Athletes generally burn more energy at rest than non-athletes … but probably not for the reasons you think.
Your basal metabolic rate (BMR) tells you how much energy your body burns to just “keep the lights on” – it’s the energy used to power the basic functions of your vital organs, to accomplish sufficient protein and cell turnover to keep your tissues functioning properly, and more. If you didn’t leave your bed all day and didn’t move a muscle, your basal metabolic rate is the amount of energy you’d still burn in a day.
There’s a general belief that athletes have higher BMRs than non-athletes because they have more muscle mass due to training. And, while it’s true that athletes do have higher BMRs, differences in muscle mass are far from the primary reason for the difference.
In a previous article in this series, we discussed the determinants of your BMR. Just to recap, your BMR is determined by the tissues composing your body, and the specific metabolic rates of those different tissues. When you split your BMR out on a tissue-by-tissue basis, it becomes clear that differences in muscle mass have a relatively small impact on your overall BMR. Muscle has a tissue-specific metabolic rate of about 13.5 Calories per kilogram. So, if you gained or lost a large amount of muscle mass – say, 5 kilograms or 11 pounds – that would only increase or decrease your BMR by about 67 Calories per day. That’s not nothing, but it’s a fairly small difference in the grand scheme of things.
Most of your BMR is determined by the mass of your high-metabolic-rate organs: your brain, heart, kidneys, and liver. These tissues all have BMRs that are about 15-33 times higher than the BMR of skeletal muscles – about 200-440 Calories per kilogram, versus 13.5 for muscle. And, as we covered in a previous article, high-metabolic-rate tissue mass doesn’t scale linearly with total lean mass. In other words, larger people generally have larger hearts, livers, kidneys, and brains, but the difference is considerably smaller than the difference in total fat-free mass. If person A has twice as much fat-free mass as person B, their high-metabolic-rate organs might only be 50% larger.
Because of this, BMR per unit of fat-free mass tends to decrease as fat-free mass increases. People with around 40kg of fat-free mass typically have BMRs of about 31 Calories per kilogram of fat-free mass, whereas people with 80kg of fat-free mass typically have BMRs of about 26 Calories per kilogram of fat-free mass. As total fat-free mass increases, the ratio of low-metabolic-rate tissues (like muscle, bone, and the lean component of adipose tissue) to high-metabolic-rate tissues (like brain, heart, liver, and kidneys) tends to increase.
If you’ve been following along with this series, I’m sure you’re already familiar with all of the information up to this point in the article. But, this recap is important, because it sets the stage for discussing how athletes differ from non-athletes.
The most important factor contributing to higher BMRs in athletes
The main reason athletes have higher BMRs than non-athletes is that everything I’ve covered in this article – and most of the articles in this series up to this point – doesn’t really apply to athletes. Larger and smaller athletes burn about the same amount of energy per unit of fat-free mass.
This was most clearly demonstrated in a pair of studies from Japan. The researchers analyzed body composition and BMR in 57 male athletes in one study, and 93 female athletes in the other. In both studies, they split athletes into three groups (small, medium, and large) based on their fat-free mass. They found that BMR per unit of fat-free mass was basically the same in all three groups in both studies. Furthermore, comparing between studies, BMR per unit of fat-free mass was basically the same in the male and female athletes (which runs counter to what we observe in the general population – women tend to have higher BMRs per unit of fat-free mass in non-athletes).
A follow-up study from the same group of researchers tells us why larger athletes have the same BMR per unit of fat-free mass as smaller athletes: in athletes, most high-metabolic rate organs do scale linearly with body size. In athletes with fat-free masses ranging from about 57kg to 85kg, muscle, liver, kidney, and heart mass all scaled linearly with total fat-free mass. So, at all body sizes, each of these tissues had a consistent relative contribution to total BMR. The one exception was the brain, which didn’t scale as strongly with total fat-free mass.
In other words, large athletes with 50% more total fat-free mass than small athletes also had hearts, livers, and kidneys that were about 50% larger, which runs counter to what we observe in non-athletes. However, athletes with more fat-free mass only had slightly more brain mass than non-athletes. As a result, you’d still expect smaller athletes to have slightly higher BMRs per unit of fat-free mass than larger athletes (which is what these studies observed), but the difference is much smaller than the one that’s observed in the general population.
These results are bolstered by an earlier study by Midorikawa and colleagues. In this study, sumo wrestlers were compared to untrained controls. The sumo wrestlers had much higher BMRs, but both groups had similar BMRs per unit of fat-free mass. Again, the researchers found that the mass of most high-metabolic rate organs (the heart, liver, and kidneys) accounted for similar proportions of total fat-free mass in both groups, explaining the similarities in BMRs per unit of fat-free mass. Much like the previous study, the brain was the one exception – brain mass was pretty similar in both groups (meaning brain mass per unit of total FFM was a bit lower in the sumo wrestlers).
So, the main reason athletes buck the trend discussed in previous articles is that athletes with large amounts of fat-free mass have granular body compositions that are different from non-athletes with large amounts of fat-free mass. Non-athletes with large amounts of fat-free mass have disproportionately more low-metabolic rate fat-free tissue than people with less fat-free mass, whereas the ratio of high- to low-metabolic rate fat-free tissue remains remarkably consistent in athletes with differing amounts of total fat-free mass.
The effects of recent training
When you have your BMR measured, you’re asked to follow quite a few pre-assessment guidelines (you should be, at least). You should be in a fasted state, have no stimulants in your system, use the bathroom before the test, and avoid strenuous exercise for at least 48 hours before the test.
All of these factors are important, because they can all skew the results of your BMR test. If you’ve eaten recently, the thermic effect of feeding (the energy you burn to digest food) will artificially elevate your resting energy expenditure. Stimulants boost your resting energy expenditure slightly. Anxiety from feeling the urge to urinate can elevate your energy expenditure a bit. And … strenuous exercise can elevate your resting energy expenditure for a day or two.
The degree to which exercise elevates your BMR will depend on how long and how strenuous your workout was. You may experience no increase at all following a relatively easy, relatively low-volume workout, you may experience an increase of 100 Calories for the next day or two following a harder workout. This increase may be due to the increased energy cost of repairing muscle damage, increases in sympathetic nervous system activity, and the general biochemical cost of returning to homeostasis (metabolizing waste products, resolving inflammatory responses, converting lactate back to glucose, etc.). You can find some outlier studies suggesting that this increase can be 400+ calories, but most studies find elevations in the range of 50-150 calories.
Most of the research on this phenomenon has been conducted in untrained subjects, but there’s evidence for it in elite athletes as well. For example, cyclists competing in the Ardennes classics (~250km/150mile one-day bike races through mountainous terrain) had their BMRs measured before a race, and the morning after a race. Their average BMR before the race was about 1936 calories (already pretty high, since they weighed just 67kg, on average). The morning after the race, their BMRs were elevated to 2055 Calories.
Most of the research assessing BMR in elite athletes will note that the athletes were required to refrain from strenuous exercise for either 24 or 48 hours prior to BMR measurements. I have some level of skepticism about how many athletes actually follow through with that requirement. People lie to researchers sometimes, and athletes who are accustomed to training every day may not want to take two days off of training to participate in a study, especially if they don’t fully understand the reason they’d need to take time off in the first place. I also have some level of skepticism about whether BMR fully returns to baseline after 48 hours in athletes who train hard day-in and day-out for months or years; we observe the BMR returns to baseline within 48 hours after a single challenging workout, but I find it plausible that the elevation may last for longer if someone’s trained hard on four of the past five days. Some older research in athletes has suggested that it may take up to five days.
However, I ultimately think that those considerations are purely academic. If athletes’ “true” BMRs (i.e. their BMRs if they were fully recovered from exercise) are slightly lower than the BMRs reported in the research on the topic, I’m not sure it matters too much, because athletes spend most of their time training consistently. If you spend the vast majority of your time <48 hours removed from your last workout, you could easily make the case that the BMR elevations associated with recovery from training are just a part of your “normal” BMR. Or, on the flip side, if athletes’ BMRs have fully returned to baseline before being measured, that would mean that the research may slightly underestimate athletes’ day-to-day BMRs (which would be elevated following workouts on most days). But, since post-training elevations in BMR tend to be around 100 Calories per day, this would ultimately be a relatively small difference.
Yes, muscle too
I’d be remiss if I didn’t call this out: exercise generally increases muscle mass, and athletes in most sports have more muscle than untrained individuals. I realize that I downplayed the importance of muscle mass earlier in the article, because it’s falsely assumed to be the only factor (or at least the primary factor) explaining why athletes generally have higher BMRs, and I wanted to push back against that erroneous belief. But, it certainly is a factor. If two people are the same height and weight, and one has 5kg more muscle and 5kg less fat (on par with the body composition differences we tend to observe when comparing athletes and non-athletes), you’d expect the individual with more muscle mass and less fat mass to have a BMR that’s 40-50 Calories higher.
To be clear, that’s a very real effect. But, it pales in comparison to the other two factors discussed above when comparing two individuals who are roughly the same height and weight.
Now, if you compared a very large, very muscular 100kg athlete to a much smaller, much less muscular 60kg individual, of course the muscular athlete will have a much higher BMR. But, that’s an apples to oranges comparison, and most of that difference will be accounted for by differences in high-metabolic rate organ masses, with differences in muscle mass playing a smaller role.
I’m getting slightly ahead of myself, but let’s assume we were comparing an athlete to a non-athlete, and both weigh 75kg. The non-athlete has 20% body fat and 60kg of fat-free mass, while the athlete has 13% body fat, 5 additional kilograms of muscle mass, and 65kg of fat-free mass in total. Based on the 1991 version of the Cunningham equation (one of the best BMR equations for non-athletes), you’d predict the non-athlete to have a BMR of 1666 Calories. Based on the equation that will be developed and discussed below, you’d predict the athlete to have a BMR of 1986.5 Calories. That’s a difference of about 320 Calories. The 5kg difference in muscle mass only explains about 20% (65 Calories) of that difference.
So, how much higher are BMRs in athletes?
To answer the question posed in the section header, I performed a meta-regression. In the interest of full transparency, I didn’t do a full systematic literature search to identify every study that could have been included. But, I was able to lean on two recent studies that did do systematic literature searches: a 2023 systematic review by Martinho and colleagues, and a 2023 meta-analysis by O’Neill and colleagues. Both of these papers found all of the studies that assessed BMR in athletes and compared those measurements to pre-existing BMR prediction equations. Plenty of studies measure BMR but don’t compare those measurements to a prediction equation, so I supplemented the studies from those review papers with some Pubmed searching of my own. As mentioned, this was not a fully systematic search. I’m sure I missed several studies. But, I was targeting a total count of around 1500-2000 total athletes. Increases in statistical precision scale nonlinearly with sample sizes. Margins of error with polling data illustrate this pretty well, as you can see below. More data does continue increasing precision, but you hit a point of diminishing returns.
So, I was content to sift through several hundred Pubmed records instead of several thousand. I’m quite confident that this analysis includes enough of the studies on the topic to fairly and accurately describe the research on BMR in athletes with sufficient precision.
For a study to be included, it needed to assess both BMR and body composition in healthy adult athletes without any known medical conditions. I included the body composition requirement because the athletes in these studies tended to have remarkably homogeneous body composition. Almost all of the men were between 10-20% body fat, and almost all of the women were between 17-27% body fat, so an equation using height, weight, and age (instead of fat-free mass) wouldn’t generalize outside of those bounds anyways.
I ultimately turned up 50 groups comprising 1950 total athletes (1146 men, and 804 women) across 29 studies. They ran the gamut from Olympians to recreational lifters, and from professional cyclists to bodybuilders to sumo wrestlers. You can see the sample size, and average age, height, weight, fat-free mass, body fat percentage, and BMR of all of the included studies in the tables below.
If you’d like to dig into any of the studies for yourself, the table below has clickable links for all of the included studies:
Individual subject data wasn’t available, so I performed a meta-regression on the group mean fat-free mass and BMR values, weighted by sample sizes.
The best-fit regression lines for only male and only female athletes didn’t meaningfully differ from the regression line for athletes of both sexes (which is what we’d expect, based on similar research in non-athletes). The linear equation predicting BMR in athletes was:
BMR = 28.9 × Fat-Free Mass (kg) + 108
Alternately, if you’re an athletic man between 10-20% body fat, or an athletic woman between 17-27% body fat, you could use this equation, which was calculated via multiple regression (again, this should not be expected to generalize very far outside of those body composition bounds).
BMR = 21.2 × Weight (kg) + 6.7 × Height (cm) – 2.9 × age – 95 × sex* – 764
*0 if male, 1 if female
Evaluating the results
When comparing this equation to the 1991 Cunningham equation (the best equation for non-athletes that predicts BMR from fat-free mass), you can see that the two equations produce pretty similar BMR predictions at relatively low levels of fat-free mass, but they diverge as fat-free mass increases. At 40kg of fat-free mass, the predictions only differ by 2.4%. At 85kg of fat-free mass, the MacroFactor equation generates a prediction that’s over 16% higher.
This is exactly what you should expect, given the information covered thus far in this series. In non-athletes, high-metabolic rate tissue comprises a smaller and smaller percentage of total fat-free mass as total fat-free mass increases. In athletes, on the other hand, heart, kidney, and liver mass scale proportionally with total fat-free mass, and only relative brain mass (as a percentage of total fat-free mass) decreases as total fat-free mass increases. So, you should anticipate that BMR per unit of fat-free mass should decrease much less in athletes as total fat-free mass increases.
Digging a layer deeper, I roughly categorized the athletes in these studies as “elite” or “non-elite.” I’ll readily admit that this is a somewhat subjective characterization. For example, are Division III collegiate athletes “elite” athletes? Compared to pros, absolutely not. Compared to the vast majority of humans, absolutely. When in doubt, I typically gave them the nod if they competed at a level of sport that most people would be unable to reach (even if that was a sub-professional level), or if the study specifically noted that they trained for more than an average of 2 hours per day (14+ hours per week).
Ultimately, it appears that athletes have higher BMRs than non-athletes regardless of competitive achievement. The BMRs of the elite and non-elite athletes in these studies weren’t discernibly different.
In fact, this isn’t a particularly surprising finding. A BMR equation that tends to perform pretty well in athletes is the 1980 version of the Cunningham equation (BMR = 22 × FFM + 500). It produces comprehensively higher BMR estimates than the 1991 Cunningham equations (BMR = 21.6 × FFM + 370), and tends to overestimate BMR in the general population. However, it was developed from data collected from a general population sample back in 1918, when typical levels of physical activity were considerably higher, but rates of structured sport participation were considerably lower. Another equation that performs well for high-level competitive athletes is the ten Haaf equation (BMR = 22.8 × FFM + 484), despite the fact that it was developed from a sample of recreational athletes. So, it appears that people who are generally active and exercise regularly tend to have higher BMRs than people who have a more sedentary lifestyle, but you don’t necessarily need to train like a professional athlete to reap the benefits.
Next, let’s turn our attention to the types of sports these athletes participated in. I categorized all of the studies as including strength/power athletes, endurance athletes, or “mixed” (for the most part, these were studies that included athletes from a variety of different sports).
In general, athletes in both strength/power sports and endurance sports tended to have slightly higher BMRs than athletes in sports with mixed demands, or athletes in studies that included a variety of different sports. For what it’s worth, I personally wouldn’t read too far into that – most of the studies were categorized as “mixed,” and one study in strength/power athletes or endurance athletes that finds a particularly low average BMR could flip the trend. Furthermore, I should point out that the research on endurance athletes firmly counters the common (but completely spurious) claim that “cardio crashes your metabolism.” It quite clearly doesn’t.
Aging
In our previous article about the impact of age of BMR, I hinted that the next article in this series might have suggestions about how we can stymie age-related declines in BMR.
By now, it should be clear: exercise and staying active.
Unfortunately, only one study included in this analysis used a sample of subjects with an average age north of 40. However, that study by Frings-Meuthen and colleagues is perfectly at home in the rest of this body of research. These athletes were in their mid-50s, on average, and their BMRs were only slightly below the general trendline.
Furthermore, after adjusting for height, fat-free mass, fat mass, and sex, the researchers found that each year of age was associated with a decrease in BMR of just 0.6 Calories. The athletes in this study ranged from 35 to 84 years old. As we discussed in the last article, BMR corrected for similar factors tends to decrease at a rate of about 2 Calories per year up to 60 years old, and 4-5 Calories per year thereafter. So, not only did they have BMRs that were comparable to younger athletes – this study also suggests that age-related declines in BMR per unit of fat-free mass occur at about one-third to one-eighth the usual rate in Masters athletes. And of course, that’s before even mentioning how regular exercise can help you build or maintain muscle throughout your life, countering the typical age-related losses in strength and muscle mass.
I should note that the researchers suggested that the subjects in their study may have had artificially elevated BMRs, because BMR was assessed in a room that was 27.5℃ (which is warmer than you’d typically keep a room for metabolic testing). However, assuming the subjects were wearing light clothing, 27.5℃ is still within the thermoneutral zone (the range of temperatures where your body can easily regulate temperature without the need to expend additional energy). Elevations in resting energy expenditure only start to reliably occur at considerably higher ambient temperatures. Furthermore, other research in both men and women has also found that older athletes have BMRs (adjusted for body composition) that are more similar to younger athletes than to sedentary older adults.
Differences vs. Changes
Before wrapping up, I’d like to address the topic of changing your metabolic rate in response to exercise. It’s possible that everything in this article has been fool’s gold, after all. “Athletes have higher BMRs” doesn’t necessarily imply, “If you start exercising more, that will increase your BMR.” It may feel like an obvious leap to make, but to apply a bit of skepticism, I’m sure you’d feel differently about the statement, “Professional basketball players have an average height of 6’6”, so playing basketball will make you taller.”
It’s possible that people who are athletes were able to become athletes because they were people who naturally had larger hearts to pump more blood, larger livers to metabolize the metabolic byproducts of exercise, and larger kidneys to flush metabolic waste out of their systems.
So, does exercise actually increase your BMR?
Yes.
A 2020 meta-analysis by MacKenzie-Shalders and colleagues analyzed the research on changes in BMR after exercise interventions. Most of the included studies lasted for around 12 weeks, and subjects experienced an average increase in BMR of about 80 Calories. As discussed above, that’s a much larger increase than you could reasonably expect due to increases in muscle mass – for an 80 Calorie increase in BMR, you’d need to build about 6kg of muscle in 12 weeks (a 1.5kg increase in fat-free mass is more typical with resistance training, and less with aerobic training). Furthermore, research also suggests that athletes’ BMRs decrease by about 5-10% if they stop training, so we have evidence of adaptability going both directions.
Furthermore, it’s well-known that the heart increases in size (in a benign or beneficial manner) as an adaptation to exercise. I’m not aware of much direct evidence of (benign or beneficial) kidney and liver growth with long-term exercise training, but I find it plausible, and there is a bit of human evidence for the phenomenon. For instance, elevated biomarkers in liver function tests following an intense exercise bout may suggest that exercise stresses the liver (again, in a benign or beneficial way; exercise is good for the liver) in a way that would lead to adaptations over time.
Finally, as previously discussed, we don’t just observe higher BMRs in elite athletes (who may have physiological gifts that aren’t available to most of us). We also observe similarly elevated BMRs in recreational lifters, totally normal CrossFit participants, and recreational athletes.
So, I can’t confidently say that exercise will fully close the gap between your BMR and the BMR of a professional athlete. But, I can confidently say that exercise does at least narrow the gap, both because it helps you build more lean tissue, and because it likely increases your BMR per unit of fat-free mass.
Wrapping it up
At this point, our BMR series is winding down. We’ve discussed the best BMR prediction equations, covered determinants of BMR, addressed how age and sex impact BMR, and now we’ve gone through the research about BMR in athletes (and, by extension, how exercise impacts your BMR). The next articles will cover weight loss and weight gain. Then, we’ll wrap up the main part of this series by using everything we’ve learned to improve upon the (current) best BMR equations. You can also try our BMR calculator, which incorporates all of the information covered in this series, in order to estimate your BMR as accurately as possible.
