Carbohydrates and Fibre: Setting an Intake That Fits Your Training

Carbohydrate is the only macronutrient with no established minimum requirement. The Institute of Medicine's Dietary Reference Intakes (2005) set a recommended allowance of 130 g per day to cover the brain's glucose demand, then acknowledged in the same document that the body can meet that demand through gluconeogenesis from protein and glycerol. People live on very low carbohydrate intakes without deficiency.

Which makes it the wrong question. Carbohydrate is not about survival, it is about fuel for a particular kind of work, and the intake that suits you follows from how much of that work you do. Someone lifting five times a week and someone walking the dog have genuinely different answers, and neither is a moral position.

Why carbohydrate is the elastic macro

In our macro calculator, protein is anchored to body weight and fat is set as a percentage with a floor beneath it. Carbohydrate takes whatever calories remain. That is not an afterthought; it reflects the physiology.

Protein has a requirement tied to lean tissue and training. Fat has a chemical minimum, because two fatty acids cannot be synthesised. Carbohydrate has neither constraint, so it is the sensible variable to absorb changes when calories move, which they do every time you shift between a deficit, maintenance and a surplus.

The practical consequence: as calories fall, carbohydrate should take most of the reduction, because the alternatives are cutting protein when the need for it is rising or cutting fat below a physiological floor. For a 68 kg woman at 1,818 kcal on the balanced setting, the calculator returns 122 g protein, 57 g fat and 204 g carbohydrate. At maintenance on 2,139 kcal the protein figure does not move at all, fat rises to 67 g, and carbohydrate climbs to 262 g. Almost the entire difference lands in one place.

What glycogen actually does

Carbohydrate is stored as glycogen in muscle and liver: roughly 300 to 500 g in trained muscle plus about 100 g in the liver, each gram bound with around three grams of water. That water is why a low-carbohydrate diet produces rapid early weight loss that has nothing to do with fat, and why reintroducing carbohydrate reverses it just as fast. It is also why the scale misleads in the first fortnight of any dietary change, a point we develop in weight loss versus fat loss.

Glycogen matters for performance in proportion to how glycolytic the work is. Fat oxidation supplies energy well at low intensities and its rate is limited; above roughly seventy percent of maximum oxygen uptake, muscle depends increasingly on glycogen because it can be broken down faster. Bergström and colleagues established the relationship in Acta Physiologica Scandinavica (1967), showing time to exhaustion tracking starting muscle glycogen concentration closely.

This is where the training-dependence comes from. Zone 2 work and easy walking run largely on fat and are relatively indifferent to carbohydrate availability, which is part of what makes them useful. See zone 2 training explained. Interval sessions, hard tempo running and high-volume resistance training draw on glycogen directly, and running them depleted degrades both the quality of the session and the recovery from it.

Resistance training depletes glycogen less than endurance work but not trivially. A hard session of multiple compound movements at moderate rep ranges can reduce muscle glycogen substantially, and the evidence that very low carbohydrate availability impairs high-volume strength work is reasonably consistent, even where maximal strength on low-rep sets holds up better.

Setting a target from training volume

The sports nutrition literature expresses carbohydrate needs in g/kg of body weight, scaled to training load. The Academy of Nutrition and Dietetics, Dietitians of Canada and the American College of Sports Medicine set out this framework in their joint position statement in Medicine and Science in Sports and Exercise (2016), drawing on the work of Burke and colleagues.

For the 68 kg woman above, moderate training puts her in the region of 340 to 475 g. Her calculator output at maintenance was 262 g, below that band. Both numbers are defensible, and the discrepancy is instructive: the sports nutrition figures assume performance is the priority and that calories will accommodate it, while a macro calculator is working inside a fixed calorie total set by a body composition goal. When those two goals conflict, something has to give, and it is better to notice the conflict than to average it away.

The reconciliation for most recreational trainees is that they are not doing the volume the upper rows describe. Three or four gym sessions a week sits closer to the moderate row than the endurance one, and a target somewhere in the 3 to 5 g/kg region alongside a fat loss goal is workable. Anyone genuinely training two hours a day and simultaneously in a deficit is attempting two things that pull against each other, and the training usually loses first.

Distributing it around sessions

Total intake matters more than timing, but the pattern is not irrelevant when sessions are close together.

Glycogen resynthesis is fastest in the first couple of hours after exercise, when the muscle is more insulin-sensitive and glucose transporter activity is elevated. That window matters if you train twice a day or on consecutive days with limited recovery. For someone training once daily or less, total daily carbohydrate restores glycogen adequately regardless of when it arrives, and the urgency around post-workout feeding does not survive scrutiny. The wider timing question is covered in meal timing and frequency.

Fibre: the target most plans miss

Fibre is not one of the three macros but it needs a number, and low-carbohydrate plans routinely fall short of it without anyone noticing.

The Institute of Medicine set adequate intake at 14 g per 1,000 kcal, which works out at roughly 25 g daily for many women and 38 g for many men. Most adults in Western countries consume around half that. EFSA's panel arrived at a comparable 25 g reference value for adults in the EFSA Journal (2010).

Reynolds and colleagues, in a series of meta-analyses published in the Lancet (2019), pooled prospective studies and randomised trials covering decades of follow-up and found higher dietary fibre intake associated with lower all-cause mortality, coronary heart disease, type 2 diabetes and colorectal cancer, with the relationship continuing across the observed range rather than plateauing early. The magnitude was large enough that fibre intake is one of the more strongly supported single dietary variables available.

Two categories behave differently. Soluble fibre (oats, legumes, psyllium, apples) forms a gel, slows gastric emptying and modestly lowers LDL cholesterol. Insoluble fibre, from wheat bran, vegetable skins and whole grains, adds bulk and speeds transit. Both are fermented to varying degrees by colonic bacteria into short-chain fatty acids, which is the mechanism behind much of the metabolic benefit.

A practical caution: increase intake gradually and with adequate fluid. Going from 12 g to 35 g in a day produces bloating and discomfort that leads people to abandon the change. Spread it over a few weeks and raise fluid intake alongside it, which the water intake calculator will put a figure against.

Quality, glycaemic index and what it is worth

Glycaemic index ranks foods by the blood glucose response to a fixed 50 g carbohydrate portion. It has real uses and significant limits.

The limits first. The measure applies to a food eaten alone in a laboratory portion, and almost nothing is eaten that way. Fat, protein, fibre and acidity in a mixed meal all blunt the response, so a high-GI food inside a normal meal often behaves quite differently. Between-individual variability is also substantial. Zeevi and colleagues, in Cell (2015), monitored continuous glucose in eight hundred people and found responses to identical meals differing enough that population averages predicted individuals poorly.

Where the concept holds up is at the extremes. Refined starches and sugar-sweetened drinks produce rapid, large excursions with little accompanying nutrition; whole grains, legumes and intact fruit produce more moderate ones with fibre and micronutrients attached. That distinction is worth acting on without needing an index number.

The more useful heuristic is degree of processing. Hall and colleagues ran an inpatient crossover trial in Cell Metabolism (2019) presenting matched diets differing in processing, and participants ate around 500 kcal more per day on the ultra-processed arm and gained weight, despite the diets being matched for macronutrients, sugar, fat, fibre and sodium. Food form and eating rate affected intake independently of composition, which is a finding no macro target captures.

Fitting it together

Choose carbohydrate last, from what is left after protein and fat, then check the result against your training. If it looks thin for the work you are doing, the honest options are to reduce the deficit, reduce the training, or accept that sessions will be worse for a while. Averaging the conflict away serves nobody.

Two refinements are worth the effort. Put more carbohydrate on training days and less on rest days, provided the week still totals correctly. And spend most of it on whole grains, legumes, fruit and vegetables, which brings the fibre target along without separate accounting.

Anyone with type 1 or type 2 diabetes, particularly on insulin or sulfonylureas, is in different territory: carbohydrate quantity and distribution interact directly with medication dosing, and changes need to be made with the clinician managing that. The same applies to inflammatory bowel disease, diverticular disease and irritable bowel syndrome, where fibre type and amount are therapeutic decisions rather than general recommendations. For everyone else, set the total with the TDEE calculator, let carbohydrate absorb the remainder, and judge the figure by how your training feels rather than by how the percentage reads.