You finish the hardest work of the week and the watch says the session merely maintains your fitness. Or the opposite: an easy run comes back as highly impacting. One cause you can check in five minutes is the maximum heart rate sitting in your profile. Get that number wrong and every score is off in the same direction, month after month — and if you steer by them, you keep adding load to a body that is not absorbing what it already has. If the setting turns out to be right, the other cause is below.
Download as PDF · 7 pagesAfter a workout, Garmin shows a training effect — a number from 0 to 5 with a label. It is not a measurement but a calculation, and the manual names what goes into it: "your user profile information and training history, and heart rate, duration, and intensity of your activity."
Next to the number there is also a phrase naming the main benefit of the session. The manual counts seven of them and ties them to your training load focus, printing one as its example: "Highly Impacting VO2 Max." The number and the phrase answer different questions — how hard the session hit, and what it hit.
And there are two scores, not one. Aerobic effect is how much the session develops endurance, the ability to work for a long time on oxygen. Anaerobic is how much it develops short, powerful efforts, where oxygen can't keep up: surges, climbs, the finish.
Bands and labels from the fēnix 8 manual; the right-hand column is our plain-words reading
| Score | Label in the manual | What it means |
|---|---|---|
| 0.0–0.9 | No benefit | The load is too small to change anything |
| 1.0–1.9 | Minor benefit | Endurance does not grow; work like this helps recovery, and a long version of it builds the base |
| 2.0–2.9 | Maintains your aerobic (anaerobic) fitness | Holds your current level |
| 3.0–3.9 | Impacts your aerobic (anaerobic) fitness | Moves the level rather than holding it |
| 4.0–4.9 | Highly impacts your aerobic (anaerobic) fitness | A strong developing effect; recovery deserves attention |
| 5.0 | Overreaching and potentially harmful without enough recovery time | Excessive load: it needs a long recovery, and repeated often it raises the risk of overtraining |
The manual is explicit about which work raises which score. Aerobic: steady work at moderate effort, or intervals longer than three minutes. Anaerobic: repeated intervals of 10 to 120 seconds at high intensity. They rest on different inputs, too — the aerobic score uses heart rate, the anaerobic one uses heart rate and speed (or power).
The main thing to understand about this scale: it does not rate your effort, it rates what the session does to your body relative to its current level. The same run reads as impacting for a beginner and as merely maintaining for a trained runner — the trained runner needs a bigger load for the same effect. For the same reason, 5.0 is not the target of every session: Firstbeat warns explicitly that frequent sessions scoring 3 to 5 without enough recovery raise the risk of overtraining.
To rate a session, heart rate on its own is not enough. 160 bpm is moderate work for one person and flat out for another. The method needs a reference, and the one it uses is your maximum heart rate — the highest your heart rate will go in an all-out effort.
Two things about that number. First, it is not a measure of fitness. Training barely moves it, and a high maximum does not mean you are well prepared, nor a low one the opposite. It varies between people the way shoe size does.
Second, if you never set it yourself, the watch falls back on an estimate from your age — the familiar "220 minus your age", and the manual says so outright. (On the fēnix 8 that default can later be overwritten by Auto Detection, which is on out of the box — see the checklist below.) The formula describes an average across a large group: its standard error of estimate is about 7 to 12 beats per minute, and a 2021 study of physically active people found it underestimating the maximum by a mean of 9 beats in the youngest group.
Your percentage of that maximum is how the method decides whether the work was hard or easy. It is not the only input: duration and how trained you are go into the calculation too. What ties them together is EPOC — the extra oxygen your body burns getting back to normal afterwards. The further a session has pushed the body out of balance, the higher the EPOC. You can only measure it directly in a lab; Firstbeat estimates it from heart rate, which is why the reference point matters.
That is how Firstbeat describes the method, and its own table of error sources names maximum heart rate first. Garmin's manual is less specific: it lists the profile, training history, heart rate, duration and intensity as the inputs without singling the maximum out, and the company does not disclose how closely the current firmware follows the Firstbeat method. So whenever the two differ below, we say which one we are quoting.
The method is not Garmin's own, either. The calculation grew out of the work of the Finnish company Firstbeat, whose papers are the most detailed published description of it we could find. Garmin bought the company's consumer side in June 2020, and the current fēnix 8 manual signs the training effect with its own name: "Training Effect technology is provided and supported by Garmin Human Performance Lab™." So we read how the method works from Firstbeat, and what is shown on the screen from the Garmin manual.
Take one session and look at how the watch sees it under two different profile settings. The athlete's real maximum heart rate is 185 beats; during the session the heart rate sits around 170.
One and the same session · 170 bpm · real maximum 185
| Profile says 200 | Profile says 185 — the real one |
|---|---|
| 85% — of maximum, as the watch counts it | 92% — of maximum, as it actually was |
| The work looks moderate. The score comes out below what it was worth | The work was near the limit. The score matches the load |
The session is identical. What gives it two different scores is a number in the settings.
Set it too low and everything flips. Real maximum 195, settings 175, heart rate during the session 160: the watch sees 91% instead of the true 82% and shows a score higher than the session earned.
TE is underestimated if maximal heart rate, maximal respiration rate or activity class has been set too high. TE is overestimated if maximal heart rate, maximal respiration rate or activity class has been set too low.
Firstbeat, "EPOC Based Training Effect Assessment", table 3
And it isn't one session that is off. While the wrong value sits in your profile, every new score comes out of the same shifted input: with a maximum set too high you spend months looking at scores that are too low and, steering by them, add load on top of a load your body is already failing to absorb. Once the setting is fixed, new scores land closer to the work you actually did.
Don't push the arithmetic too far, though. The percentages show the size of the error at the input, not a ready-made correction to the score: Firstbeat does not publish the conversion formula, and Garmin does not disclose how closely its firmware follows it. The conclusion to draw from the examples is not "mine should have been 3.7 instead of 2.5" but "my scores are shifted in one direction, and that is worth fixing".
Maximum heart rate is the main source of drift, but Firstbeat names four more.
Table 3 of the Firstbeat training effect paper; the right-hand column is our plain-words reading
| What is happening | Where the score goes | Why |
|---|---|---|
| Illness, including one that is only starting | up | Heart rate and breathing run above usual at the usual work |
| Heat, humidity, altitude | up | The heart has to beat more often to hold the same pace |
| A long period of high volume | down | A tired body responds less |
| A period of high intensity | up | The body reacts more sharply |
Two things follow for practice. In heat and at altitude, ease the pace — you will get the same score for less work. And falling scores after a hard block do not prove overtraining: Firstbeat makes the point separately that the main sign is a drop in performance, not the numbers on the watch.
Two caveats that are not steps. At high heart rates and during sharp accelerations a wrist sensor drops beats and misreads, so a peak from a single session is worth confirming with a chest strap before you type it into your profile. And if you want the real figure rather than a working approximation, a graded test under supervision in a lab gives it, along with a look at how your heart handles maximal work; a do-it-yourself test to exhaustion is not for everyone, and most people do not need one.
What does a training effect of 3.5 mean?
It falls in the 3.0 to 3.9 band, which the manual labels as impacting your fitness: the session moves your level rather than holding it. The full table of bands is above. Aerobic 3.5 and anaerobic 3.5 are not the same achievement — one is about endurance, the other about holding short, very hard efforts.
Is training effect the same as training status?
No, and the words make them easy to confuse. Training effect scores a single session and appears as soon as it ends. Training status describes a stretch of weeks and is built from your VO2 max., acute load and HRV status. "Maintaining" is a training status label, not a training effect one: the training effect label for the 2.0 to 2.9 band reads "Maintains your aerobic fitness".
Should I be aiming for 5.0?
No. The manual labels 5.0 as overreaching and potentially harmful without enough recovery time, and Firstbeat warns explicitly that frequent sessions scoring 3 to 5 without enough recovery raise the risk of overtraining.
I fixed my maximum heart rate. Will Garmin recalculate the sessions already recorded?
We could not confirm that it does, and the manual does not say. Treat the history as it stands and judge the change by the sessions that come after the fix.
My maximum is right and hard sessions still only maintain. What then?
Then the setting is not your answer and the other cause is: the scale measures a session against your current level, so work that once moved your fitness only holds it once you have grown into it. Give it two or three comparable sessions to be sure the score is steady, then change the training rather than the profile — a bigger or a different stimulus, not a different number in the settings.
A series of explainers, "Garmin metrics", built from primary sources. Published wording, thresholds and limits come with a link to the Garmin manual, to the Firstbeat papers or to the study quoted; our explanations, calculations and examples are marked. Fizkult is not affiliated with Garmin or Firstbeat; the names belong to their owners.
Fix the maximum and the number stops misreading the work. It still will not tell you what 3.5 means on the third hard day in a row, after a short night, on top of a block you have not absorbed yet. Reading one against the others is a coach's job, not a table's.
Fizkult reads your Garmin data itself and sends a plain-words breakdown to Telegram every morning: what your sleep, recovery and load are saying and what to pay attention to today. It builds your norm from your own history, and while the days are still too few it says so outright: “I don’t know your usual yet: counted over 9 days out of the 14 I need.” You connect the watch once; don't skip the Historical Data box — without it your norm builds only from new days, and more slowly. First week free, everything included. The bot writes in English if your Telegram is set to English; /language switches it either way.
Connect your watch in Telegram →This is a coach's read, not a medical one. If you don't feel you're recovering, see a doctor.