. After watching your YouTube videos I went out and bought one. The videos are well presented and factual. You are doing great work, so I am happy that I have found your blog and YouTube channel. I have a couple of questions. I have watched your running escapades in Rome and Wien. They are brilliant to watch, so can you tell me what camera you are using please ? I run trail ultra marathons and would love to capture the footage you have caught on your video camera. The footage is so steady. BTW my wife and I do the same thing on New Years Eve. We go to bed early and get up early for a trail walk or run. PS Thanks also for sharing your ultra marathon displays too on the Ambit 3 Peak. I have the same ones as they for me are logical. Schöne Grüße aus den subtropischen sonnigen Australien Thomas

    1. Australische Grüsse nach Austria, schön ;)

      Funny thing. You’re the third person in what feels like a very short time to ask about the camera.

      Thing is, it’s not the camera (a Sony HDR-AZ1 action cam, by the way. For the size.). It’s that I slow down, concentrate on running steadily and holding the camera as stable as possible – and somehow, it seems to work quite alright, most of the time.

      Now if only the problem weren’t that I still feel a bit shy / unworthy :-p

      1. Thanks for your reply. I guess running on the beautiful places should be shared to encourage many more people to see what we see…. I will take a look at that action cam and thanks for your tips. It really captures the whole experience without being there. Your context on your blog and videos are great !!! I have passed your blog details onto many other who have just bought the Sunnto watches… Thanks again :) PS No need to be shy, there are only a few thousand people watching you ;)

        1. Nice way of putting it… and well, I just partly undressed for my viewers, so, there goes the shyness. Hopefully not the visitors! ;)

Feel free to contribute