If your Garmin watch has been flashing the dreaded “overtraining” status during your marathon build, you’re not alone. Runners across the world are questioning the validity of their devices’ recovery recommendations, especially as training intensity ramps up. The issue often isn’t that you’re doing too much. It’s that the way your watch interprets threshold heart rate zones is fundamentally limited, leading to misleading stress signals that can derail smart training decisions.
Let’s unpack the data behind why this happens and what a real-world case study reveals about the gap between algorithmic feedback and actual physiological readiness.
The Problem with Default Threshold Heart Rate Zones
Most Garmin devices estimate your threshold heart rate using a combination of age-based formulas, VO2 max modeling, and your observed performance data. The default zone, usually Zone 4, is meant to represent the heart rate at which lactate begins accumulating in the blood faster than it can be cleared. It’s supposed to mark the boundary between sustainable aerobic work and the beginnings of anaerobic stress.
But here’s the catch: threshold HR is not static. It fluctuates with fatigue, hydration, heat, sleep quality, altitude, and even caffeine intake. Your watch doesn’t recalculate it in real time based on those daily variables. Instead, it uses a rolling estimate that may lag behind your true current threshold by days or even weeks.
So when you complete a long run or a hard workout, Garmin compares your effort to a threshold that may no longer reflect reality. If your real threshold is actually higher than the watch thinks, your device will misclassify a normal aerobic effort as high stress. If your real threshold is lower because of accumulated fatigue, the device will tell you that you weren’t working hard enough, encouraging you to push on a day when rest would be smarter.
A Real-World Case Study: The Sub-3 Hour Build
Consider a 34-year-old runner training for a sub-3:00 marathon over 16 weeks. During week 10, his Garmin began showing a rising “training load” ratio, eventually crossing into the “overtraining” red zone by week 11. He felt physically capable. His pace on tempo runs was consistent. His sleep scores were above 75 most nights.
Confused, he dug into the raw data. His watch was still using a threshold HR of 168 bpm, estimated back in week 3 after a VO2 max test. But after a heat wave in week 7 and a high-volume block, his real lactate threshold HR, verified through a blood lactate test, had drifted to 162 bpm. His watch was classifying his easy runs as moderate effort because his heart rate was spending more time in what the algorithm thought was “threshold territory.”
Meanwhile, his actual easy pace was producing a heart rate well below his true threshold. The problem wasn’t overtraining. It was a misaligned reference point that made every run look harder than it was.
Why the Training Load Algorithm Compounds the Error
Garmin’s training load metric uses something called EPOC, or excess post-exercise oxygen consumption, to estimate how much physiological stress each workout creates. The system then compares that stress against your recent training history to generate an acute-to-chronic ratio. When that ratio climbs above 1.5, the watch flags you as at risk for overtraining.
The math relies on accurate HR zone modeling. If the zones are off, EPOC is off. If EPOC is off, the load ratio is off. One small error at the foundation of the model cascades into a false overtraining alert.
In the case study above, the runner’s “moderate” easy runs were being scored as high-intensity efforts because his heart rate sat in what Garmin believed was Zone 4. Each run generated an inflated EPOC score. Over a week, the cumulative effect pushed his training load ratio past the warning threshold, even though he was running fewer total miles than his peak week and recovering well.
What Marathon-Build Runners Should Track Instead
If you want a clearer picture of your actual recovery status, supplement your Garmin data with metrics it doesn’t prioritize:
- Resting heart rate trends taken manually each morning. A sustained rise of 5+ beats above your baseline often signals accumulating fatigue before the algorithm catches up.
- Heart rate variability (HRV) using a chest strap or morning measurement app. Trends matter more than single numbers.
- Subjective wellness logging: mood, soreness, motivation, and sleep quality. These correlate strongly with overtraining risk in research studies.
- Performance drift: if your easy pace is getting slower at the same perceived effort, that’s a clearer signal than any training load number.
How to Calibrate Your Garmin for a Marathon Block
A few practical adjustments can reduce false overtraining flags without abandoning the data entirely:
- Re-run a threshold test every 6 to 8 weeks during a build phase, especially after a heat exposure period or a peak training week.
- Disable auto-detected max HR changes unless you’ve done a verified max effort test. Spikes from hard intervals can artificially shift your zones upward.
- Use a chest strap for threshold workouts. Wrist optical sensors are still less accurate during high-intensity efforts, which can throw off zone distribution.
- Set manual lactate threshold HR if you have lab data or recent field test results. This prevents the watch from drifting based on outlier sessions.
- Watch the trend, not the daily status. A single “overtraining” day is noise. A week of climbing load, rising morning HR, and falling pace is a real signal.
The Bigger Lesson: Algorithms Are Tools, Not Coaches
Your Garmin is remarkable at collecting data. It is not remarkable at interpreting that data the way an experienced coach or a sports physiologist would. The training load metric was designed for general fitness populations, not for the specific demands of a marathon build where high volume and frequent intensity are expected and necessary.
When your watch says you’re overtraining, the right first question isn’t “should I take a rest day?” It’s “what is this algorithm actually measuring, and is it measuring it correctly?” Treating the alert as data rather than a verdict gives you the freedom to make better decisions about your actual recovery needs.
Conclusion
Overtraining alerts from Garmin during marathon training are often more about misaligned threshold zones than true physiological breakdown. Real recovery assessment comes from combining multiple data sources, recalibrating your watch regularly, and trusting your subjective experience alongside the numbers. The goal isn’t to silence the alerts. It’s to understand what they’re really telling you.
