Garmin displays a real-time stress level from 0 to 100. Oura calculates “daytime stress” and resilience metrics. For Whoop, it’s the stress monitor; for Fitbit, a “stress management score.” However it’s branded, some version of a “stress score” has become ubiquitous across smartwatches and wearables. This number is marketed as a window into our internal emotional state, turned into quantified proof of how our day is really going. The only issue: these numbers aren’t all that accurate.
What your “stress score” actually tells you
The scores lighting up our wrists aren’t measuring what most of us think they’re measuring. When you check your smartwatch and see that your stress level spiked, you might assume the device somehow detected your anxiety about some direct stimulus, like a difficult conversation or frustrating traffic. But that’s not totally accurate.
Sure, your watch might have detected physiological arousal—changes in your heart rate variability, skin conductance, or movement patterns. And while those signals do tell us something real about the nervous system, they don’t really tell us about stress in the psychological sense you actually care about.
“Part of the discrepancy can be explained by different definitions of how stress is conceptualized,” says Eiko Fried, who co-authored a 2025 study that found smartwatch stress measures did not align with self-reported stress scores for most individuals. The way most people understand the term “stressed”—as in “I was really stressed today!“—isn’t the way Garmin defines its stress score, which measures physiological stress. So, your watch is not necessarily telling you how stressed you feel, just how your nervous system is behaving. “Such elevated activity can come from various sources,” says Fried, “including many we would not typically consider a stressful experience.”
Physiological arousal shows up in response to all kinds of experiences that have nothing to do with distress. “What most smartwatches call a ‘stress score’ isn’t stress itself,” says Erwin van den Burg, a physiologist who specializes in the biology of stress. “It’s usually based on indirect physiological signals like heart rate variability, skin conductance, or movement patterns. Those signals tell us something about arousal in the nervous system, but arousal can come from many sources—physical activity, excitement, caffeine, poor sleep, illness, or emotional engagement—not just psychological stress.”
The oversimplification becomes even more problematic when we consider that most stress algorithms fail to account for sex-specific physiology, particularly the menstrual cycle. Because hormonal fluctuations can meaningfully alter heart rate, heart rate variability, and temperature, “a perfectly healthy physiological shift can be interpreted by a wearable as ‘high stress,'” says Emile Radyte, CEO at Samphire Neuroscience. This means women are more likely to receive misleading stress alerts for standard human biology, which can be confusing at best and anxiety-provoking at worst.
Can you trust your “stress score” at all?
Even setting aside the definition problem and the sex-bias issue, there’s a basic question of measurement accuracy.
“When you have problems with your heart, your cardiologist may ask you to wear a chest-worn device for a few days to monitor your heart rate and heart rate variability. This is a highly accurate medical-grade device,” Fried says. “Your doctor will not ask you to wear a smartwatch, because there are many issues that make wrist-worn measurement less reliable. This affects in particular heart rate variability, for which we need highly accurate measurements.”
What do you think so far?
Heart rate variability is the cornerstone of most smartwatch stress scores, yet wrist-worn devices struggle to measure it with the precision required for medical-grade insights. The data isn’t worthless, but it’s noisy, and building definitive claims about internal states on top of noisy data is, well…scientifically dubious.
So is your wearable useless? Of course not. My critique here isn’t that wearables have no value—it’s just that the value they provide is being misrepresented. Your smartwatch’s “stress score” claims to tell you far more than the science really supports. And in some cases, a less-than-ideal score may even increase stress, rather than help people understand what their body is responding to. The great irony of the wellness industry persists.
The bottom line
The way you think about “stress” doesn’t translate to a single biological state, let alone one that can be captured by number or “score.” Your watch simply detects signs of arousal in your nervous system, which could mean almost anything.
This distinction doesn’t make the data useless, but it should make you a more informed consumer. It’d be nice if companies could stop using the word “stress” for what they’re actually measuring—perhaps “physiological arousal” or “autonomic nervous system activity,” which would be more accurate, but less marketable, so I’m not holding my breath. (Although, if I did, I’m sure my stress score would skyrocket.)
A device marketed to help you manage stress may actually create more of it by generating anxiety-inducing alerts about normal physiological variation that it misinterprets as distress. The sooner we’re honest about that gap, the sooner these devices can actually help us, rather than selling us a quantified illusion of self-knowledge they don’t really have.
