The refuelling window: how real is it, and when does it matter?

Few pieces of sports nutrition advice have travelled as far on as little as the thirty-minute window. The version most riders carry is roughly: get carbohydrate and protein in within half an hour of finishing or the session is wasted. It is repeated in gyms, printed on recovery drink tubs, and quietly responsible for a lot of unnecessary anxiety about a shake.

The underlying research is real. The interpretation built on top of it is not. What the evidence supports is narrower and more specific: the rate at which muscle refills glycogen is highest in the first couple of hours after you stop, and that rate advantage matters when your next session is soon. If your next hard ride is tomorrow, the rate hardly matters at all, because you have time to get there anyway.

So the useful question is not "is the window real?" It is "how long have I got, and does the difference in rate change where I end up?" Here is how to answer that.

What the original research actually showed

The foundational study is from 1988. Cyclists rode to deplete muscle glycogen, then received a carbohydrate feed either immediately or two hours later. In the two hours after each feed, glycogen synthesis was substantially faster in the immediate group. Two hours of near-nothing followed by normal synthesis leaves you behind two hours of good synthesis.

That finding is solid and has not been overturned. The muscle is transiently primed after exercise: glucose transporters are translocated to the membrane, the enzyme glycogen synthase is more active, and the muscle is more sensitive to insulin. Carbohydrate arriving in that state is stored more quickly than carbohydrate arriving six hours later.

What the study did not show — and could not, because it only measured four hours — is what happens by the following morning.

Why the window shrinks in importance as the gap grows

Look at the shaded area. It is wide early and narrow late. If your next demanding session falls in the wide part, the two curves put you on the start line with genuinely different amounts of fuel in your legs. If it falls in the narrow part, they do not.

Modern reviews of post-exercise glycogen resynthesis land on exactly this. When recovery time is around eight hours or less, timing and rate of intake matter. When you have 24 hours, the dominant variable is total carbohydrate over that period, and whether you started eating at twenty minutes or two hours becomes a rounding error — provided the total is achieved.

That last clause carries more weight than it looks like it does. Delaying makes it harder to hit a high daily total, simply because you have fewer waking hours in which to eat 8–10 g/kg. For a 70 kg rider that is 560–700 g of carbohydrate. If you do not start until late evening you will not get there, and the failure will look like a timing failure when it is really an arithmetic one.

The numbers

Two practical notes on the top row. Frequent feeding beats bolus feeding for rate — small amounts every fifteen to thirty minutes keep blood glucose and insulin elevated more consistently than one large sitting. And higher-glycaemic carbohydrate is genuinely better here, which is one of the few situations in nutrition where that sentence is true without qualification. This is not the moment for lentils.

Does adding protein help?

This is where the recovery drink industry lives, and the answer is conditional rather than yes or no.

When carbohydrate intake after exercise is below the rate that maximises resynthesis — below roughly 0.8 g/kg/hour — adding protein increases glycogen storage, probably through a greater insulin response. Studies using carbohydrate plus protein hydrolysate mixtures showed higher synthesis rates than carbohydrate alone at those intakes.

When carbohydrate intake is already at the optimal rate, adding protein does not increase glycogen storage further. There is nothing left to add.

So protein is useful insurance rather than a separate mechanism. In practice most riders do not manage 1.2 g/kg/hour of carbohydrate for four hours — that is a real effort — so a mixed feed is a sensible default. And there is an entirely separate reason to eat protein after a ride, which is muscle protein synthesis rather than glycogen; do you need protein after every ride covers that side, and recovery drinks or a meal covers whether you need to buy anything at all. Usually you do not.

When it actually matters

The bit everyone forgets: liver glycogen

All of the above is about muscle. Your liver holds a much smaller store — on the order of 80–100 g against several hundred in the muscles — but it does a different job. Liver glycogen maintains blood glucose, which is what keeps your brain supplied when you are not eating.

It also empties overnight, because you spend eight hours not eating while your brain keeps drawing on it. This is why an evening ride followed by a hard morning session is a specific case: your muscle glycogen may be fine by morning, and your liver glycogen will not be, because sleep spent it. Riding hard on a depleted liver store tends to feel like a general flatness rather than dead legs, and it comes on early rather than at hour three.

The fix is not a timing trick, it is breakfast. Something with carbohydrate an hour or two before a hard morning session restocks the liver, and if you are short of time even a small amount helps. What to eat before a morning ride covers the options when you have twenty minutes rather than two hours. Riding fasted has its place, but the morning after a depleting evening session is not it.

The case for not bothering

There is a reasonable argument, and it deserves airing rather than dismissing. Muscle glycogen is not only a fuel store — low glycogen is itself a signal that drives some of the adaptations you train for. Deliberately training with reduced availability is a real strategy, and if you refuel maximally after every session you never experience it.

The honest caveat is that this is easy to overdo. The evidence for periodised carbohydrate availability improving actual performance, as opposed to improving molecular markers, is mixed. And the failure mode — chronically under-eating around training — is common, damaging and slow to reverse. Under-fuelling takes riders backwards far more often than aggressive refuelling blunts anyone's adaptation.

A sensible resolution: refuel hard when the next session is soon or important, eat normally the rest of the time, and let the occasional easy ride that follows an ordinary evening meal be the low-availability session. That happens by itself without a protocol.

What to do

• Work out your gap first. Hours until the next demanding session is the only input that changes the answer.
• Under eight hours: start eating now, keep eating, use high-glycaemic carbohydrate. This is the case the thirty-minute rule was written for, and here it is broadly correct.
• Over eight hours: eat a normal meal and hit the day's total. No shake required, no stopwatch required.
• Do not let the ritual replace the arithmetic. A rider who drinks a recovery shake at fifteen minutes and then under-eats for the rest of the day is worse off than one who eats nothing for two hours and then has three proper meals.

Before a big event the same logic runs in reverse, over days rather than hours — that is carb loading, and it is a more reliable lever than anything you do in the half hour after a training ride.