What Is a Normal Oxygen Saturation During Exercise?

For most healthy people, blood oxygen saturation stays between about 95% and 99% during exercise, and a reading that remains at or above 95% is generally considered normal. Small dips of a percentage point or two during intense effort are common and harmless, but drops below roughly 88% to 90% raise clinical red flags. The line between a normal dip and a worrying one depends on your fitness level, altitude, underlying lung health, and even how your sensor is measuring you.

What Happens to Your Blood Oxygen During a Workout

At rest, oxygen saturation (SpO2) for a healthy person typically sits between 96% and 100%. When you start exercising, your muscles demand more oxygen, your heart pumps faster, and your breathing rate climbs. In a well-functioning respiratory system, ventilation keeps pace with demand. The lungs load oxygen onto red blood cells efficiently enough that the percentage of hemoglobin carrying oxygen barely changes, even when your body’s total oxygen consumption has multiplied several-fold.

That said, oxygen saturation doesn’t stay perfectly flat. During very hard exercise, it’s common for SpO2 to nudge down by one or two percentage points. The most noticeable shifts tend to occur above about 70% of your maximum effort, where the body’s ventilatory system is working near its ceiling. A study examining both arm and leg exercise found that the greatest changes in saturation happened at work rates above 70% of maximum oxygen uptake, and that the pattern was similar regardless of whether the exercise involved the upper or lower body.¹

For a typical gym-goer or recreational runner, finishing a hard interval with an SpO2 of 94% or 95% is nothing to worry about. It usually rebounds within seconds to a minute of easing up.

When Fit Athletes Drop Lower Than Expected

One of the more counterintuitive findings in exercise physiology is that the fittest people are sometimes the ones whose oxygen levels fall the most. Exercise-induced arterial hypoxemia (EIAH) is a recognized phenomenon in highly trained endurance athletes. In one study of trained athletes and untrained subjects, none of the untrained or moderately trained individuals experienced EIAH, defined as saturation dropping to 91% or below. But about half of the highly trained endurance athletes did, and the effect was highly reproducible across repeat tests.²

The reason seems to be a mismatch between how much oxygen these athletes can consume and how quickly their lungs can deliver it. Research on elite endurance athletes found that all of them showed a significant drop in arterial oxygen pressure from rest up to 80% of maximum, along with a rise in carbon dioxide, consistent with the lungs not ventilating fast enough relative to the body’s demand. The athletes who trained the most hours per week showed a continuous oxygen decline all the way to peak effort, driven in part by a widening gap between the oxygen levels in lung air and in the blood leaving the lungs.³

So if you’re an elite cyclist or competitive rower and your coach straps an oximeter on your finger during a maximal test, seeing readings in the low 90s or even high 80s doesn’t necessarily mean something is wrong with your lungs. It may reflect that your cardiovascular engine has outgrown your respiratory system’s capacity to keep up. This phenomenon has practical implications for competitive athletes: inspiratory muscle training, which strengthens the breathing muscles, has been shown to reduce EIAH and improve rowing performance, with a particularly strong effect in female rowers.⁴

Altitude Changes the Rules

Everything about “normal” oxygen saturation shifts when you exercise at elevation. At sea level, resting SpO2 runs in the high 90s. At high altitude, the thinner air means less oxygen pressure pushing into the blood, so even resting saturation falls substantially. In one study at 4,559 meters (roughly 15,000 feet), median resting SpO2 was about 87%, and at peak exercise it dropped to around 82.5%, compared with about 97% at peak exercise at sea level.⁵

The decline is not linear, and it gets more complicated the higher you go. At 4,300 meters, the drop in maximum oxygen uptake roughly tracks the drop in peak saturation. But at 5,260 meters, maximum oxygen uptake plummets far more than saturation does, in part because the heart’s maximum output also falls sharply.⁶ Acclimatization helps over days to weeks, but even well-acclimatized mountaineers work with saturation values that would trigger alarm bells in a hospital at sea level.

This matters for hikers, skiers, and anyone who exercises above about 2,500 meters. If you’ve just arrived at a mountain resort and your fitness tracker reads 90% while you’re jogging, the altitude is the most likely explanation, not a heart or lung problem. But persistently low readings, symptoms like confusion or extreme breathlessness, or numbers that don’t recover with rest warrant medical attention regardless of elevation.

Why Your Pulse Oximeter Might Be Lying

Most people check their oxygen saturation with a fingertip pulse oximeter or a smartwatch. These devices work by shining light through tissue and estimating how much hemoglobin is carrying oxygen versus how much isn’t. At rest, they’re reasonably accurate. During exercise, accuracy drops, sometimes a lot.

Finger-based sensors face two problems when you’re moving hard: motion artifact from jostling, and reduced blood flow to the fingertips as the body redirects circulation to working muscles. One validation study found that finger sensors during exercise had low precision and a consistent tendency to read about 2% lower than the true value, with individual readings wandering by several percentage points in either direction.⁷ That means a reading of 92% during a hard run might really be 94%, or it might be 90%. The error band is wide enough to make single readings unreliable for clinical decisions during vigorous movement.

Forehead and earlobe sensors tend to perform a bit better during exercise because those sites retain more blood flow, but they are less convenient and less commonly available to consumers.

Smartwatches Add Another Layer of Uncertainty

Wrist-based SpO2 sensors, like those built into many smartwatches, face all the problems of fingertip sensors plus extra challenges. The wrist has more ambient light leakage, more motion, and the sensor sits further from major arteries. A study comparing smartwatch readings from an Apple Watch and a Samsung Galaxy Watch to arterial blood gas measurements in hospitalized patients found that the Apple Watch had a sensitivity of only about 35% for detecting low oxygen states, meaning it missed roughly two out of three cases of true desaturation. Overall accuracy was about 85% for the Apple Watch and 79% for the Samsung device, but the Apple Watch’s very low sensitivity for hypoxemia is concerning for anyone relying on it as a safety monitor.⁸

In COPD patients specifically, a comparison of smartwatch SpO2 to arterial blood gas found only a moderate correlation, with the watch tending to read higher than the actual blood oxygen level. The average error was about 1.8% too high, but individual readings could be off by as much as 7 percentage points in either direction.⁹

The practical takeaway: a smartwatch SpO2 reading during exercise is best treated as a rough trend indicator, not a precise measurement. If you see a consistent pattern of low readings across multiple workouts, that’s worth discussing with a doctor. A single alarming number during a hard effort is more likely sensor error than a real emergency.

Lung Disease and Exercise Desaturation

Where exercise-related oxygen drops matter most clinically is in people with chronic lung diseases. In conditions like COPD and interstitial lung disease (ILD), the lungs’ ability to transfer oxygen into the blood is impaired, and exercise amplifies the shortfall. Clinically, exercise-induced desaturation (EID) is typically defined as a drop of 4% or more from resting SpO2, or a nadir that falls to 88% or below.¹⁰

Researchers often break this into severity tiers. In one study of COPD patients, mild EID was defined as a drop of 4% or more with the lowest SpO2 still at or above 90%, while severe EID meant a drop of 4% or more with the nadir falling to 90% or below.¹¹ These categories are clinically useful because patients with severe desaturation tend to have worse quality of life and more limited daily function.

Idiopathic pulmonary fibrosis (IPF), a condition in which lung tissue progressively scars, tends to produce even more dramatic desaturation during exercise than other forms of interstitial lung disease. One study found that patients with IPF had a significantly lower nadir SpO2 during a six-minute walk test compared with patients who had other types of ILD, averaging roughly 86.5% versus 88.7%.¹² That may sound like a small difference, but at those levels every percentage point matters. Some IPF patients drop below 85% during exercise, reaching a range where supplemental oxygen becomes necessary to exercise safely.¹³ Simple functional tests, including a one-minute sit-to-stand test, can detect this kind of desaturation in pulmonary fibrosis patients nearly as well as a formal six-minute walk test, making screening more accessible in clinical settings.¹⁴

Supplemental Oxygen During Exercise Training

For people with chronic lung disease who desaturate during activity, supplemental oxygen during exercise is sometimes prescribed not just as a safety measure but as a way to train harder. In COPD patients who were not hypoxemic at rest, breathing supplemental oxygen during high-intensity exercise training allowed them to push their training workloads higher. By the end of the program, the oxygen-trained group was working at about 62 watts on average versus 52 watts for the air-trained group, and their endurance in constant work-rate tests improved more, with gains of about 14.5 minutes compared with 10.5 minutes.¹⁵

The picture is more nuanced in pulmonary rehabilitation programs. A trial of supplemental oxygen during rehab for COPD patients with exercise-related low oxygen found that the oxygen group reported less breathlessness after the program, but other outcomes like walking distance, overall quality of life, and anxiety or depression scores didn’t differ from the group that trained breathing regular air.¹⁶ The benefit seems to be most clear for enabling higher training intensity rather than producing across-the-board health improvements.

How Age Affects Muscle Oxygen During Exercise

Blood oxygen saturation and muscle oxygen saturation are related but different measurements. SpO2 tells you how much oxygen the blood is carrying. Muscle oxygen saturation (SmO2), measured by near-infrared spectroscopy sensors placed on the skin over a muscle, tells you how much oxygen the muscle tissue itself is using versus storing. In older adults, the muscle side of the equation changes dramatically.

In people over 65, resting muscle oxygen saturation is reduced by about 38% compared with younger adults. During submaximal exercise, the gap narrows somewhat but persists at about 24% lower. During maximal exercise, older muscles are wrung out to an even greater degree, with SmO2 roughly 59% lower than in younger exercisers. Recovery is also slower: it takes more than 50% longer for muscle oxygen to return to baseline after intense effort.¹⁷

These changes reflect age-related declines in capillary density, mitochondrial function, and the ability of small blood vessels to dilate. They don’t necessarily show up on a standard fingertip pulse oximeter, which only reads arterial blood oxygen. An older person can have a perfectly normal SpO2 of 96% during exercise while their muscles are extracting oxygen far less efficiently than a younger person’s. This gap helps explain why the same SpO2 reading can correspond to very different exercise tolerance at different ages.

Children and Exercise-Induced Drops

EIAH isn’t limited to adults. A study of 24 trained prepubescent children, averaging about 10 years old and doing roughly 10 hours of weekly physical activity, found that about a third of them developed a drop of at least 4% in oxygen saturation during maximal cycling. The children who desaturated tended to have lower lung volumes relative to their oxygen demand and a lower breathing reserve at peak exercise. Their maximal oxygen uptake was actually modest (about 42 mL/min/kg), suggesting that it doesn’t take an exceptionally high fitness level for a child’s growing lungs to fall behind their muscles’ oxygen appetite.¹⁸

This finding is more of an academic curiosity than a clinical concern in most cases. The children in the study were healthy and athletic. But it’s worth knowing for parents and pediatricians who might spot lower-than-expected oximeter readings in a young athlete and wonder whether something is wrong. In a trained, otherwise healthy child, a transient dip during hard exercise is likely the same benign mismatch seen in adult athletes, scaled down to a smaller body.

How Recovery Affects Muscle Oxygen Rebound

After you stop exercising, oxygen saturation in both blood and muscle tissue rebounds. Blood SpO2 in healthy people typically returns to resting levels within a minute or two of stopping. Muscle oxygen recovery is slower and more variable, and it’s an area where researchers are actively exploring interventions.

Compression garments, for instance, have been shown to speed the recovery of muscle oxygen saturation in the upper limbs. In one study, wearing a compression sleeve reduced the half-recovery time for muscle oxygen from about 10 seconds to about 8.5 seconds, and the time constant for full recovery dropped roughly in half. The compressed muscles also overshot their resting oxygen levels by a larger margin after exercise, suggesting enhanced reperfusion.¹⁹ Whether faster muscle oxygen recovery translates to meaningful performance gains or reduced soreness is still an open question, but the physiological effect itself appears real and measurable.