Carb loading

At miscusi we've taken carb loading seriously since before it was a trend: we're the pasta restaurant of Francesco Puppi and the other Power of Carbs athletes. On the fuel tank of anyone who runs, we have a clear view.

Carb loading is the deliberate topping-up of glycogen stores in the days before a race. The reference protocol is easy to say and hard to do: 10–12 grams of carbohydrate per kilo of body weight per day, for 36–48 hours, while cutting training. It only applies to races beyond 90 minutes of effort. Under that duration, you don't need it.

Below we answer the questions you might ask: grams scaled to your weight, the one-day short protocol, why the scale goes up, where the wall really hits, and why a half marathon makes it wasted effort.

It's a strategy to start with muscles and liver fuller of glycogen than usual — and it has a precise duration threshold.

Glycogen is how the body stores carbohydrate. When it runs out, the race often ends with it. Filling it beyond the usual level only works with lots of carbs and little training. The joint position of the American College of Sports Medicine, Academy of Nutrition and Dietetics and Dietitians of Canada (2016) distinguishes two situations that keep getting mixed up ACSM 2016 [1]:

Daily grams look similar — that's where the confusion starts. The difference isn't how much you eat in one day; it's how many days you do it and what you do in between. The 24-hour fueling is your usual diet at the high end of the range and fills a tank that stays the same size; loading asks for two full days at that level with reduced training, to enlarge the tank beyond normal.

So the answer: under 90 minutes you don't have time to burn the extra glycogen, so loading won't get you there sooner — it isn't a wrong strategy, it's the answer to a question you don't have. Eating well in the 24 hours before still helps, and that's enough.

One figure circulating for years also needs correcting. The "load in 24–36 hours" formula is attributed to the guidelines on dozens of Italian pages. The document says 36–48 hours. The 24 hours come from a different study, covered below.

Ten or twelve grams per kilo is far more than most people imagine.

For someone at 70 kg that's 700–840 g of carbohydrate a day, for two days. In food terms: you don't get there by adding one pasta plate at dinner — you get there by rebuilding every meal around carbs and cutting fat and fibre to make room. Trying the night before is a big dinner, not a load.

The old protocol — three low-carb depletion days and hard sessions before three reload days — is no longer necessary. Jeukendrup put it bluntly in the Journal of Sports Sciences (2011): starting with high glycogen helps, but it "does not necessarily have to be achieved with the traditional supercompensation protocol" Jeukendrup 2011 [2].

At miscusi carbs almost always mean a plate of pasta — but for a marathon we all know one extra plate isn't enough. Seven hundred grams is more food than the stomach happily handles for two days straight; at some point the limit isn't hunger, it's space. Guidelines account for that and recommend low-fibre, easy sources, including liquids, because juice delivers grams without taking plate space ACSM 2016 [1].

So work on two tracks: meals as the base, with simple shapes and light sauces, while liquid and spreadable carbs close the gap. How heavy the second track is depends on your weight: at 60 kg little is missing after meals; at 85 kg a lot is missing, because needs rise and stomach space stays the same. Fruit juices, sports drinks, honey and jam matter — even if you love pasta as much as we do. Maltodextrin in water is handy because it's almost tasteless, but gram for gram it does nothing a juice wouldn't do for carb loading.

None of this is race fuel. Gels and salts are taken while you run; they have another job, covered in what to eat before a marathon. From a place that lives on pasta, take that as a honesty check.

It exists, it works, and it requires a condition almost nobody respects: staying still.

Bussau and colleagues (European Journal of Applied Physiology, 2002) had eight trained athletes take 10 g of carbohydrate per kilo per day from fast-absorbing foods, without the classic low-carb depletion phase and without training on reload days. Muscle glycogen rose from 95 to 180 mmol per kilo of tissue in a single day, then stayed stable for the next two days Bussau 2002 [3].

The authors conclude that rest plus high carbohydrate intake maximises stores in 24 hours. The practical point is the first half of that sentence: without rest the sums don't add up — loading while still training fills a leaking tank.

A detail that explains why loading belongs at the end of preparation, not in any random week: supercompensation concerns muscles that have worked beforehand. In Solem, Clauss and Jensen's 2025 Frontiers in Physiology review the phenomenon is "limited to previously active muscle cells", as in the classic experiment where glycogen rose only in the leg that had pedalled Solem 2025 [4]. So the final two rest days come after training — not instead of it.

Because glycogen holds water — and the order of magnitude is around a kilo.

Shiose and colleagues (Journal of Applied Physiology, 2016) followed eight athletes through a 72-hour load at 12 g per kilo per day: muscle glycogen more than doubled and total body water rose by about 0.9 kg, all inside cells and especially in the legs Shiose 2016 [5].

A phrase circulates as if it were a constant — three grams of water per gram of glycogen. Fernández-Elías and colleagues (European Journal of Applied Physiology, 2015) measured that ratio during recovery from exercise in the heat and saw it change with how much you drink: with little fluid the ratio is about one to three; with full rehydration the muscle holds far more, up to one to seventeen Fernández-Elías 2015 [6]. Three grams is therefore the observed minimum, not a physical law.

For the eve-of-race scale: if you weigh a kilo more on race morning, the load worked. It isn't fat — it's fuel with its water around it.

Not at kilometre 30 — a bit further on.

The reference work is Rapoport (PLoS Computational Biology, 2010): a mathematical model of marathon metabolism. The author places the peak incidence of hitting the wall around mile 21 — kilometres 33 to 34 and cites earlier literature that more than two in five marathoners face reserve depletion that can limit performance, with 1–2% of starters dropping out Rapoport 2010 [7].

In the same model are the two numbers that explain what loading does: leg-muscle glycogen is about 80 kcal per kilo of muscle in a trained athlete and reaches 144 after a maximal load, while the liver holds about 350, up to 650 at most. Loading doesn't create new energy: it enlarges the tank by more than half — and that margin is the distance from kilometre 33 to the finish.

Two studies say it directly — and none say the opposite.

Sherman and colleagues (International Journal of Sports Medicine, 1981) tested loading on a 20.9 km trial: starting glycogen was higher and more was used during the run, but times didn't change and glycogen remained in the muscles at the end Sherman 1981 [8]. Twenty-four years later Burke and colleagues (2005) gave eighteen highly trained runners a gel with about 1.1 g of carbohydrate per kilo during two half marathons: the gain was 0.3% — fourteen seconds, indistinguishable from zero. Three of eighteen had GI issues with the gel and worsened by 2.4% Burke 2005 [9].

For the half, normal 24-hour fueling at 7–12 g per kilo applies, with a carb-heavy dinner the night before. What to put on the plate the eve and morning of the race — and what changes between half and full — is in what to eat before a marathon.

You need a food that delivers lots of carbs with little fat, digests well, doesn't tire after the third meal, and can be counted. Four requirements — and a plate of pasta has them all.

The decisive advantage in loading is that pasta can be weighed. When you need 700 g of carbohydrate in a day, going by feel doesn't work: you need to know what's on the plate. A miscusi spaghetti plate has 86 g of carbohydrate miscusi data [11] — about one eighth of the daily quota for a 70 kg person in loading phase.

Ours is Superpasta: fresh, egg-free, just water and organic semolina from durum Kronos and stone-milled ancient grains like Perciasacchi, bronze-drawn. On fibre the logic flips versus the rest of the year: in the two loading days, higher-fibre shapes stay for the lunch far from race day, while the eve dinner stays light.

You came here to learn how carb loading works — and the hardest answer to accept is that it means eating far more than you thought, two days ahead instead of the night before, while staying still.

The rest is detail that matters less than the stories. What decides your race is how much fuel you have in the bank at kilometre 33.

In training weeks, when lunches of 80 or 90 g of carbohydrate become routine, you build them with us as you need: pick a dish from the menu or make your own, choosing shape, sauce and protein one by one. Runners, dieters, plant-based — almost anyone eats at miscusi.

Francesco Puppi, who knows kilometres, has a motto that sums it up: eat pasta, run fasta. Book your table at the nearest miscusi and, after the race, come tell us how it went and whether the carbs were enough.

How do you do carb loading?

Bring intake to 10–12 g of carbohydrate per kilo of body weight per day for 36–48 hours before the race, cutting fat and fibre to make room and limiting physical activity. For someone at 70 kg that's 700–840 g a day. The old depletion phase is no longer necessary.

How many days before do you carb load?

One to two days. Guidelines say 36–48 hours; a muscle-biopsy study showed 24 hours at 10 g per kilo is enough if you rest. The "24–36 hours" formula attributed to the guidelines does not match the text, which says 36–48 hours.

Can you load in a single day?

Yes, on two conditions: arriving from real preparation and staying still for the 24 hours of reload. In Bussau et al. 2002, eight trained athletes nearly doubled muscle glycogen in one day with 10 g of carbohydrate per kilo and no training. What the short protocol removes is the old low-carb diet — not the training that precedes it.

Which races need carb loading?

Those beyond 90 minutes of continuous or intermittent effort: marathon, ultra, long triathlon, gran fondo. Under that threshold, normal 24-hour fueling at 7–12 g per kilo is enough.

Do you need carb loading for a half marathon?

No. The two studies that tested it on half distances found no time gain, and guidelines reserve it for races beyond 90 minutes. A carb-rich dinner the night before and a digestible breakfast are enough.

Why do you gain weight during the load?

Because glycogen is stored with water. In a 72-hour load study, total body water rose by about 0.9 kg, all inside muscle cells. It's fuel, not fat.

Does loading work at any point in preparation?

No: supercompensation concerns muscles that have worked beforehand, so loading makes sense at the end of preparation, in the two days before the race, with reduced training. It isn't a phase to bring forward at will or to do while resting for a week.

What should you eat on loading days?

Pasta with simple sauces, rice, white bread, potatoes, ripe fruit, jam, a lean protein portion. Reduce wholegrains, whole legumes, raw veg and fats, which take space and slow digestion. Nothing you haven't tried before.