Fuelling cycling training around rotating shifts
Every article about eating for cycling assumes your Tuesday looks like your Wednesday. On a rotating roster it does not. The shift moves the training window, and the training window moves the meals — and if you also keep carbohydrate low, you have removed the one lever most cycling nutrition advice pulls. This is the method I use to decide when and how much to eat around a roster that never repeats. It is a method, not a meal plan: there are no gram weights here, because yours would be different from mine anyway.
Anchor the day to sleep, not to the clock
The mistake I made for months was keeping breakfast at breakfast time. On a night shift that means eating three hours after waking from a daytime sleep, which is not breakfast at all — it is the middle of the night as far as your body is concerned.
Number the meals from the end of your main sleep, whenever that sleep happens. First meal roughly an hour after waking. Last substantial meal at least three hours before the next sleep. Everything else fits between those two. Once the day is numbered that way, a night shift stops being a special case and becomes just another day that starts at two in the afternoon.
The four day shapes
A rotating roster produces a small number of repeating shapes. Once you have a rule for each, you stop deciding and start executing — which matters, because the decision always arrives when you are tired.
| Day shape | Where the ride sits | How the eating works |
|---|---|---|
| Night shift | late afternoon, after the daytime sleep | Two meals before you leave, and a packed portion for the shift itself. The meal before leaving is the one that decides how the night goes — protein and fat, nothing that needs digesting at speed. |
| Afternoon shift | morning, before work | The ride goes first, fasted. Main meal after it, a second packed to eat partway through the shift. Nothing after you get home: eating at midnight costs you the sleep that the next day's training depends on. |
| Early shift | evening, after work | You have been awake since before five. Eat after the shift, ride, then eat again. Keep the second meal early enough that it is finished well before bed. |
| Day off | anywhere | The only shape where you can put the long ride and still eat normally around it. This is where the week's biggest intake belongs, because it is where the week's biggest ride belongs. |
Scale the food to the session, not to the day
A single daily calorie target is the wrong tool when one day is a rest day and the next is two and a half hours. I use four tiers instead, set once and then simply selected:
- No ride — the floor. This is the only number you should set deliberately; the other three are built from it.
- Base ride — the floor plus roughly 200.
- Intervals or a test — the floor plus roughly 400. Hard sessions in a deep deficit are how you lose power rather than fat.
- Long ride — the floor plus roughly 600.
The deficit lives in the weekly average, not in any single day. That one change — moving the deficit off the hard days and onto the easy ones — is the difference between losing fat and losing both fat and fitness.
Setting the floor: your measured resting metabolic rate is a floor, not a target. Eating at it on a twelve-hour shift leaves nothing for the shift, let alone the ride.
Protein is the part you do not cut
In a deficit, the body is choosing what to break down, and training is the signal that tells it to keep the muscle. Protein is the material. Commonly used ranges for an athlete losing weight sit around 1.8 to 2.2 g per kg of bodyweight, and the deeper the deficit, the further up that range you want to be.
Keep it flat across every day — rest days included. The tier system above moves fat up and down; protein does not move. If you are low-carb, this is also the easiest macro to hit by accident, because the foods you are already eating are mostly protein and fat.
What low carbohydrate actually changes
Three things, and only one of them is a problem:
- Long aerobic rides get easier. Fat oxidation at endurance intensity is where a fat-adapted rider is at no disadvantage at all. The long ride is the session that works best on this diet, which is convenient, because it is also the session that moves body composition most.
- Hard intervals get harder. Anything above threshold leans on glycogen, and you are running a smaller tank. Expect the last interval of a set to feel worse than the numbers say it should. This is a reason to keep interval sessions short and well fuelled, not a reason to skip them — they are what raises FTP.
- You lose sodium faster and hold less water. This is the one that catches people out, and it is covered next.
Electrolytes, and the hunger that is not hunger
Low carbohydrate means lower insulin, and lower insulin means the kidneys excrete sodium rather than holding it. Add a sweaty indoor session and a run of night shifts, and the symptoms arrive together: headache, flat legs, a heart rate that will not come down, and a powerful craving at three in the morning.
That craving is frequently sodium and tiredness wearing the costume of hunger. The test costs nothing: drink salted water, wait fifteen minutes, then decide. Most of the time the decision makes itself. Fermented vegetable brine does the same job and tastes like food rather than medicine.
Night shifts make this worse on their own, independent of diet — appetite regulation is one of the first things disrupted circadian rhythm breaks.
The weekly loop that keeps it honest
None of the numbers above are right on the first attempt. They are a starting position that one weekly measurement corrects. Once a week, same conditions, I take body composition from a scale and recovery and training load from the watch, and then apply exactly one adjustment:
| What the week showed | What changes |
|---|---|
| Fat down, power holding or rising | Nothing. This is the state you are trying to stay in. |
| Nothing moved for two weeks running | Take roughly 150 off the floor. Not off the hard days. |
| Power down, or recovery scores sitting on the floor | Add roughly 150 back and hold it for a fortnight before judging anything else. |
| Weight up but lean mass up with it | Nothing. Scales of this kind are noisy, and glycogen and water move first. |
One change at a time, then wait a fortnight. Two changes at once means you learn nothing from either. The waiting is the hardest part of the method and the part that makes it work.
Two things this page is not
It is not a meal plan. I keep my own gram weights and menus private for the simple reason that they are fitted to one 45-year-old on one roster with one set of numbers, and copying them would be copying the least transferable part.
It is also not medical or dietary advice, and I am neither a dietitian nor a doctor. A low carbohydrate diet interacts with medication, with several medical conditions and with pregnancy. Shift work complicates all of it further. If any of that applies to you, the person to ask is a professional who can see your bloods, not a cyclist with a website.
What I will happily share is what the method produced. The training side of it is on the shift work page and the rules that pick each session are on the rules page. The plan those rules are currently producing is on the front page, updated weekly with real numbers, including the weeks it goes badly.