My Track Record
GuidesToolsGet the app

Guides › Race fueling

Running

Why do runners carb load before a race?

What glycogen is, what happens when it runs low, and which races it matters for. · 8 min read · Updated October 24, 2026

Key takeaways

How we made this guide: we built it from the research below and checked every number against it: Bergström et al. (1967); Sherman et al. (1981); Hawley et al., carbohydrate-loading review (1997); Hawley, Palmer & Noakes (1997); ACSM, Academy of Nutrition and Dietetics & Dietitians of Canada joint position stand (2016); Rapoport, marathon metabolic model (2010); Hearris et al., muscle glycogen review (2018); Widrick et al. (1993); Bussau et al. (2002); Walker et al. (2000); Solem et al., glycogen supercompensation meta-analysis (2025); Shiose et al., glycogen and body water review (2022); Fernández-Elías et al. (2015); Podlogar & Wallis (2022); Brandt et al., HYROX physiology study (2025); Rappelt et al., HYROX results (2026).

This guide is for runners who keep hearing “carb load” and want to know what it does and whether it applies to their race. Carb loading means eating extra carbohydrate for a day or two before a long race so your stores start as full as they can get. This page covers the why. For the how, see our carb-loading guide.

What is glycogen, and where does your body store it?

Glycogen is how your body stores carbohydrate. Most of it sits in your muscles, where it fuels those muscles. The rest is in your liver, which releases it to keep your blood sugar steady. One review puts a typical adult at about 400 g in muscle and about 100 g in the liver (Hearris et al., 2018). A marathon model used somewhat lower values for a 70 kg man, about 310 g in the leg muscles and 88 g in the liver (Rapoport, 2010). At roughly 4 calories a gram, that is 1,600 to 2,000 calories.

You burn fat too, but stored carbohydrate is the fuel that runs out in a long race, and the one you can change with food beforehand. In the same model, a 75 kg runner finishing a marathon in 3:42, with no carbohydrate taken during the race, needs at least 1,500 calories of glycogen at the start, close to everything the legs and liver typically hold. Pace, body size and diet all move that answer.

What happens when glycogen runs low?

The classic test is from Bergström and colleagues (1967). Nine healthy students cycled at about 75% of their maximum until exhaustion, after three different diets. Muscle glycogen ranged from 0.6 to 4.7 g per 100 g of muscle. Time to exhaustion averaged about an hour after a fat-and-protein diet, two hours after a mixed diet and three hours after a carbohydrate-rich one, and it tracked starting glycogen closely. It was a lab cycling test, not a race, but it is the idea behind carb loading.

A review reports that in events lasting more than 90 minutes, exhaustion usually coincides with very low muscle glycogen, and that blood sugar cannot fully make up the shortfall (Hawley et al., 1997). Runners call the result “the wall”: heavy fatigue and a slowdown you did not plan. A 2010 modeling paper cites studies in which more than two-fifths of marathoners report hitting it (Rapoport).

Where does it happen? You will often hear mile 18 to 20. The paper reports that wall reports among marathoners peak around mile 21 (kilometer 33 to 34), and its model puts glycogen running out at about the same spot for runners of many builds racing at 80 to 95% of their aerobic capacity (VO2max, the most oxygen your body can use). Few would hit it before mile 11, or at all in a marathon run below about 55% of capacity, and harder pacing brings it sooner. Treat any single mile marker as a rough average. Glycogen is one cause of late-race slowing, and the one you can influence with food beforehand.

Which races does carb loading help?

It depends on how long you are running, not how far.

RaceWhat the research foundOur read
5K, most 10Ks (about 15 to 60 min)A review found no benefit from higher starting glycogen in running or cycling lasting 60 to 90 minutes, because plenty is left at the end (Hawley et al., 1997). In trained cyclists, three days at about 9 instead of 6 g/kg of carbs made no difference to distance in a one-hour time trial (Hawley, Palmer & Noakes, 1997).A 5K is shorter than those tests. A normal carb-rich day is enough.
Half marathon (about 1:15 to 2:30)Trained runners covered 20.9 km (13 miles) in the same time whether or not they loaded, although loading raised their muscle glycogen (Sherman et al., 1981). The guideline cutoff is about 90 minutes of sustained effort (Thomas et al., 2016; Podlogar & Wallis, 2022).Finishing under about 90 minutes: a normal carb-rich day. Slower than that: consider a load.
Marathon (about 2:30 to 6:00)Past 90 minutes, higher starting glycogen postponed fatigue by about 20%, and performance in set-distance events improved by 2 to 3% (Hawley et al., 1997). The position stand recommends loading for events over about 90 minutes (Thomas et al., 2016).Worth planning a load.

“Our read” is ours, not a study finding. Finish-time ranges are typical spans.

For scale, 2 to 3% of a 4:00 marathon is about 5 to 7 minutes, if the review’s range applied to you. It comes from a mix of studies, and the 20% figure comes from lab tests to exhaustion, which differ from racing.

In the half-marathon study, the loaded runners also burned glycogen faster (about 5.0 versus 3.1 mmol per kilometer) and finished no sooner, which fits extra fuel mattering only when a race is long enough to use it up.

How much does carb loading raise your glycogen?

It depends on where you start. These studies measured muscle glycogen with and without a higher-carb diet. The units differ between studies, so the percentage is the easiest comparison.

StudyWho and whatMuscle glycogen
Bergström, 19679 students, mixed diet versus carbohydrate-rich diet after hard exercise emptied the musclesAbout 1.8 to 3.3 g per 100 g of muscle (roughly +90%)
Bussau, 20028 endurance-trained men, before loading versus after 1 resting day at 10 g/kg of carbs95 to 180 mmol/kg wet weight (+89%)
Sherman, 1981Trained runners, 6 days at 50% carbs versus 3 days at 50% then 3 days at 70% (with exercise tapering)159 to 203 mmol/kg wet weight (+28%)
Hawley, 19976 trained cyclists, about 6 versus about 9 g/kg of carbs for 3 days459 to 565 mmol/kg dry weight (+23%)
Walker, 20006 trained women, about 48% versus about 78% carbs625 to 709 mmol/kg dry weight (+13%)
Solem, 2025Meta-analysis of 8 running studies: hard exercise, then 3 to 5 days of high-carb eatingAverage gain of about 157 mmol/kg dry weight, with large differences between studies

Percentages are calculated from the values reported in each study.

The “glycogen doubles” figure comes from two studies: the oldest, where students first emptied their muscles with exhausting exercise and a low-carb diet, and one that measured trained men before and after a day of rest and loading. When studies instead compared loading with a moderate-carb diet (about half of calories from carbs, or about 6 g/kg), the gains were 13% to 28%, and results varied a lot between people. Fewer studies include women, and in its cycling studies, the same meta-analysis found a tendency toward a smaller average gain in women (Solem et al., 2025), so these numbers may not describe everyone.

Why does the scale go up after carb loading?

Glycogen is stored with water: a lab study found at least 3 g of water held in muscle per gram of glycogen during recovery (Fernández-Elías et al., 2015). A 2022 review puts the typical weight gain from carb loading at about 1 to 1.5 kg (2 to 3 lb), mainly water, and in the studies it describes the gain was not body fat (Shiose et al., 2022). It also notes that the water bound directly to glycogen has not been measured. Plan for a couple of extra pounds on race morning.

Does carb loading matter more if you eat little during the race?

It can. Think of the race as a bill paid from two accounts: the glycogen you start with and the carbohydrate you take in on the way.

Widrick and colleagues (1993) tested this with eight cyclists in four simulated 70 km time trials of about two hours. Starting glycogen was either high (about 170 to 180 mmol/kg wet weight) or low (about 100 to 110), and riders drank either a carbohydrate drink (116 g over the ride, roughly 57 g an hour) or a drink without carbohydrate. Over the final 14% of the ride, power and pace dropped only with low starting glycogen and no carbohydrate drink. High glycogen alone, or the drink alone, kept pace up. It was a small lab study on cyclists, so treat it as an illustration.

If you know you will struggle to take in much during the race, a full start matters more. How many grams per hour to aim for is in our carbs-per-hour guide.

What about HYROX and other hybrid races?

HYROX combines eight 1 km runs with eight workout stations (our HYROX guide has the format). The ELITE division’s median finish in the latest season studied was about 57 minutes for men and 63 for women (Rappelt et al., 2026). In a simulated race, 11 recreational athletes took a median of 86.5 minutes, and heart rate stayed at 90 to 100% of maximum for about 80% of it (Brandt et al., 2025).

We found no study of carb loading in HYROX, so any advice is borrowed from running research. By the table above, an athlete finishing in an hour to 90 minutes sits where studies found no benefit from extra starting glycogen. Someone finishing well past 90 minutes, or racing a long hybrid event, is closer to where the endurance research applies. We have not found a measurement of how repeated bursts of lifting, carrying and running drain glycogen in HYROX. If you try a bigger carb day, test it before a full simulation, not for the first time in race week. More in our HYROX fueling guide.

How do you check that your plan covers the race?

  1. Estimate your finish time. Over about 90 minutes, loading is worth planning. Under it, eat a normal carb-rich day.
  2. Plan the loading. The grams and timing are in our carb-loading guide, and the foods are in best foods for carb loading.
  3. Plan what you will take during the race. The less that is, the more your starting glycogen matters. See carbs per hour during a race.
  4. Rehearse it on a long run and write down what you ate and how you felt.

That last step is where a record helps. We built My Track Record so logging a meal, or what you ate on a run, is one spoken sentence, with the carbs estimated for you. Note what you ate in the two days before your best long run, and you have a plan to repeat.

Log your long-run fueling in one sentence.
Get My Track Record on the App Store
Andrew, founder of My Track RecordBy Andrew, founder of My Track Record. Andrew has been active all of his life across a variety of sports and continues to stay active through climbing, swimming, weight training, golfing, snowboarding, and more. He built My Track Record so logging food and training takes one spoken sentence and can support any activity. He puts these guides together from published sports science research cited on each page. About Andrew

Sources

General information for healthy adult runners, not medical advice. If you have a medical condition, are pregnant, or have a history of disordered eating, talk to a doctor or registered dietitian before changing how you eat.

Related