Cyclists are obsessed with weight. Pick up any forum thread on climbing performance and within ten replies someone will quote the watts-per-kilogram number they're chasing and the kilograms they plan to shed to get there. The problem is that the methods most cyclists use to get lighter — cutting calories, skipping breakfast, riding fasted — actively work against the goal.
Alex Larson is a registered dietitian who specialises in endurance sport nutrition. On episode 2088 of the Roadman Cycling Podcast she made a case that most cyclists have the body composition equation exactly backwards. The athletes who get lean and stay lean are not the ones eating the least. They are the ones eating the most strategically.
This article pulls together the core of her argument, with the research context it sits inside.
Why restriction fails for cyclists
The instinct is logical: eat less, weigh less. In a sedentary person with no training load to protect, a significant calorie deficit produces weight loss and some of that weight is fat. In a cyclist training 8–15 hours per week, the same deficit produces a different outcome.
Training is a stressor. Recovery from that stressor — the adaptation that makes you fitter — requires raw material: calories, protein, carbohydrate, micronutrients. When raw material is short, the body prioritises survival over adaptation. Muscle protein synthesis drops. Hormone production shifts. Fat oxidation, paradoxically, slows.
Larson's clinical observation matches what sports scientists have documented in the lab: cyclists who under-eat do not get leaner over a training season. They get slower, more fatigued, more injury-prone, and their body composition often deteriorates as lean mass is cannibalised to fuel training.
The cyclist who wants to be leaner at the start of next season is better served by asking how much they should eat, not how little. Understanding your energy availability before you touch your food intake is the starting point, not an optional extra.
The metabolic adaptation trap
When calorie intake drops significantly below expenditure, the body responds in predictable ways. Resting metabolic rate falls. Thyroid output decreases. Non-exercise activity thermogenesis — the fidgeting, the walking, the general movement that burns calories outside formal exercise — drops substantially, sometimes by 200–300 kcal per day, without the athlete noticing.
This is metabolic adaptation. It is not a myth or a social media talking point. It is a well-documented physiological response to sustained energy deficit. The result is that a deficit which initially produced change gradually closes as the body tightens its belt.
The cyclist then faces a choice: eat even less, or accept a plateau. Eating even less typically means training at intensity without adequate fuel, which is where performance collapses and injury risk spikes. Asker Jeukendrup's work on carbohydrate availability during exercise makes clear that glycogen-depleted muscles are less trainable muscles — the quality of each session falls before the body composition goal is reached.
Larson's approach treats metabolic adaptation as an adversary to be avoided rather than a hole to dig out of. The strategy is a modest, periodised deficit rather than an aggressive one — close enough to maintenance that adaptation is minimal, sustained long enough that real change accumulates.
Protein timing for body composition
Getting enough total protein is the first priority. Larson's recommendation for cyclists working on body composition is 1.6–2.2 g per kg of bodyweight per day — the upper end of that range when training load is high and the calorie budget is tighter. A 70 kg rider is looking at 112–154 g per day.
The second priority is distribution. Research on muscle protein synthesis consistently shows a ceiling effect per meal, somewhere around 40 g of leucine-rich protein for a trained athlete. Beyond that ceiling, the excess protein is oxidised for energy rather than used for synthesis. Eating 120 g of protein in a single evening meal is significantly less effective than spreading the same 120 g across three or four meals throughout the day.
This matters practically. A rider who skips breakfast, eats a small lunch, and then back-loads their protein in the evening is leaving adaptation on the table. Pre-ride, post-ride, and evening protein windows each have a role.
The downstream effect on body composition is real. Higher protein intake at adequate distribution preserves lean mass during a deficit, which means a greater proportion of any weight lost comes from fat rather than muscle. That shift in composition is more valuable to performance than the number on the scale.
Carb periodisation done right
Carb periodisation gets misread as a low-carb strategy. It is not. The principle, as Larson applies it with her athletes, is matching carbohydrate intake to the energy demand of each session — high on the days that need it, lower on the days that don't. Total weekly carbohydrate often stays similar to a non-periodised approach. The distribution changes.
On a hard threshold or VO2max day, carbohydrate availability before and during the session is non-negotiable. Asker Jeukendrup's work on multiple transportable carbohydrates shows that trained athletes can absorb 90–120 g of carbohydrate per hour during high-intensity effort when the gut is trained for it. Showing up to a hard session under-fuelled limits the training stimulus and starts a recovery debt that affects the next two or three days.
On a recovery ride or rest day, carbohydrate needs are lower. Protein and vegetable intake can fill that space without adding unnecessary energy. This is where the modest deficit lives — not in the hard sessions, but in the recovery days where the body is doing less work and needs less fuel.
The practical article on fuel for the work on this site covers this day-by-day approach in more detail. The point Larson makes is that most cyclists get this exactly backwards — they under-eat on hard days because they're watching calories, and then compensate by over-eating on rest days because they're depleted and hungry.
The scale vs the mirror
Body weight fluctuates by 1–3 kg day to day based on hydration, glycogen storage, gut contents, and sodium intake. A rider who eats high-carbohydrate the day before a big session will weigh more the following morning because glycogen is stored with water. That is not fat gain. That is performance readiness.
Cyclists who weigh themselves daily and react to every fluctuation create anxiety around fuelling that directly undermines body composition goals. Larson's position is that daily weigh-ins can be used as data — trends over weeks, not individual readings — but that for athletes with any history of disordered eating patterns, the scale should come out of the equation entirely.
The metrics that actually track progress are more useful: how does your kit fit over a 12-week period? Are your power numbers at a given effort level holding or improving? Is training consistency better? These are the questions that body composition work is ultimately trying to improve.
For cyclists who want an objective reference point, the race weight calculator gives a framework based on height, training history, and sport-specific norms rather than arbitrary targets pulled from a forum.
Larson's clinical experience is consistent with the broader evidence: athletes who focus on performance metrics as the primary feedback loop make better food choices, train more consistently, and end up in a leaner, more functional body than athletes who chase a number on the scale.
Sustainable habits for 12 months
The research on weight loss maintenance is unambiguous: short-term restriction produces short-term results. Studies following individuals post-diet show that the majority regain the lost weight within two years, and a significant proportion end up heavier than before they started. The body composition literature in endurance sport follows the same pattern at a smaller scale.
Larson's framework asks a different question: what eating pattern can this athlete sustain across a full 12-month training year, including off-season, base phase, build, race season, and transition? That question rules out a lot of the strategies the cycling internet promotes.
It rules out long fasting windows that make high training loads unsustainable. It rules out very low-carbohydrate approaches that degrade high-intensity capacity. It rules out severe calorie deficits that drive metabolic adaptation. What it leaves is a pattern that is specific, structured, and quietly consistent.
Prof. Stephen Seiler's work on training distribution offers a useful parallel here. Around 80% of training volume at low intensity, 20% at high intensity — not because it's a comfortable middle ground, but because the physiology supports that split for long-term adaptation. The nutrition equivalent is not the dramatic intervention. It's the practice that works at 80% effort across 52 weeks.
For cyclists who want that support structured around their specific training load, Roadman's Not Done Yet programme builds nutrition as one of five integrated pillars alongside training, strength, recovery, and accountability. The coaching is 1:1, and the nutrition approach reflects exactly what Larson describes — not a template, but a plan calibrated to the individual's energy demands, timeline, and training phase.
Larson's final point on the podcast was the one that stays with you: the cyclists who are still lean at 45 or 50 are not the ones who dieted hardest in their 30s. They're the ones who built an eating pattern they didn't have to think hard about. Sustainability is not a compromise on ambition. It is the strategy.
If you're starting from scratch on this, the most useful first step is calculating your current energy availability accurately and honestly. Start there before you touch your food intake in either direction.