Your watch says 28 and “Time to slow down,” but your legs feel fine. Don’t cancel the session yet. A low Garmin score can mean you trained hard, or it can mean you recovered badly, and those two need opposite responses.
This post shows how to tell them apart in about 30 seconds.
What Garmin Training Readiness Measures
Garmin’s manual lists exactly six inputs: last night’s sleep score, recovery time, HRV status, acute load, sleep history (last 3 nights) and stress history (last 3 days). The score updates through the day as new data arrives.
HRV status compares your heart rate variability to your own normal range. Higher variability usually signals a rested nervous system. Acute load is your training stress from the last 7 days, with recent days counting most.
In short: sleep, stress, HRV and recent training all go into one number.
Think of it like the check-engine light in a car. It tells you something is off. It doesn’t tell you whether you’re low on oil or just ran the engine hard. You still have to look under the hood.
What Garmin Doesn’t Publish
Garmin doesn’t publish input weights. It doesn’t say how missing data is handled. We also found no published accuracy figure for the combined score. Some third-party sites list Body Battery as an input, but Garmin’s manual doesn’t.
We also won’t quote a number for how long Garmin’s HRV baseline takes to form. Sources disagree, and Garmin’s docs are the place to check.
That’s not an accusation. It means the score is a useful flag, not a measurement you can audit.
When the Score Is Wrong About Your Body
One reviewer at the5krunner reported a readiness of 1 (“Poor”) while HRV status showed High. A deliberate overload block had pushed acute load so high that load alone dragged the score down. That’s one writer’s account, and the same article argues the inputs aren’t additive. Treat it as opinion, not proof.
But the pattern is easy to understand. Hard training is supposed to raise load. If load alone sinks the score, the number is describing your plan, not your recovery.
A blended score can hide its own cause. Your job is to find it.
There’s a second weak spot: sleep tracking. A reader in the same article reported two identical watches on different wrists. They gave sleep totals more than an hour apart and sleep scores 10 to 15 points apart. That’s anecdotal, not a study. Still, sleep score is one of the six inputs. A noisy night can nudge the whole number.
If your sleep score doesn’t match how you felt, discount it. Your own notes on mood, soreness and motivation are data too. Write them next to the score for a week. See which one predicts your sessions better.
A Decision Rule for Every Band
Before you act on the band, check two things. First, is HRV status in range? Second, did you sleep badly for more than one night? If you want the case for trusting the trend over any single morning, read why the 7-day HRV trend beats the daily score.
| Garmin band | HRV status | Sleep | Likely driver | What to do |
|---|---|---|---|---|
| Prime / High (75-100) | Any | Any | Fresh | Do the planned session. Hit the key workout. |
| Moderate (50-74) | In range | Fine | Normal training load | Keep the plan. |
| Low (25-49) | In range | Fine | Load-driven | Keep the plan. Cut the extras, not the key session. |
| Low (25-49) | Suppressed | Poor | Recovery-driven | Swap to easy or rest. |
| Poor (1-24) | In range | Fine | Heavy block | Keep easy work. Watch the trend for 2-3 days. |
| Poor (1-24) | Suppressed | Poor | Recovery-driven | Back off. Rest or short easy session. |
Translation: the band tells you to look. HRV and sleep tell you what you’re looking at.
What the Research Says (and Doesn’t)
We found no peer-reviewed validation of Garmin’s composite. The nearest evidence is on HRV-guided training in general, and it’s modest.
| Study | Design | Result |
|---|---|---|
| Düking 2021 | 8 studies, 198 participants, wearable HRV | Small gain on submaximal markers (g=0.296, p=0.028). Performance not clearly better (g=0.079, p=0.597) |
| Granero-Gallegos 2020 | 6 RCTs, 195 athletes | VO2max effect 0.402 for HRV-guided vs 0.215 for control. Most studies had high risk of bias |
| Nuuttila 2022 | 30 recreational runners, about 15 weeks | 10K time improved 6.2% vs 2.9% with a fixed plan (P=0.002) |
Translation: guiding training by recovery data helps a little. It shows up more in how easy a pace feels than in race results.
One detail matters. Nuuttila’s runners didn’t follow a single number. Their load was adjusted from nocturnal HRV, perceived recovery and a heart-rate-to-speed index. That’s several signals, checked twice a week. It’s closer to the decision table above than to obeying one score.
What a Form Score Measures Instead
Form is fitness minus fatigue. Both come from your training log, not from sleep or stress. Fitness builds slowly and fades slowly. Fatigue builds fast and fades fast. Here’s how fitness, fatigue, and form are calculated.
It rests on the fitness-fatigue model reviewed by Imbach and colleagues in 2022. Load is the only input. The review also notes there’s no consensus model, and results shift depending on how you quantify load.
In plain English: Form tells you how much you’ve been training. It can’t tell you how well you slept.
Be careful with injury claims. A popular 0.8-1.3 workload ratio zone, with risk flagged above 1.5, is contested. Critics cite arbitrary windows and the fact that the ratio isn’t a mechanical load measure. So treat Form as load context, not an injury predictor.
Putting Both to Work: Dana’s Week
Take Dana, 41, training for her first half marathon off 35 miles a week. On Wednesday her watch showed readiness 31 (“Low”). HRV status was in range. She’d slept seven hours.
Under a one-number rule she’d have skipped her Thursday tempo. Under the table she kept it, since the low score was load-driven. She ran it fine.
Two weeks later, readiness read 33, but HRV was suppressed and she’d slept under five hours twice. Same band, different cause. She swapped the tempo for an easy 40 minutes.
Same score. Opposite calls. That’s the point.
Why the Number Should Change the Workout
Garmin Training Readiness describes your state. It doesn’t change tomorrow’s session. You do the thinking.
AthleteOS reads sleep and HRV from Garmin, Apple Health or Health Connect, plus your training load. Form feeds the plan engine, which can scale or swap a session and log the reason in the Adaptation Log. Strava data isn’t used by the AI.
Our Garmin Connect+ comparison covers pricing. For the HRV measurement question, see HRV4Training vs AthleteOS. You can start a 14-day trial and see how it handles your own data.
Quick Checklist
- Read the band, then check HRV status.
- Check the last 2-3 nights of sleep.
- Low plus in-range HRV: keep the key session.
- Low plus suppressed HRV and poor sleep: back off.
- Trust the trend over one morning.