Published June 14, 2026 03:33AM
Everyone is hyper-focused these days on how many carbs they can cram in during a race. But those efforts are wasted if you start the race with a partly empty fuel tank. Carbohydrate-loading protocols have been around for more than 50 years now, but there’s still debate about the best way to ensure that your pre-race fuel stores are fully maxed out. A new study offers some useful data.
The study comes from a group at Liverpool John Moores University led by Robyn Jones and Julien Louis, and is published in the Scandinavian Journal of Medicine & Science in Sports. It tests three different two-day carb loading protocols under realistic conditions—that is, with trained endurance athletes eating and training as they normally would in the days before a big race—and finds that, within the range of carb intakes tested, more is better, with no significant downsides.
The Carb Loading Backstory
The idea of carb loading emerged in the 1960s, when scientists in Scandinavia figured out how to take muscle biopsies to measure exactly how much carbohydrate (in the form of glycogen) was stored in leg muscles. They noticed that endurance was improved when athletes started with more glycogen in their legs, and also that the amount of glycogen could be increased—supercompensated—if you first emptied your glycogen stores by exercising to exhaustion.
These insights led to the first carbohydrate loading protocols, which were complicated and demanding. The classic week-long protocol involved a long bout of exercise to exhaustion, three days of low-carb diet, and another long bout of exercise to exhaustion, and then three days of high-carb diet. It may have boosted glycogen stores, but it wasn’t a very pleasant way to spend your final week before the big race.
Subsequent studies suggested that the elaborate depletion protocols weren’t really necessary, in part because trained athletes are always exercising anyway, even in the week before a race. By the early 2000s, the pendulum had swung so far in the opposite direction that some scientists were suggesting that all you needed was one day of eating more carbs than normal (10 grams per kilogram of body weight, or g/kg, an amount we’ll return to) to max out your glycogen stores.
These days, the American College of Sports Medicine’s official stand suggests spending two to three days eating 10 to 12 g/kg to max out muscle glycogen. That’s standard practice among elite athletes, but when Jones reviewed the literature as part of her doctoral studies, she found the evidence for how much and when wasn’t as clear-cut as she expected—which is why a new study was needed.
The New Study
Jones’s study tested a two-day carb-loading protocol with three different amounts of carbohydrate: 6, 8, and 10 g/kg, compared to a baseline diet of 4 g/kg. She recruited 11 trained endurance athletes (8 men, 3 women), each of whom completed three trials of a five-day protocol designed to mimic a real-life race taper:
- Day 1: 60 minutes moderate cycling; eat 4 g/kg of carbs
- Day 2: 30 minutes easy, then hard intervals (8 x 5:00) with muscle biopsies before and after; eat 4 g/kg
- Day 3: 60 minutes moderate cycling; eat 6, 8, or 10 g/kg of carbs
- Day 4: 30 minutes moderate cycling; eat 6, 8, or 10 g/kg of carbs
- Day 5: Muscle biopsy
The key result is that eating more carbs led to more glycogen stored in their leg muscles. Here’s the data for the four different conditions (including the pre-load measurement from 4 g/kg of carbs):
Jones and her colleagues had hypothesized that there would be a “ceiling effect,” so that going from 8 to 10 g/kg wouldn’t give as much of a boost as going from 6 to 8. But there’s no evidence of any ceiling effect: 10 g/kg is much better than the alternatives, which leaves open the question of whether 12 g/kg might be even better.
There are some potential caveats to carb loading. One is that 10 g/kg is a lot of carbohydrate. A one-cup serving of cooked pasta supposedly has 42 grams of carbohydrate; if you weigh 150 pounds, 10 g/kg is the equivalent of 16 servings. Feeling absolutely stuffed and bloated is hardly what you’re aiming for the night before a marathon.
The study measured a bunch of GI outcomes, but contrary to expectations, they didn’t find any higher prevalence of symptoms like nausea, cramps, or flatulence with the highest carb loads. The subjects did report feeling more full with 10 g/kg, but not to a problematic degree. They even ate two blue-dyed muffins on Day 4, and didn’t observe any significant difference in how long it took for blue dye to show up in the toilet (20.2, 19.0, and 18.7 hours in the low-, medium-, and high-carb conditions, if you’re curious).
The other concern is that carb loading can lead to weight gain and bloating, since every gram of glycogen is thought to be stored with as much as four grams of water. But Jones and her colleagues didn’t see any differences in total body weight. Here’s the data:
Carb Loading in Practice
What I take away from this is that two days at 10 g/kg is a good carb-loading protocol—if you can handle it. It seems clear that less is worse, and the volunteers in this study didn’t seem to have any problems at 10 g/kg. On the other hand, it’s possible that more is better: the ACSM guidelines, after all, go as high as three days at 12 g/kg. But that’s getting into some seriously challenging eating.
In this study, the researchers provided subjects with all their meals. There isn’t a ton of detail about the meals, but for the two higher carb levels, they added things like “jelly sweets,” orange juice, and sports drinks: good sources of carbohydrate that have minimal fiber, in order to reduce the risk of GI symptoms. In general, that’s probably a good principle to keep in mind even for the meals you eat during those days. The cyclist Mike Woods once told me that his taper meals followed what he called a “five-year-old’s diet,” avoiding things like whole grains and high-fiber vegetables.
In Maurten’s account of Sabastian Sawe’s sub-two-hour marathon fuel plan, they mention that he was adding an extra sports drink with 80 grams of carbs two days before the race. The day before the race, he drank down 160 grams of carbohydrate in sports drink, and added another 40 grams in an energy bar. If Sawe is aiming for around 600 grams of carbohydrate per day (based on a guess that he weighs around 130 pounds), then he’s already getting 200 of those grams from his easily digested sports nutrition products.
There’s one final wrinkle to consider. Even after all that carb loading, the glycogen in your liver—the second most important storage depot after your leg muscles—gets depleted overnight because you’re still fueling your brain. For that reason, you still need to top up your stores on the morning of the race. Sawe reportedly had a light breakfast, sipped sports drink on the bus to the start line, and took a gel with 25 grams of carbohydrate five minutes before the start. Yomif Kejelcha, who also broke two hours in coming second behind Sawe in London, was reportedly sipping sports drink on the morning of the race and took a gel with 45 grams of carbohydrate shortly before the start.
Of course, it’s worth remembering that there are no medals given for the highest glycogen levels. The South African sports scientist Tim Noakes, once a booster of carb loading, now believes that glycogen levels are irrelevant to race performance and you only need enough carbohydrate to keep your blood sugar high. Most sports scientists believe otherwise—and so, apparently, do Sawe and Kejelcha.
For more Sweat Science, sign up for the email newsletter and check out my new book The Explorer’s Gene: Why We Seek Big Challenges, New Flavors, and the Blank Spots on the Map.