Training Metabolic Efficiency

One of the critical aspects of Ironman racing is maintaining a fuel balance. We’re looking to reach the final hours with muscle glycogen to spare and the ability to continue the effort to the finish. Pacing and feeding are in fine balance, race too hard and fuel will soon become a limiting factor. As intensity increases so does the utilisation of carbohydrate (CHO) – an easy run might be largely fuelled by fat, but a 5K race is highly glycogen dependent.

Metabolic Efficiency Chart - Fat/CHO Utilisation against Power

An Ironman sits between those intensities the optimal point being determined by your metabolic efficiency. The chart shows results from my metabolic testing, as the power increases so does my utilisation of CHO. If I can’t supply sufficient amounts of glycogen to the muscles then I’ll be unable to reach higher levels of effort. As we deplete our stores we’re effectively diminishing our ability to work further down the line, you may not know it at the time, but you’ll find out in the latter half of the marathon.

Eating carbs helps delay the onset of fatigue from depleted glycogen stores as muscles utilise the newly available CHO in place of their own. Fuelling is an important aspect of an Ironman plan for this reason. What has to be remembered is there are limits to the rate the body can absorb CHO from the intestine. Above that rate food will sit undigested in the stomach potentially causing GI issues, Ironman eating is not an easy game.

A good pacing and fuelling plan should deliver you to the start of the marathon with most of your glycogen spared. To manage this you need to work at an intensity that allows you to consume and absorb CHO as fast as you utilise it. Using the maximum value of 1.7g CHO/min I need to ride below 250W to achieve it. Eat enough and keep my efforts below that level and in theory I start the marathon ready for a good run.

The absorption rate of CHO is fixed, but training can adjust it’s utilisation rate. Increasing Ironman bike power is effectively raising my power at a given level of CHO utilisation (from 250W utilising 1.7g CHO/min to 270W) and developing the ability to sustain that power for the duration (actually hold 250W for 5 hours). An Ironman training program needs to address this issue, the question is how?

Reading Jesse Kropelnicki’s article on starvation workouts kick started my recent thoughts on metabolic efficiency. I largely agree that there’s little place for starvation workouts in most plans, despite that I’ve ridden long on no food. My training history contains a lot of long steady running and riding often without much in the way of nutrition. Volume was high, but intensity was sufficiently low that I didn’t need to eat; over time I could train harder and still not consume much food.

All training provides adaptations towards metabolic efficiency, training performed in a glycogen depleted state provides a strong stimulus for increased fat utilisation. It wasn’t by design, but my preference for riding long without eating has effectively built my present efficiency. The problem is that whilst I’ve become very efficient the intensity of riding was less effective at developing my ability to sustain higher power. I can sufficiently fuel riding at 250W, but I can’t sustain it for the duration of an Ironman.

Most athlete’s need to work on this ability to maintain race paced efforts and have insufficient free time for high volumes of riding. What works when you’re training thirty hours a week needs adjusting for those training less than fifteen. The reality is working athletes ride three or perhaps four times a week, typically only one of those being long. My own training has moved to similar lines as I realised the limitations of constantly riding for distance.

Shorter, harder sessions such as threshold work or intervals require sufficient fuelling to manage the intensity. The objective isn’t to develop fuel efficiency, but to improve other parameters of fitness, perhaps pace or power at threshold or VO2max. To get the most from the workout you need the glycogen to sustain the efforts. As I learnt in the build to Kona, failure to replenish well after can lead to a meltdown.

Longer sessions give an opportunity to train in a glycogen deprived state and encourage increased fat burning. As a rule I never eat on runs shorter than two hours, I can manage this whilst finishing strongly and recovering for subsequent training. However if I run more than 2.5 hours without food performance fades and recovery is notably impaired. Finding the duration you can run without fuelling takes experimentation, but I’d encourage some exploration in the first half of the year. Always eat well afterwards.

During the early part of my season I tend not to eat on any ride shorter than four hours (my ability to manage this has developed with training). I’d suggest an occasional fuel limited long ride in the schedule is a good thing, but should be used with caution and not mixed with any intensity at first. Don’t approach every long ride in this way, whilst it may be effective in improving fat utilisation it’s less effective for other aspects of fitness. Always eat well afterwards.

Initially the effort level you can sustain will be disappointing and for this reason use these sessions sparingly. They’re effective for developing fat utilisation, but not so much for other fitness parameters. These aren’t time efficient workouts. Torbjorn Sinballe mentions a compromise in a recent Inside Triathlon (unfortunately not available online), by only eating after the first two hours of a session he combined the stimulus of reduced reserves with the ability to work harder later.

In practice one low fuel ride per month without pressure on pace or distance should be enough. Knowing intensity will be lower it’s a good combination in a big weekend, one ride fuelled and at a solid pace, the other low fuel. I’d include at least one long bike incorporating your race fuelling strategy per month, it’s going to take practice to achieve close to optimal CHO intake. Otherwise the Torbjorn style sessions seem a good compromise. Always consider recovery and the impact on subsequent training if you feel they’re impaired then eat.

Metabolic efficiency can be overlooked in favour of power and pace at threshold, but it’s a major factor in Ironman execution. For those looking to push performance incorporating some fuel limited training into their programs can yield results. Emphasise it early in the season and always fuel properly when your sessions are more race specific. Like any aspect of training it shouldn’t be rushed, you need to gradually build how much you can do without food.

Ironman Training Library

From nutrition to pacing - a collection of CoachCox blog posts focused specifically on Ironman training and racing.


  • I question this statement:

    “The absorption rate of CHO is fixed”

    Why do we not believe we can train ourselves in this regard just like we’re able to train ourselves in every other aspect of this sport?

  • Well there will be some variance, but it’s ultimately capped by the channels in the small intestine’s cellular lining. Previously the limit was placed at around 1.2g/min based on the rate at which glucose could cross the intestinal wall and enter the blood stream. The higher figure of 1.7g/min is based on their being separate channels for fructose allowing a greater overall crossing of CHO by combining different sources.

    On an individual level they’ll be variance I’m sure we all have varying levels of appropriate channels in the cell lining of the small intestine. I couldn’t honestly say if this is trainable – could we eat in such a way as to alter the rate at which carbs can enter the bloodstream from the intestines? Something to look into, but I’d suspect there’s not going to be significant gains to be made there, nor am I sure what training would provide the stimulus. Perhaps a higher CHO diet may result in the stomach lining optimising for CHO absorption. Effectively the result would be an increase in the number of CHO transporting mechanisms in the intestinal lining. A quick search hasn’t turned up any articles specific to variations in the intestines rate of CHO absorption.

    What I would say is that whilst there’s an upper limit to the body’s rate of CHO absorption across the stomach lining I suspect most of us struggle to eat close to this amount during a race (or training). I saw figures recently for some pros where they talked of consuming over 400 cals/hour in the first few hours of the bike easily at the suggest limit to absorption rate. I know I’ve never consumed that much at any point in a race, clearly if I want to hit that level I do need to train myself to eat more. It’s why i think it’s worth having a regular ride utilising race feed strategy – it’s a chance to test it and to work towards the higher end of the spectrum.


  • Thanks for the thoughts Russ.

    Regarding training your digestive tract. Sure, I agree that the pathways for sugars to get across the membrane are fixed. That is almost definitely not the rate limiting step though. When you’re racing your body decides where it’s going to send the blood. If you’ve got minimal blood in your digestive system and the CNS isn’t telling you to operate your digestive tract it’s not going to be used. That is a decision made by your nervous system about how fast your body is going to digest food. The way you train that is making sure that you feed yourself at race intensity so that your body learns that it needs to send blood to the intestines and to process digestion.

    So, the physiologists out there measured how fast glucose gets across the membrane. The diffusion equation is related to two things though, one is the permeability of the membrane, the second is the concentration gradient across it. This one changes all the time based on how much blood your body is pumping through the capillaries that line the duodenum.

    Let’s think about why people puke when they exercise super hard. It’s not because what was in their stomach suddenly became sickening, it was a nervous system decision that the body wanted to purge the load on the digestive tract because it felt like it was advantageous as a survival response.

    I think this is a case of exercise physiologists and physicians getting hung up on being able to measure something and then deciding that it’s the most important thing to measure. When HRMs first came on the scene people LOVED HR zones, now that we’ve got power we say that HR lies, there’s a disconnect between how fast our body circulates blood and how much work it is doing. Well, what happens when we have a chip that we can implant in our spinal cord that calculates how fast our body is processing oxygen? Then we’re going to say that power lies due to fluctuations in our gross efficiency. The scientific literature is good enough to tell us about the permeability of the membrane but it hasn’t told us how to control the concentration gradient in the stomach to trick your body into doing what you want to to do better.

    I agree with your response though, race specificity is the answer. So let’s deal with it and get better at it instead of suggesting that it’s a hopeless cause. I’m only about 3% away from Kona qualification, if no-one is training their digestive system and I am, then I don’t think it’s unreasonable to expect I could outperform my competition by 3% there. Then I just need to match my competition elsewhere instead of needing to be better than them.

    Today I do the things that you will not so that tomorrow I can do things that you can not.

    My thoughts regarding Jesse’s article:!/jdkrabbe/status/8497415015374848

  • Sure – blood flow to the intestinal lining is going to influence rate of absorption. Figures like 1.2g or 1.7g per minute are based off recorded data in a sample set of subjects. So they probably represent ‘typical’ maximal figures under the conditions of the test. I’ve no idea if anyone has researched whether practicing eating at racing pace improves blood flow to the gut (and hence absorption rate) or could equally enable you to eat more simple by leading to greater suppression of nausea related reflexes due to familiarity (desensitisation of the reflex).

    From the perspective of the athlete, I’d suspect most of us could gain the 3% without too much emphasis on what that maximal figure is. The things to train are your ability to eat sufficiently to get close to your maximal absorption rate and be comfortable eating at race pace. Also supposing we’re capping bike power by a rate of CHO utilisation again we need to train to be able to sustain that cap.

    Practical terms that means some longer rides where we eat like we intend to on race day and we ride like it too. Pace + feeding strategy over a few hours should tell us a lot about how it will work for us and hopefully help us adapt to using it on race day. I think that’s where Jesse is right – it’s not that there’s no value in training under limited fuelling (so long as you don’t impair other sessions), but generally there’s a lot to be gained from training like you’ll race.

    I think I under eat on the bike in races – the truth is I don’t practice race level eating enough. It’s something to incorporate as the next season progresses.

  • One point worth clarifying is that though blood flow into the small intestine is an influencer on absorption rate. The absorption of glucose or fructose through the intestinal lining is by active means through the membranes of the cells. It’s not diffusion over a permeable membrane. The rate at which CHO crosses these membranes will be dependent on the number of transporters available in the membrane and potentially a degree of competition or inhibitory effects of other nutrients.

    My initial thought when you talked about training was that you were suggesting that you could potentially increase the number of available transporters in a similar way that training induces an increase in cellular mitochondrial density. It’s been a long time since I’ve studied cellular biology so was a little uncertain, yet to find a paper that really gives a good answer. A few searches on rates of absorption and mechanisms of transport helped clarify thoughts there.

    I also suspect the issue of GI distress relates to increased sensitivity of the gut to its contents when exercising. We can consume a lot of sugar when inactive, more than we can absorb via the cellular membrane. We can consume enough to induce nausea, I have first hand experience from overdosing on Strawberry Cream Quality Street one Christmas (I was under 10, but a sugar addict even then). When we’re active we add mechanical distress to the stomach, an unusual position (aero on the bike) and probably increased sensitivity to the components of its contents. Training under race intensity with race feeding should help suppress the instinct to empty the stomach as the CNS adapts.

    Which really leads back to the general conclusion – you need to practice race feeding and pace in order to know if it’s up to the job, to help the body adapt to it (in terms of components of fitness, potentially components of the absorption and utilisation of metabolites and any neurological aspects in terms of CNS responses). I’d say the biggest return comes from improving your ability to sustain power for the duration of your event whilst eating sufficiently.

    Training to enhance metabolic efficiency essentially make it easier to spare glycogen at lower power levels, or more importantly for those looking to race their hardest to work harder on the limit of glycogen sparing. The issue remains that if sessions designed to improve metabolic efficiency cost in terms of subsequent training or poorer improvement in other fitness parameters they’re less useful for time limited athletes. So whilst I like them I think using sparingly is the way to go.

  • You might consider textbook on the subject:

    Biochemical, Physiological, & Molecular Aspects of Human Nutrition. – Martha H. Stipanuk

    I like this one, it’s rather detailed but also references you to the sources of original research when you decided you need to know even more. I find it’s a good reference when there’s something you want to look up. Not something you would ever think of reading cover to cover, I certainly don’t use it like that.

  • Cheers

    I’d forgotten how expensive text books are!

  • Hi Russ,

    Here’s a golden nugget


  • That’s a great article Josh. Good to see the numbers.

    Essentially then the more you can eat without vomiting the better and the only way to improve that is to train it. I’m going to have to get in the habit of stuffing my face during some of my longer rides.