Skip to content

There are no fat sprinters

This post is by Phil.

A little over three years ago I wrote a post about exercise and weight loss in which I described losing a fair amount of weight due to (I believe) an exercise regime, with no effort to change my diet; this contradicted the prediction of studies that had recently been released. The comment thread on that post is quite interesting: a lot of people had had similar experiences — losing weight, or keeping it off, with an exercise program that includes very short periods of exercise at maximal intensity — while other people expressed some skepticism about my claims. Some commenters said that I risked injury; others said it was too early to judge anything because my weight loss might not last.

The people who predicted injury were right: running the curve during a 200m sprint a month or two after that post, I strained my Achilles tendon. Nothing really serious, but it did keep me off the track for a couple of months, and rather than go back to sprinting I switched to cycling, which is much easier on my knees and ankles (which are pretty messed up after years of field sports). At this point it’s fair to call me an avid cyclist, of the mockable sort (lycra, shoes you can’t walk in, etc).

People who suggested my weight would return were wrong, at least so far. I stabilized at around 193-195 and stayed there for about three years. During that time I’ve gotten my heart rate close to my maximum twice a week or more in most weeks, by riding up steep hills as hard as I can. I occasionally wear a heart rate monitor, but at this point I can also gauge my effort well enough to avoid cheating. (This is harder than it seems: it’s tempting to merely go very very hard, rather than literally as hard as I can; the former is uncomfortable, the latter feels awful, but only for a very short time). Of course I’m also getting other exercise on my bike — I ride for hours sometimes — but for reasons described in that previous post I think it’s the high-intensity exercise that has a big effect on my weight.

Several months ago, some friends and I signed up for a bike trip in the Pyrenees in July (in two weeks now). A few legs of the route are pretty intimidating — long, steep climbs at fairly high elevation — so after signing up for the trip I started training more: getting an hour ride in before work, riding farther and harder on weekends, etc.  After a couple of months I was a stronger biker than ever, but my weight was still around 193-195 pounds. That’s not bad for someone over 6 feet 3 inches tall, but it’s about 8 pounds heavier than when I last tried to play competitive sports at a reasonably high level, fifteen years ago, and 14 pounds heavier than a pro cyclist who is more than an inch taller than I am (as you can see in this touching article about Taylor Phinney.

I’ve never been concerned about weight loss per se; I think of it as an unintended (but welcome) consequence of my activities. Weight loss has never been a goal, in other words.  But about a month and a half after I started training for my trip I was chatting with a biking friend who is much faster and fitter than me, and who weighs even less than Taylor Phinney in spite of being almost my height, and he suggested I lose some weight before my Pyrenees ride. I told him that although it would be great to haul 10 fewer pounds of fat over every mountain with me, I wasn’t willing to diet to do it. I enjoy exercising, and even get satisfaction, though not enjoyment, out of maximum-intensity exercise, but I do not enjoy not eating!

But, upon reflection, I did decide to try to do two things to see if I could lose weight before my big ride: (1) reach my max heart rate 3 or 4 days per week instead of 2, and (2) weigh myself every morning. The theory behind (2) is that if (1) was working then this would provide encouragement to continue, plus perhaps my subconscious would control my appetite a bit in response to the weight on the scale. So for the first time in my life I had a weight loss goal!

That was about 10 weeks ago, and I now weigh under 188 pounds, so I lost about 6 pounds in 10 weeks. I think this puts me at around 10% body fat, maybe just slightly higher. I’m about 20 pounds lighter than I was three years ago when I put maximum-intensity intervals back into my routine.  I seem to have stabilized, with no change in the past two weeks, which is too bad since I’d love to have gotten down to my playing weight from 16 years ago. But I’m not going to give up chocolate crepes and other pleasures of life just to ride my bike uphill a little bit faster.

In conclusion:

  1. I continue to believe that studies that claim that exercise doesn’t lead to weight loss are wrong. I think this may be true of low- and moderate-intensity exercise,  but I think that if you really go all-out, a few times per week, you will end up with low body fat (at least, low by current U.S. standards).  There are plenty of chubby joggers, even ones who regularly run marathons, but there are no fat sprinters.
  2. For me, when it comes to weight it doesn’t seem to matter whether the intense exercise is in the form of running or biking.
  3. To try to discourage a theme that cropped up in the previous post, three years ago: I am _not_ claiming that high-intensity exercise is part of the solution of America’s obesity problem, or anything of that sort. There is just no way most people are going to give maximum effort three or four days a week. I will probably slip back to two days a week when I get back from my trip, in fact, in which case I expect my weight to climb back up in to the low 190s.

This post is by Phil


  1. jonathan says:

    My anecdotal evidence is the label “high-intensity exercise” masks the actual behaviors of the person. That is, we read the label and expect one does the same thing over and over. I never see that in high-intensity exercisers, including myself. The actual behavior is one of continuous attempts to stress the system. That requires improvement, sometimes in technique or by reducing rest time or engaging in complementary exercise that then feeds back as improvement. In other words, I’d say it’s less the intensity than that the label masks the actual behavior of essentially continuous effort towards improvement.

    I see many people who don’t improve. Who don’t try to improve. Who do versions of the same routine over and over with relatively minor fluctuations in effort over time. A smallish number of people attempt to improve on a regular basis, meaning not every time but with a real trendline “up”.

    For example, I use a Precor elliptical. When I’m in top shape, I can do sprints up to 250 rpm or a steady pace of perhaps 225-230 at a resistance level of 4, 5 or 6. (I find I wear down if I keep at that level and the signal to me is the need to drop resistance to keep rpm’s up.) I know a few people who exercise at over 200 rpm. Most just read or watch TV while moving at a comfortable rate.

    My division of workouts is as follows. I do strength for a while. That means trying to lift more each time or so. I then switch to muscle endurance, so I cut the weight – sometimes in half – and lift for 100 reps (in as few sets as possible). It’s amazing how quickly the body adapts to more weight and then to more reps. I then switch to aerobics and wean myself from lifting until I only do aerobics. When I reach that stage, I push the rpm’s as high as possible with sprints and repetitions that really stress me (like trying to do 30 secs on, 30 off with the off not dropping below 200. Really hard.). The point isn’t to list what I do but to point out they’re really ways of seeking improvement by shifting focus. I can’t do strength and endurance, for example, because then I can’t really focus on one. I put on weight during the strength phase and cut it over the rest.

    So when I read studies about these exercise effects, I tend to disbelieve them because I don’t think they are actually testing the underlying behaviors very well.

    • jonathan says:

      I meant also to say that I used to sprint. My recommendation: don’t do it. Your body will fall apart. Sprinters are always injured because they are always pushing to limits of what their muscles, tendons and joints can handle. There is too much mental work required and a small mistake can mean a torn hamstring. Took me 2 years to recover and I learned from that 2 lessons. First, sprinting as you get older is dumb. Second, if I try to do a light sprint workout, I will not be satisfied and I’ll do it until I get the kinks out … and I’ll be sore for 2 weeks. There is nothing more fun than sprinting when you’ve cleared out all the mental and physical inhibitions to completely free-flowing movement and you feel like you’re flying until you run out of air near the line and the oxygen deprivation hits you like a smack in the head. Best drug in the world but bad for your body.

    • Phil says:

      I perhaps should have been explicit in the post about “high intensity,” because you’re right, people mean it to mean different things. In my case, I mean that my heart rate is within 2 or 3 beats per minute of my maximum for at least 30 seconds or so. You’re right, that means that a specific effort that used to get me there, like biking up such-and-such a hill in under 3 minutes and 15 seconds, no longer does the trick now that I’m in better shape: now I have to do the same hill in under 2:55.

      Also, workouts like the recent New York Times “scientific workout” do not get me there. That’s a very tiring workout and there’s definitely a sense in which it is intense, and it does elevate my heart rate to quite a high level, but it does not max out my heart rate. When I say “high intensity” I mean something akin to the original Tabata workouts as described in the Wikipedia page on High-Intensity Interval Training..

  2. K? O'Rourke says:

    Fascinating what remains unknown when it could potentially be increasingly sorted out by using randomised experiments (at increasing costs/decreasing feasibility).

    The conjectures here seem reasonable especially Jonathon’s about continuous effort towards improvement.

    And my favourite activity when at Oxford was sprinting on the Ifley track against other members of the boxing team. They were about 20 years younger than me and at first I would take a 10 or 15 yard head start to make crossing the finishing line first a challenge. It did not take that long for the needed head start to decrease and the enjoyment increase. Fortunately no injuries.

    • Phil says:

      Boxing! Funny, I just took up boxing in November…or, rather, I took up training for boxing. Biking is great for my aerobic capacity and leg strength, but does nothing for upper body strength, hand-eye coordination, balance, or agility, so I figured I should do something besides just ride the bike. Boxing is great, and a great workout, although it’s disappointing to learn that I can’t take a punch: in sparring, if I take a decent hit to the face — even just in light sparring — it’s enough to discombobulate me for half a second or so. Which is, of course, plenty of time to get hit another couple of times. I don’t much like getting hit in the face (although I quite enjoy hitting other people) but it’s definitely a good incentive to put in the training. However, for about the past six weeks I’ve only been going to the gym once a week or so: what with the nice weather and the stepped-up bike training, I just haven’t wanted to spend the time indoors that way.

      • Even moderate levels of getting hit in the head such as typical high school football, or being a soccer forward who frequently heads the ball seem to have measurable long term negative consequences for the brain. Since you’re a guy who uses your brain a lot it’s worth doing a little research on this issue and deciding for yourself whether you really want to risk that.

        • Phil says:

          Thanks for the concern, Daniel. As you can imagine, I’m very attentive to those studies about soccer players and football players, and the effects of sub-concussive impacts in general. I’m comfortable that at the level I’m doing it — about one light sparring session per month, typically involving three short (two-minute) rounds against opponents who are pulling their punches — any damage is negligible. A soccer player warming up for a single game probably takes more impacts to the head than I take in several months, the way I’m training.

          I did once have a slight headache the day after a session; if that happens more than very rarely, I’ll stop sparring.

          Anyway, thanks.

      • K? O'Rourke says:

        > light sparring — it’s enough to discombobulate me for half a second or so. Which is, of course, plenty of time to get hit another couple of times.

        Does not quite seem like light sparring – especially the couple follw on hits. If you are not trying to be competitive, you _should_ be able to avoid most of the risk. There is a lot to learn before you need to experience a lot of the brute force reality – which unfortunately is where a lot of the incentive comes from (no one can say they did not block a punch because they just weren’t trying to.)

        And I do remember the students at Oxford whose parents forbade them to Box but settled on them joining the Rugby team instead!

        Except for when I was in Oxford, I have done Kickboxing which I do believe is less risky. Have been training with this guy’s group for over ten. Never had the aspiration to compete and the only notable injury was a bruised/cracked rib. But then when sparring, if I felt a punch to the head was really hard, I stop for that evening or even week. Still a lot of fun.

        Except for when I was in Oxford, I have done Kickboxing which I do believe is less risky. Have been tarining with this guy’s group for over ten. Never had the aspiration to compete and the only notable injury was a bruised/cracked rib.

        Except for when at Oxford

        • K? O'Rourke says:

          Of course maybe it does explain these posting slip ups ;-)

          • Phil says:

            Ha, yes. I tell you, boxing is not as dangerous as some thinkle peep.

            Nah, trust me, the sparring is light. Perhaps “discombobulate” leaves the wrong impression. When I take a solid jab to the nose it sort of stops me in my tracks for a fraction of a second; I just have to learn not to do that.

            • K? O'Rourke says:

              Concentrate on trying to hit back/move right away (I wish I was better at that).

              And can’t help but mention RA Fisher was the only Cambridge professor (that I know) who knocked out a professional boxer. (The newspaper report did not indicate if it was with the first punch.)

              My inference is that he was the only academic statistician to do so!
              (Somehow this post had to come back to statistics)

  3. Anonymous says:

    If the only thing that changed was exercise, there would have to be an effect on weight. But everyone will increase food intake if they make a significant jump in exercise level. The binary interpretation of the P-value masks what is happening at the individual level and is really the difference between a true effect if we could clone ourselves (potential outcome?) and an effect estimated by treatment means. The explanation for the lack of a population level effect is that the increased exercise is balanced by the increased calorie consumption. But at the individual level, some people are losing weight because they are losing more calories via exercise than gaining via increased food intake while others are gaining weight because they are losing less calories via exercise than gaining via increased food intake. So when Time magazine says “exercise doesn’t work”, the only correct way to interpret this is at the population (public health) level. At the individual level it works for many, many people and I suspect it would work for everyone if they exercised by running 1+ hours per day. It’s pretty hard to eat that much food.

  4. Bill Mill says:

    The usual wisdom I hear in training circles is “you can’t out-train a bad diet”. That is, diet dominates exercise.

    My hypothesis is this: you generally held your diet constant at about the same caloric intake no matter how much you exercised. That was enough to make you ~207 without exercise, ~200 with moderate exercise, and ~190 with lots of exercise.

    I don’t think what you’re experiencing is against normal training wisdom, which suggest that exercise burns calories, but not as many as people think it does. Many people compensate for their exercise by eating more, thereby offsetting the calories they burned at exercise.

    Does that seem plausible to you? Do you think your caloric intake is roughly the same when you’re sedentary and when you’re exercising? That’s a variable you don’t discuss in either post, unless I missed it.

    • Phil says:

      Bill, I think I’m in mild disagreement with just about everything stated or implied in this post, at least inasmuch as it applies to me.

      “Can’t out-train a bad diet”, I’m sure that’s true at some level but my diet has always been pretty good and I haven’t changed it in recent years. I eat much more fruits and vegetables and much less fried food and junk food than most people, and that’s been true for a long time. I do have a major sweet tooth and it would be unusual for me not to eat dessert of some kind at both lunch and dinner. I’d say my diet is healthier than most people’s but far from what would be recommended for an elite athlete. But at no point in the past several years have I tried to change it.

      As for your second point, my caloric intake is not remotely close to constant. I don’t track it or control it, but my daily intake varies wildly depending on my level of activity. On a hilly 60-mile bike ride I might burn something like 3000 calories in just a few hours. When I come back from a ride like that, I eat like a horse…which is why I wasn’t losing weight when I started training for my trip with rides like that. (A Tour de France rider eats over 9000 calories per day. There was an amusing story a few years ago on about a non-bike-rider who tried to eat that much in a day.) I eat what I want when I want, so my caloric intake is closely tied to how much I exercise.

      I think that with “normal” exercise, I eat more to make up the calories. I suspect that for high-intensity exercise, I don’t. A few months ago, when I started doing more rides of 60 miles or more, and more shorter (30-60 mile) rides at moderate intensity, I burned more calories but I ingested more too…no net change. But when I added a few short intervals of maximum intensity, I started losing weight. I don’t think that for me the total calories burned in those high-intensity efforts is the cause of the weight loss — in fact, there’s no way I’m burning enough calories with a few minutes per week of maximum effort to cause anything like the weight gain I’m seeing. Maybe high-intensity exercise suppresses appetite a bit. Or maybe it changes the way I metabolize food — perhaps we’ve evolved so that, if we are at maximum effort like that, our bodies try to store glycogen and shed fat. I don’t have a proposed mechanism, just the observation that, for me, high-intensity exercise leads to fat loss, but medium-intensity exercise (of any duration) does not.

  5. FD says:

    My (limited) understanding is that “you can’t out-train a bad diet’ is less about total caloric intake and more about the composition of a diet. You can consume the same quantity of calories in sugar and carbohydrates or in lean meat and greens and have very different results. This applies to exercise. My understanding is that if Phil is already eating healthy (regardless of how much), and specifically if is sugar intake is relatively low, then increased exercise should lead to decreased body fat. I don’t think that applies if he does not eat healthy, regardless of how many net calories he consumes.

    • Phil says:

      I eat a lot of sugar, though! Actually I’m not sure if that’s true by general U.S. standards, maybe it’s not that unusual. I get hardly any “hidden” sugar — I don’t drink sugar-laden lattes, or snack foods with lots of sugar in them — but I often have a brownie or a couple of cookies at lunch, and a slice of cake or something at dinner, and maybe a couple of squares of chocolate before bed. And when I come back from a long bike ride, my favorite meal includes (but is not limited to!) a big serving of French toast with lots of real maple syrup. Mmmmm.

  6. lemmy caution says:

    You went from 207 pounds to 195 pounds. Your “lose 10+ pounds on strenuous exercise alone” diet is sure to be big seller.

    • Phil says:

      No, I went from 207 pounds to under 188 pounds, so that’s about a 20-pound loss.

      As I said at the end of my comment, I’m not trying to promote this as a solution to anyone’s problems. I am trying to debunk the claim that “you can’t lose weight by exercise alone.” Maybe some people can’t, but I can.

      • Chris G says:

        Years ago I dropped from 207 to 157 running 20-25 miles per week and w/o making any conscious changes to my diet. That said, I suspect I did eat less while I was running a lot than when I wasn’t. The weight loss was over a three year period – give or take. My goal was to get in decent shape and run 10k races in a respectable time not to lose weight per se. I took the experience to be proof that it is possible to lose weight by exercise alone. (FWIW, it took me about five years to put the weight back on after I dropped my mileage to zero per week.)

  7. Janne says:

    You lost ~10kg by high-intensity exercize. Some people lose that amount by adopting a high-protein, low “carbohydrate” (really sugars) diet. I lost about 20kg by accident, adopting balanced, set meals and walking – that is, low-intensity, sustained exercize. There are clearly a number of disparate ways to reduce (and increase) your weight.

    I’ve come to suspect that what really matters is adjusting ingrained habits and body-related expectations. I moved, to a different country with completely different culture, where I knew nobody and didn’t even speak the language. At the time I decided to quit tobacco; it was remarkably easy once the physical craving was gone, as I’d left all my old habits behind.

    But without realizing it, I’d left my old eating habits behind too, along with expectations of how much I “should” eat, how much food is “plenty”, how hungry-feeling is “hungry” and so on. I adopted the food and eating habits of my new place, and lost 5kg (10 pounds) in the first year before I even realized I had done so.

    So I suspect that what unites many successful weight loss events really is that mental and physiological recalibration; our habits and expectations has a large say in how much and what we “should” eat, and our bodies has its own ingrained idea of how big it should be. Recalibrate this through exercize, diet changes or otherwise, and the body weight will adjust to those new set points. Maybe.

    • Phil says:

      My (and Andrew’s) friend Seth has proposed a “just so story” for why we should have evolved a way for high-intensity exercise in particular to reduce one’s fat. I hesitate to even mention it because you can make up a story for anything; that’s why they’re “just so” stories. But here’s the story:

      In most of our evolutionary history, why would our heart rate be really close to the maximum? Either because we desperately need to catch something or we desperately need to avoid being caught. Either way, it’s potentially a life or death situation. If this happens only rarely, then perhaps we’re better off (on average) not shedding fat, because those fat calories can also help in a life-or-death situation when the lean season comes. But if it happens frequently — if we are frequently sprinting for our lives — then we had better not be fat, because (1) fat makes us slow, and (2) we’re more at risk of injury (death!) if we’re sprinting when we’re heavy. So we evolved to become thin if we do all-out sprints (and to store more of our energy in glycogen, which is the power source for intense exercise, and less of it in fat).

      Hey, it’s just a story, don’t take it too seriously. But it could be true!

  8. Byron says:

    I think the difference may actually be that there are a large number of recreational runners while there are very few recreational sprinters. However, I think your point is valid in the sense that the sprint workout brought you a level of focus that meant you were working harder over a particular session than someone puttering along on a treadmill reading a magazine, If you have mental capacity to spare for reading, you’re probably not working tooo hard.

    • Phil says:

      Byron, I agree that there are few recreational sprinters, but why is that? I think it’s partly because jogging for an hour at a moderate pace isn’t as unpleasant as running for 30 seconds as fast as you can, at least for most people. This is definitely true of me, whether running or biking. It sometimes takes a considerable effort of will to go ahead and do my maximum intensity intervals, there’s always part of me saying “why not just ride for an extra half hour instead?”

      But I’ll reiterate that I don’t think the reason the high-intensity training leads to weight loss is that I burn more calories when I do it. A cumulative few minutes per week at maximum intensity doesn’t consume nearly enough calories to account for the weight loss. I think it triggers something else — slightly decreased appetite, perhaps.

  9. Kafka says:

    Quite frankly, I think you are arguing against a straw man. I have seen several studies saying that HIIT (which is basically what you were doing) has several positive metabolic effects. It actually seems to be the latest fad in weight loss circles. Physiologically, there is no reason whatsoever why exercise should not decrease your weight if this exercise is sufficient to create a caloric deficit, and I would be quite surprised if you were able to produce a study claiming that exercise does not help you loosing weight. In fact, in your older post, you were linking to a study by King et al. “Beneficial effects of exercise: shifting the focus from body weight to other markers of health”. They talk about “lower than expected exercise induced weight loss”. Lower than expected does not equal “no weight loss”, and note that they explicitly talk about lower than expected “exercise induced” weight loss. The second article you posted by Melanson et al. with the nice title “When energy balance is maintained, exercise does not induce negative fat balance in […] individuals.”. Note that they were making sure that the exact amount of calories they burned with exercise is being added to the diet again, which is why they talk about “energy balance is maintained”. Since you did not count your caloric intake, I find it hard to see how your data point can be related to this study.

    What you will definitely find are studies saying that diet is more efficient in loosing weight. To get rid of body fat (not weight per se), you need to create a caloric deficit. Exercise is one way of doing so. However, you can burn maybe 500 calories in 30 to 60 minutes of moderate cardiovascular exercise (how much exactly depends on many factors), but you can achieve this 500 calorie deficit much easier by just not having a coke and a donut. I think this is quite well explained here:

  10. x says:

    Misleading title: There are fat sprinters! They just run at another pace…

    • Phil says:

      I know what you’re saying, but I don’t think so! What makes it a sprint is that you are running as fast as you can for a short period; as with a run of any distance, that means different things to different people. There are slow sprinters and fast sprinters, but most people are non-sprinters.

  11. RAstudent says:

    Andrew, love your blog and interesting post. I must say I agree with you about HIT training. Back in my wrestling days I was as lean as could be and in great shape. My diet (wish I would have been more strict about increasing protein and decreasing sugar for strength purposes) was total crap, yet I was lean. Fast forward about 10 years plus kids, phd program, and being tired and I gained about 50 pounds. I recently remotivated myself and have been doing circuit training (primarily by relying on high intensity no rest weight training) and have lost about 25 pounds in about 6 months. My diet is still crappy (love me some barley sodas and salty snacks). Since plateauing at about 195, I have added a 2 mile run 1 or 2 days a week.

    Blah blah aside, how long do you maintain max heart rate? Are you talking 30 second intervals cycled over like 30 minutes or more like and hour?

    • RAstudent says:

      Correction. Nice post Phil.

      • Phil says:

        I don’t claim to be doing anything optimal — if you’re interested in designing an optimal fitness workout for cycling, I highly recommend “The Time-Crunched Cyclist”, by Chris Carmichael — so I hope you don’t take this as training advice.

        My maximum heart rate is about 186 bpm. I count myself as having attained my heart rate goal if my heart rate monitor reads 182 bpm or over; given its response time, that usually means I’m at that heart rate for at least 15 seconds. I rarely maintain it for more than 30 seconds, although I am capable of doing so. However, I often will take a short (15- to 45-second) recovery interval and then go hard again. When I say I try to hit my maximum 3 or 4 days a week, that’s all I mean: at least 15 seconds within a few bpm of my literal maximum.

        However, once a week or so I do a ride in which I hit my max, or close to it, repeatedly. For instance, for the past month I often do a Tuesday evening ride with a group of guys who are stronger than me. The usual route has several steep hills, and I often max out on each of them. But if I skip that ride some week and end up riding on my own and only hitting my target once, I still count it.

  12. Kafka says:

    I just read your old post again and stumbled on this: “The theory that the weight loss/exercise researchers seem to be following is that, since a pretty low level of exertion maximizes the rate of fat-burning — at higher intensity your body switches to using glycogen — this low level must be optimal”.

    Let me correct this perception: among exercise researchers and other informed people, it is widely known that you burn more fat at higher intensities. It is just that the relative amount of fat burned to glycogen burned decreases at higher intensities. The absolute amount of fat burned is nevertheless higher at higher intensities. Google “fat burning zone myth”, if you like. I would be very surprised if exercise researchers would be that uninformed. Maybe you should talk to them to see what research is out there, and how to interpret corresponding research, before jumping to the conclusion that the corresponding research is as fundamentally limited as you believe it is.

    • Phil says:

      Kafka, what you say here contradicts what I have read elsewhere. For instance, this paper shows fat oxidation (measured in grams per minute) falling off rapidly above 75% of VO2(max).

      In fact, Wikipedia says that 85% of calories come from fat at 70% of maximum heart rate, falling off to 5% at around 90% of maximum heart rate and 0% at 100% of maximum heart rate. Sure, caloric output is higher at 90% of maximum heart rate than at 70% of maximum heart rate, but not by anything like a factor of 15!

      If it were just Wikipedia, I’d shrug and say “well, maybe Wikipedia has it wrong and Kafka has it right”, but the research that I’ve seen suggests that you’re wrong about this. Or maybe you’re right and the researchers are wrong. You said you’d be “very surprised if exercise researchers would be that uninformed”…are you very surprised?

      • Kafka says:

        Phil, Wikipedia is right, but it seems the point was not well expressed or received. The lower the intensity, the higher is the fraction of energy used from fat in relation to overall energy being used. Actually, when you sleep, almost all of your energy used comes from fat, relatively speaking. Does this mean that you can slim down by sleeping?

        No, because for weight loss it does not matter much where your energy comes from, only how much energy you burn. To be even more precise: the absolute caloric deficit matters, i.e. absolute number of calories burned minus absolulte number of calories eaten. If you burn 100 calories and 90% comes from fat, how much calories makes this? 90. How many grams weight did you get rid of? 10. You lost 10 grams because there are roughly 9 calories in each gram of fat. Now, if you burn 1000 calories and 20% comes from fat, how many calories does that make? 200. How much weight did you get rid of? You can do the math yourself. But wait, it is even better. These calculations only tell you how much of your weight loss comes from fat loss during the exercise. These 80% in the latter example from carbs are gone as well! Your muscle glycogen storages are emptied! Next time when you will eat, they will be refilled. If they were not empty, where would the same caloric intake go? Right, into your fat cells. This is why for weight loss purposes, the total amount of energy burned matters, not useless fractions.

        But don’t just take my word for it. I could be just a random idiot on the internet who doesn’t know what he is talking about. Maybe you believe the “For Dummies” series more than me. Read up on myth number 2 here: . Don’t trust this source? Not scientific enough? Let’s take this one here, where you find proper references:

        Neither Wikipedia, nor the research you are citing, nor other researchers have it wrong. I rather think that you have not yet fully understood what this research is saying. But that is fine, it took me quite some time to understand all these issues myself. However, I think it is unfair to go around bashing research you just do not fully understand.

        • Phil says:

          Either I’m confused or you are. Obviously each of us thinks the other one has it wrong.

          The paper I linked to shows fat burning at about 0.5 g/min at 65% of VO2max and 0.4 g/min at 75%. That’s grams per minute. Higher exercise intensity yields less fat burned in a given amount of time. The Wikipedia percentages yield the same behavior: higher exercise intensity means less fat burning.

          You give a hypothetical example: burn 100 calories, 90% from fat, that means you lose 10 grams of fat. Burn 1000 calories, 20% from fat, you lose 20 grams of fat. Your math is right you have used crazy numbers to get things to come out that way.

          Suppose I bike for an hour at a fairly easy pace; I might burn 600 calories per hour, about 85% of them from fat…that’s going to be about 50g of fat.

          Suppose, instead, I bike for 30 minutes at an easy pace, 25 minutes at a hard pace, and I do 5 1-minute max intervals. That gives me 0.5 hours x 600 cal/hour x 0.85% fat + (25/60 hours) x 900 cal/hour * 10%fat + (5/60 hours) * 1100 cal/hour * 0% fat = 292 calories from fat, which is only about 25 g. I’ve burned a lot more calories (840 instead of 600) but I’ve burned a lot less fat. This is why the standard recommendation was to exercise in the “fat burning zone” under 70% of maximum heart rate.

          The first of the articles that you mention asserts (wrongly) that you burn more total fat when exercising intensely than when exercising moderately for the same amount of time. That is contradicted by studies like the one that I cited.

          The second of the articles makes a totally different claim: it claims that “fat oxidation during exercise tends to be higher in low-intensity treatments, but postexercise fat oxidation and/or energy expenditure tends to be higher in high-intensity treatments.” Note the first half of that sentence.

          So, the articles that you cite contradict each other. The second one may very well be correct that post-exercise fat oxidation is higher if the exercise intensity was high. I find that easy to believe.

          In short, exercise researchers do think that you burn the most fat per minute while exercising at low-moderate intensity (indeed, not only do they think it, they have evidence to back it up). It does not necessarily follow that exercising at low-moderate intensity is the best way to lose fat, and indeed, many of us believe it is not.

          • Kafka says:

            Phil, at least we agree on that we both think the other person is wrong. I maintain that you misunderstand the research, namely the difference between fat oxidization during exercise and the absolute level of weight loss.

            Let me sum up: you were claiming that a) exercise research does not consider high intensity exercise as a weight loss option, and b) exercise research says that for weight loss low intensity exercise is best. I have provided resources that clearly show your first claim is wrong (see Lyle McDonalds’ or the For Dummies article for a quite accessible explanation), and that your second point is wrong as well. Just to quote my article by Aragon: “In long-term studies, both linear high-intensity and HIIT training is superior to lower intensities on the whole for maintaining and/or increasing cardiovascular fitness & lean mass, and are at least as effective, and according to some research, far better at reducing bodyfat.” Note the very last point: high intensity exercise is FAR BETTER at reducing bodyfat. Did you not just claim that such research does not exist, and that existing research suggest otherwise?

            In any case, I have other things to do than to convince you here, and it seems that this has become more a discussion for the discussion’s sake. Therefore, I am going to stop discussing here. I wish you good luck with your exercise and your goals, though.

            • Phil says:

              I think we agree on a lot of things and disagree on one.

              You said (in the comment that started this thread) “among exercise researchers and other informed people, it is widely known that you burn more fat at higher intensities.” That is not true: in fact, you do not burn more fat at higher intensity. The maximum rate of fat burning — in absolute terms, not as a percentage — is at around 70% of maximum heart rate.

              Or perhaps, when you said “you burn more fat at higher intensities”, you meant “you burn more fat after high-intensity exercise than after medium-intensity exercise”, or something like that…in which case it’s hard to interpret your made-up numerical example, but whatever. Only you can really know what you meant to say.

              I think we agree on most of the rest, including the fact that some researchers recommend high-intensity exercise for weight loss. Perhaps this is becoming more common. However, many researchers interested in exercise and weight loss have chosen to study exercise intensity in the “fat burning zone”.

              I think we also agree that this comment thread is no longer of interest to anyone, including either of us!

              • Phil says:

                Oh, one more thing. You said that I claim that “exercise research says that for weight loss low intensity exercise is best.” This is exactly the opposite of what I claim! Maybe you just said it backwards?

                Many exercise researchers used to assume that for weight loss low intensity exercise is best, but there is no evidence to back that up, as far as I can tell.

  13. Anon says:

    6ft 3in tall. Went from 198 to 185 pounds by avoiding sugar and refined carbs. Other than that I eat anything, including fried chicken. I do jog a few miles a week, but it was the diet change.

    I too have a sweet tooth. What I do is have Lindt 80 or 90 percent chocolate after meals, 1 or 2 squares. Now even 70% chocolate tastes sickly sweet.

    Your sugar craves shoot up your insulin and signals your body to accumulate fat. High intensity sports may counterbalance by signalling to build muscle. My intuition is that is better to get rid of one signal than to have two signals competing against each other.

    • Anon says:

      PS Sugar is addictive, and you may have an addiction problem.

    • Phil says:

      There seems to be no dispute about the fact that high-intensity exercise changes the way the body responds to sugar. (Search for [high intensity exercise sugar] or [high intensity exercise insulin] to find many examples).

      As for whether I’d be better off with (no high-intensity exercise, no sugar) or (high-intensity exercise, sugar), I’m not going to test that, or at least not until age or injury force me to do so.

      I may be “addicted to sugar”, but I don’t consider it to be a problem. Perhaps I’d be even healthier if I cut down on sugar, but I’m pretty healthy now, so any gains would necessarily be small.

      • Anon says:

        A structure subject to multiple stresses is typically more prone to failure.

        I wonder if you are still at risk for insulin resistance.

        • Phil says:

          I have a physical (including a blood test) every few years, and so far my numbers are good across the board. I keep an eye on it.

          Also, it’s always hard to know if I’m giving the wrong impression… I eat “a lot of sugar” compared to what would probably be optimal for me, but it’s not like I’m drinking a Big Gulp at every meal or anything.

          Thanks for your implied concern, in any case. I’ll keep keeping an eye on my numbers.

  14. jim says:

    There are fat sprinters. For example, here are two fat guys out-sprinting a field of 50 skinny guys in Harlem a few days ago:!i=2581892212&k=VwGsD6k

    • Phil says:

      Ha, nice. Did they win the sprint, or were they in a breakaway that survived?

      • jim says:

        That was a field sprint.

        We see fat sprinters win all the time because sprinting uses a different muscle type/energy metabolism (i.e., fast twitch) as well as coordination which depends on the nervous system. It has little to do with how much fat is stored. Indeed, the default (couch potato) state of muscle is fast twitch. Muscle becomes more oxidative with any type of exercise – even weight training! This is a peripheral adaptation. Physiologists categorize the response to exercise into “central” and “peripheral” adaptations. All of these are well studied and quantified by intensity.* Sprinting is clearly high intensity.

        High Intensity does not mean a short or perceived as hard workout but is defined as a percentage of maximal sustainable work for a given interval of time. This is important to understand.

        Also heart rate is highly variable and not really a good indicator of intensity nor work. Just hitting your max does not imply that you sprinted or even really did a high intensity workout. I have data from a race in the heat where I spent a total of nearly 60 minutes (not all at once) at max hr, but I never sprinted. It was extraordinarily intense though. Conversely, one could hit max heart rate in less than a minute, but that doesn’t mean the workout was particularly intense. Instead of heart rate monitors, power meters are the standard for taking a precise measurement of exercise and energy expended.

        Now weight loss comes simply from expending more energy than is consumed. That’s basic physics. So that means your hypothesis is really that high intensity training has a different energy efficiency or that you’re able to use more energy during the high intensity exercise than you would otherwise.

        Even is this were true, high intensity exercise is simply not sustainable as regular exercise (via injury as you experienced or certain fatigue). If you are able to do it regularly (i.e, more than about 20% of the training time) then it is no longer high intensity by definition. The proof is that at some point you would have adapted to actually be able to continue the routine effort for longer than the routine duration. In other words, your maximal power for the duration will have increased.


        • Phil says:

          You say “Now weight loss comes simply from expending more energy than is consumed. That’s basic physics. So that means your hypothesis is really that high intensity training has a different energy efficiency or that you’re able to use more energy during the high intensity exercise than you would otherwise.”

          That’s not my hypothesis, no. In fact, I’ve said repeatedly that the extra energy I expend in high-intensity intervals cannot be enough to account for the weight loss. I don’t think the energy I spend during those intervals has anything to do with it. It can’t, really; the number of calories burned in a few maximal efforts that last a minute or so are way too small.

          I wouldn’t wager on the mechanism(s) but there is ample evidence that high-intensity exercise has effects beyond simply burning a few extra calories. For example it affects ghrelin, which helps control appetite (search for [high intensity exercise ghrelin] to find plenty more), and it affects glycemic control, and it affects blood cholesterol levels. Perhaps there are other changes too.

          It’s a good point that heart rate is presumably a proxy for something more important, rather than being the directly relevant parameter. It’s not clear to me why power output would be the directly relevant parameter either…not that it couldn’t be, but I don’t see why it has to be.

          My observation is that when I add short intervals of exercising as hard as I can, I lose body fat…or at least, that has been true in the range of about 20% body fat down to about 10% body fat; obviously it would have to end somewhere.

          My hypothesis is that the reason this happens is because of changes in appetite — I still eat to satiety at every meal, but perhaps I reach satiety at a lower calorie count than if I didn’t do the intense exercise. (This is not the only possible explanation. For example, perhaps my body stores a calorie as glycogen less efficiently than it stores a calorie as fat, so even if I ingest the same number of calories I retain fewer of them in my body. Or perhaps my basal metabolic rate is increased by the intense exercise. Perhaps there are other possibilities that are also consistent with the observation. Perhaps it’s some mix of these.)

          I’m not positive about the hypothesis/hypotheses but at this point I’m quote sure about the observation.

          • Phil says:

            Oh, one more thing. When I say “high intensity” in this post and others, I mean “an energy output that I can sustain for a minute or two at most”; in my parlance, if I ran a marathon as fast as I possibly can, I wouldn’t call that “high intensity.” Searching around online it is clear that people use “high intensity” to mean all sorts of things. I guess I’m like Humpty Dumpty: “When I use a word, it uses exactly what I intend it to mean.”

            I apologize if that isn’t exactly the way others use the term. I’ll try to be clear in future posts, and perhaps use different terminology.

  15. Andrew Dolman says:

    There’s an interesting discussion about the direction of causality that is mostly missing in these comments.

    The statement that: weight change = energy in – energy out * (some constant) is an identity, basic law of physics etc…

    The thing is that most people who make this statement are assuming that causality runs strictly (energy in – energy out) -> weight change. This is not necessarily the case. It is possible that our behaviour (e.g. level and type of excercise and type of food consumed) can trigger a physiological response to add or lose fat, this then alters appetite, the result being a change in (energy in – out).

    We don’t make this mistake when we think about children growing. When a child grows their energy balance is positive, e.g. energy in > energy out; but we don’t say that they grew *because they ate a lot*, we understand that they grew because that’s what a human does at that age if it can – i.e. unless food is restricted they will eat a lot *because they are growing*.

    The fact that people who are putting on fat are eating more energy than they are expending is a simple law of thermodynamics, but perhaps we should more often think about it as people eating more energy than they expend because they are putting on fat. The ‘just so’, sprinting to avoid being eaten, story would be an evolutionary mechanism to explain why our behaviour can influence a physiological demand to add or lose fat.

    • jim says:

      That’s not missing. Why make guesses involving too many vague parameters when you can directly measure the energy consumed (by weighing your food for example) and energy expended (by using a power meter)? This discussion then reduces to two questions:

      1. Does type of exercise change the efficiency of the human body?
      2. Does type of exercise change appetite?

      The type of exercise is quite clearly defined using energy and time. The rest is really asking “why do people eat too much?”

      • Phil says:

        I don’t know how I’d use a power meter to measure the energy I expend in a day, though.

        You can certainly look at this at several levels of abstraction or specificity, including “how many calories did I consume and how many did I expend”, or “how many grams of various components of my body did I consume, and how many did I build”, and many things in between.

        Andrew is right that people seem to assume that to first order “all else is equal.” Chris G, in a comment below, is an example but I’m not trying to pick on him, lots and lots of people start with an ‘all other things being equal’ way of looking at things. That just makes no sense to me at all, as a first-order model. To me, the default assumption is that if you expend X calories today, you’ll consume X calories tomorrow. The idea that I could go out on a long bike ride and consume an extra 5000 calories compared to a typical day, and not be hungry when I get back and want to eat more than normal…it’s ridiculous. And yet, people really do think that way. I know people (really!) who switched from Coke to Diet Coke and expected that this would help them lose weight, as if they’re not just going to make up those calories some other way. There’s something seductive about the idea that energy consumption should be roughly independent of energy expenditure, I guess… I don’t know, I’m not subject to this feeling, but it seems that many people are.

        Maybe my body temperature goes up, maybe I toss and turn more in my sleep, maybe I fidget more, maybe my appetite is less, maybe I pay a caloric penalty by storing energy as glycogen rather than fat… as I’ve said, I don’t know the mechanism (and I don’t think it’s as easy to figure out as Jim suggests).

        • jim says:

          I wasn’t implying any of that. I was referring to only the exercise itself not the rest of your life:). The changes it induces in the body have been well known for a long time. Again, here’s a link to Coggan’s table of expected physiological changes from exercising at various power outputs:,-by-andrew-coggan.aspx#table2. I’m not motivated enough to dig up all the original publications, but we don’t need to re-derive 50 years of exercise physiology. So I disagree that it’s not easy to figure out. It already has been figured out.

          • Phil says:

            Jim, yeah, I know you weren’t implying that people’s intake is independent of their output, I was just using that comment to kill a few birds with one stone.

            Coggan’s table differs somewhat from Chris Carmichael’s advice (Carmichael was Lance Armstrong’s coach), with Carmichael recommending fairly long intervals just above lactate threshold, whereas Coggan recommends maximal efforts for that. I don’t know who is right.

            I am not very familiar with the literature on training; I have read a few books but nearly none of the original articles. But in this post and my previous one, I’m not really trying to address training for performance, just the relationship between exercise intensity and fat loss. That relationship may be well known by somebody, but if so there is a big signal:noise problem because there are people out there saying all sorts of things, including that “exercise alone does not lead to weight loss.” For me, it can; that’s really the only point I’m trying to make!

            • jim says:

              I wouldn’t believe anything C.C. says, although he must be clever because he’s the only one to have come out of the Armstrong scandal unscathed. Here’s Armstrong’s actual coach: Anyway, hormone injections and blood transfusions skew the data a bit don’t you think?

              It’s great that you’re concerned with weight loss rather than performance because that’s simpler. Weight loss is pretty much linearly related to work (if you don’t change your diet) whereas performance has to do with fatigue which is exponentially related to work. Look up Monod’s critical power concept to get an idea of what I mean. Coggan’s zones are based on what is essentially an updated version of Monod’s model. If I were to “believe” anybody, I would pick the nobel laureate over the sham coach.

              To be precise, it is correct that if you generate more fatigue, then you use more energy at rest to recover. Also, if you increase muscle mass then you use more energy all the time. This is why you don’t see bulky muscled marathoners (that are any good anyway). So I didn’t mean to say these things don’t have something to do with weight loss. What I’m skeptical about is that the only difference between your present workout and pre-weight loss workout is “intensity.” Without data I don’t really know what exercise you were actually doing. My guess is that the difference between your intense and not intense workouts is simply the difference between highway and city miles. That is, you threw in some hard accelerations which requires more energy and thus burns more calories just the way a car burns more gas. It may also be that your old workout simply wasn’t hard enough to elicit any physiological adaptations. So to lose the most weight you just do the workout that leaves you the most tired which is not really conducive to being productive in the rest of life:)

              The bottom line is that on the bike you have to workout with intensity and/or go for a longer ride to get any noticeable changes in the body.

              FYI, this chart translates Coggan zones into HR zones: But beware that HR takes a while to catch up with work when you start an interval and may not indicate really short stuff like the 1-2min intervals you’re doing.

              • Phil says:

                Thanks for the thoughts. Several interesting things there, and I will check out the link. But I really don’t think it’s possible that my weight loss is due simply to increasing the amount of energy I use during workouts by adding higher intensity. I’m talking about literally just a few minutes per week that I am getting up to a heart rate of 182+ rather than 176 or so. Even if I somehow use an extra 600 calories per hour during those few minutes, that would come to only a few hundred extra calories per week. What’s more, if nothing else changed I would simply make that up by eating a few hundred extra calories per week!

                I suspect, though I’m by no means sure, that these brief, intense bouts of exercise somehow decrease my appetite very slightly for a day or two, so if I repeat them, I lose an ounce or two per week until I reach a plateau (which I may be at now).

                I could be wrong, though. Anyway thanks for your thoughts.

  16. Chris G says:

    A simple model:

    dm/dt = C – m*r/b

    m = body mass
    C = food consumption rate [mass per unit time]
    r = burn rate [ calories/(mass-time) ]
    b = average calories per mass

    At steady-state dm/dt = 0 so

    m = C*b/r

    Drop your consumption rate and, all other things being equal, your body mass drops. Change your exercise habits and b and r will change. Lower pct body fat should translate to lower b value. Increased metabolism should increase r. In both cases I’d expect increased exercise to lead to lower body mass w/o any change in C.

    • Paul says:

      m*r is an oversimplification. Calories burned is driven by muscle and brain mass, but not fat. Fat cells are very efficient and burn very few calories regardless of exertion.

  17. Adam says:

    Years ago on Seth Roberts’ board there was post on a book called “Survival of the Thinnest”. The thesis was that 30 minutes of daily exercise in a certain heart rate zone would change your set point. It would be interesting if reaching that thirty seconds of maximum exertion every day would change the set point even further.

  18. pedro says:

    “There are plenty of chubby joggers, even ones who regularly run marathons, but there are no fat sprinters.”
    i think there’s maybe a message about association vs. causation here, but i can’t seem to be able to put my finger on it… ;-)