bike75.gif (2872 bytes)


  Last updated: 1/15/2020

Building Blocks of all Foods

Fats provide 20 to 40% of the Calories in an average daily American diet.

More than 95% of dietary fat is in the form of triglycerides which are constructed of three fatty acid (FA) molecules linked to a single glycerol molecule. Cholesterol and phospholipids make up the other 5%. Cholesterol and phospholipids are important as building blocks for cell growth while triglycerides are an important energy source.

Fat digestion takes place in the small intestine where triglycerides are cleaved into their component molecules - glycerol and fatty acids. The fatty acid molecules are then transported through intestinal lining cells and into the blood. As they circulate throughout the body, they diffuse through cell membranes where they can either be metabolized as an energy source or reconstituted into triglycerides for storage and future use. The bulk of the body's triglycerides are found in fat cells (lipocytes), and a small percent are stored in muscle cells and are more readily available for the production of ATP to power muscle cell contraction. Of the body's 50,000 - 60,000 stored triglyceride Calories, 2,000-3,000 are in the muscle cells. Stored triglycerides are a bigger potential energy source for the muscles than the glycogen in muscle and liver combined (about 1,500 Calories).

Although both triglycerides and glycogen can be metabolized into ATP, only glycogen supports exertion approaching VO2max. As VO2 increases above 50% VO2max, the percentage of exercise Calories provided by glycogen increases until it approaches 100% at VO2max.

What keeps us from using more fat as an energy source for high level exercise? This article suggests it is from bottlenecks in intracellular metabolism, in the conversion of fat into the ATP. There is evidence that training increases the energy provided by fat (and in turn less from glycogen) for any specific riding speed. The result? More riding time before you run out of glycogen and bonk. But even though it can be delayed, you cannot increase your "bonk speed", the %VO2max you can achieve on fat ATP alone. This level of performance is ultimately limited by the rate at which cells can produce ATP from fat, which is about one-third the maximal rate of ATP formation from glycogen/glucose.

For those interested, this article provides further details on fat metabolism in skeletal muscle cells.

This graph (from Romjin et al, we see the relative energy contribution from intracellular glycogen versus triglycerides (along with additional energy released from metabolism of blood glucose and FFAs) as the level of exertion increases. You can see the plateau in total Calories from fat, and as a result its providing a lower percent of total expended Calories as muscle cell energy needs increase. It is this plateau in fat ATP (Calorie) production that limits the maximal %VO2max that can be sustained by fat alone.

The ability to extract FFA from the blood to support exercise reaches a plateau as well and as a result the percentage of total Calories provided from FFAs actually goes down at higher levels of exertion.

The result of these limits in intracellular fat and blood FFA metabolism is a progressive decrease in the percentage of Calories provided by fats as the level of exertion approaches 90 to 100% VO2 max. And above 100 %VO2 max (anaerobic sprints as an example) the contribution of fat Calories is basically non existent with carbohydrates alone fueling the muscle cells.

The graph also shows us the reason for "the Bonk". Once your glycogen stores have been depleted, no matter how hard you try, fat alone provides only 50 - 60 % of the total ATP (Calories) available from a combination of fat and glycogen metabolism. And this caps the upper level of muscle cell performance.


Originally, ketogenic diets were promoted as a way to "force" the body to "burn fat" to fuel daily activities and in turn decrease the amount of body fat. The idea was to substitute fat for carbohydrates in the daily diet leaving only fat as a cell energy source. When there are inadequate carbohydrate calories available, cells do use fat with the metabolic end products being ketones. Thus the "keto"genic diet. (Ketosis happens in diabetics where carbohydrates cannot get into the cells. The cells turn to fat for fuel, and the patient becomes ketotic with an acetone (a ketone) smell on their breath).

The idea of increasing the use of fat for cellular fuel was picked up by endurance athletes as it implied the body's cells could be coaxed or trained toward a fat based metabolism and, as a result, improve endurance performance. That is the muscles would use more of the bodies vast store of fat Calories (and less of the limited glycogen stores) for those long, medium intensity rides.

Let's review the physiology and specific studies that might help us decide if this is a possibility.

There is little question, based on multiple physiologic studies, that fat will not sustain aerobic activity much above 65% VO2max. The following is an example, demonstrating that carbohydrates are needed for performance at higher aerobic (and all anaerobic) levels.

The lack of adequate muscle glycogen limited the level (%VO2max) of performance which in turn meant a longer total time to complete the time trial.

In addition to being unable to replace carbohydrates as a muscle energy source for high level aerobic (and sprint events), a high fat/ low carbohydrate diet can also result in inadequate (or suboptimal) muscle glycogen stores as

Over time, this is a vicious cycle which leads to chronic muscle glycogen depletion and its associated poor athletic performance.

Fats can't substitute for carbohydrates to fuel high level aerobic and sprint events, but what about those endurance events ridden at 50 - 60% VO2max? We know that a trained athlete gets a significant percentage of their energy Calories from fat when exercising at 60 - 65% VO2max. Will a high fat, low carbohydrate (ketogenic) diet provide additional "training" of cell's metabolic machinery, increasing the number of fat Calories used for any specific endurance riding pace, and thus ride further on their muscle glycogen stores? That remains controversial.

As you read the various studies, think about them as being in one of two groups:

Here are a several studies/reviews of the idea that a high fat diet leads to an adaptation of fat metabolism and improved endurance performance. My conclusions?

  1. There is limited data that training on a high fat/low carbohydrate diet beneficially changes the fat/CHO fuel ratio for ATP production for endurance events.

  2. If you want to ride faster, you need carbs to train at a higher %VO2max as well as for the event itself. If you've been training on a ketogenic diet, consider pre-loading with just enough carbohydrates to cover the Calories you expect to expend on the ride.

  3. Any anecdotal (self reported) improvements from a ketogenic diet may be:
  4. If you want to improve your performance in competitive events (even a weekend ride with your cycling group) you should think about pre-loading enough carbohydrates to cover the Calories you expect to expend on the ride.

The comments in this New Zealand study reinforce my conclusions. It describes athletes' real life adaptation to a ketogenic diet. Participants in the study slowly migrated back to a diet containing limited carbohydrates even after admitting to an increased sense of well being on the ketogenic diet. To quote: "Being a translational study, we followed up participants informally 12 months after the study concluded. They were all still competing in endurance events, and while not eating a ketogenic diet, none of them had returned to their previous high carbohydrate, low fat style of eating. Collectively, they reported that once the study concluded they gradually increased their carbohydrate intake until the point at which they felt their performance at high intensities return. They were still restricting carbohydrate and eating more fat than mainstream guidelines recommend, and reported having discovered the optimal macro-nutrient ratio that satisfied a performance, body composition and a health goal."

It is supportive of the suggestion that a low carbohydrate diet (ketogenic diet) may increase the percent of fat Calories used (and in turn lessen the need for glycogen Calories) for endurance level exercise (< 65% VO2max). But limiting dietary carbohydrates did not significantly increase the intensity of exercise that can be achieved from fat metabolism alone. Carbohydrates will still be needed for sprints and other periods requiring more intense exercise.

It suggests (to me) that the answer is not at the extremes - full ketogenic versus unrestricted carbohydrates, but in between. Applying the old adage "moderation in all things" should be interpreted as reasonable fat for a palatable diet, modest protein (meat or plant) and just enough carbohydrates to replace the Calories used in highly intensive athletic pursuits - both for training as well as the events themselves.

A final point to consider is training.

We know that you cannot increase the upper limits of fat based performance (%VO2max) on a carbohydrate free diet. Physiologic barriers in fat based ATP production do not change with training on a carbohydrate free diet. But what if you train on a ketogenic diet and then add back carbohydrates just for that special ride or event, allowing short term VO2max level performance?

If you are competing with equals, who trained on a balanced diet, you will lose. A friend of mine commented that training on the South Beach diet (a low carbohydrate and thus relative high fat weight loss diet) was one of the more difficult things he had ever done. He rode with a feeling of fatigue every day. And as a result, he could not push his training to levels above ~ 65% VO2max. He lost weight (his goal) but in turn felt he had lost his high end aerobic training edge.

Fasted cardio training is another example of a recent training trend, that just as with a South Beach diet, or a ketogenic diet, means you won't have the glycogen stores required to stress (and improve) your VO2max.

You need to push your VO2max in training if you hope to use its improvement to win an event. And if you limit your training, any slight improvement in fat metabolism will not make up the difference. Thus carbohydrates are needed for both training as well as competition. The only exception might be an ultraendurance event where you will never need to perform at > 65% VO2max. For that select sub group of elite ultra-endurance athletes, where long distances are to be covered at a < 65% VO2max level of exertion - and a final sprint is not planned as part of the competition plan.

Final take aways?

Or stated a bit differently:
  1. While you have glycogen stores available, the body metabolizes a combination of glycogen and fat to produce ATP energy for muscle contractions. At lower VO2max, a higher percentage of that total energy is from fat (than from glycogen). But above ~ 65%VO2max almost all ATP comes from glycogen stores and very little if any from fat.
  2. Once glycogen stores are depleted, you produce your ATP from fat alone. But bottlenecks in the cell enzyme "machinery" limit ATP production which in turn caps aerobic performance at ~65% VO2max.
  3. If you train glycogen deficient, there is little evidence you will change either


From the prior section we know that it is unlikely that a full ketogenic diet will change the source of energy Calories from carbohydrates to fat and improve endurance activity. Up to now its support has been the anecdotes (personal stories) of adherents. But now we have some science that suggests unexpected side effects, a loss of muscle mass. As we go to the gym to build muscle mass to be more competitive, anything that causes muscle loss should harm, not help, performance.

The basis of the Atkins low carbohydrate, and more severely carbohydrate restricted ketogenic diet, was the thought that restricting carbohydrates would force fatty acid metabolism and combined with caloric restriction increase the rate of weight loss. A reasonable assumption supported by the logic of fat/carbohydrate based energy physiology. The blind spot is that we forgot about protein, usually a source of cell energy only in starvation situations.

. Initially, a ketogenic diet gives dramatic results. On a standard calorie restricted diet, a study group lost about a pound a week. Switched to a ketogenic diet this increased to 3 1/2 pounds the first week. But on further investigation of their body composition, it was found that the rate of fat loss had decreased by 50% with the bulk of the extra weight loss from water weight. It seemed that the with carbohydrate restriction, the body began cannibalizing its own protein instead of using fat. They lost less fat mass and more fat-free (protein and water) mass.

In a real life scenario, a group of CrossFit trainees switched to a ketogenic diet based on the assumption that a switch to a fat based metabolism would help their endurance performance. They found their leg muscle shrank by 8%. Not the trade off they were expecting.

My take-away from the referenced study?

Fat metabolism decreases on a ketogenic diet with the needed calories to support daily activities now being derived from protein (muscle) breakdown. Fat alone does not fill the energy deficit.

You could argue that if you ate a pre competition diet with enough carbohydrate calories for the planned activity, there might be a tiny bit of extra fat utilized for every calorie of muscle performance. But does this potential offset the risks of a poorer performance from both a less effective training program (see the section above) and loss of muscle tissue?

Your body evolved to use carbohydrates to fuel muscle activity. Excess carbs (beyond those needed to support that work) are a problem as they will be converted into fat. But too few carbs is also bad as the body scavenges needed calories from muscle mass.

And if you have any doubt, this article directly addresses the question of using a ketogenic diet for high level (more than endurance at 50 - 60% VO2max) performance.

Even without addressing the additional risks of osteoporosis, kidney stones, and the rare case of pancreatitis.

A ketogenic diet is not the ultimate diet for a serious athlete.


Brown fat versus white fat. What is the difference? There are definitely health benefits (brown fat is better) and there has been speculation that brown fat (which is generally subcutaneous (under the skin) is the reason Eskimos and other cold climate dwellers tolerate cold so well (insulation as well as access to extra calories for heat). This post from Dr. Mirkin is a nice summary of the current literature. As I read the post, it crossed my mind that the development of brown fat cells is another small piece to explain the benefits of training. Brown fat cells contain extra mitochondria, suggesting they more easily provide fat energy. And as exercise appears to stimulate the development of brown fat, this change may be one of the many training effects that leads to performance improvement (in endurance sports) by supplementing the calories available to the active muscles from your internal glycogen stores.


In terms of overall health, several decades of research and clinical studies have led the US Surgeon General and numerous other health authorities to recommend a diet that is higher in carbohydrate, lower in fats, and rich in fruits, vegetables, and whole grains. Such a diet is associated with a lowering of the risk of major chronic diseases including hypertension, atherosclerosis and heart disease, and certain cancers. But in the past few years the validity of that recommendation has been brought into question. Volunteers fed a diet with 25% of Calories from fat diet versus 45% fat Calories had lower HDLs (good cholesterol) without a change in LDL (bad cholesterol) questioning the idea of substituting CHO for fat to lower risk. (Am J Cardiol 2000:85:45-48.) And for a weight loss diet, a 33% fat diet was more effective in achieving weight loss than a 20% diet. (Int J Obesity 2001:25:1503-1511).

Consider replacing as much saturated fat as possible with mono-unsaturated fats. You also need to be sure to get adequate omega-3 and omega-6 fatty acids -- the essential fatty acids found in many foods, including fish, flax seed, walnuts, and wheat germ. Use nuts such as almonds and walnuts as a topping in your cereal or add them to salads. Substitute monounsaturated (olive oil, canola oil, and avocado) for saturated oils and eliminate trans-fatty acids.

As (1) carbohydrates are superior to fats for high intensity events (both for training and on event day), and (2) fats may, EVEN AT THEIR BEST, only be equal to carbohydrates for lower intensity, endurance events, there is no reason to emphasize fats for training or on the day of an event. And for those who still aren't convinced, it should be remembered that even the leanest athlete has plenty of stored fat available (approximately 100,000 Calories worth in a 70 kg male) without any need for any dietary fat supplementation.

TYPES OF FAT (vegetable vs animal, saturated vs non saturated, LDL vs HDL).


Q.I am trying to chase nutritional information down to primary references. For example, you and many others state that one pound of body fat = 3,500 calories, but give no reference. I calculate 9 calories/gram of fat x 453 grams/pound = 4046 calories per gram of fat.

A.I did find this footnote in Understanding Nutrition 4th edition, 1987, by Whitney and Hamilton; West Publishing Company (a text on nutrition) on page 90. "The reader who knows that 1 pound = 454 grams and that 1 gram of fat = 9 Calories, may wonder why a pound of body fat doesn't equal 9 x 454 Calories. The reason is that body fat contains some cell water and other materials; it's not quite pure fat."

Questions on content or suggestions to improve this page are appreciated.

Cycling Performance Tips
Home | Table of Contents | Local Services/Information