Professional road cyclists take in about 90 g of carbohydrate an hour in races, some up to 120. The guidelines said 30-60 g an hour.
In brief
- Glucose alone has a ceiling: it is set by the gut’s transporter, not by appetite.
- Adding fructose opens a second door and lets the body burn more sugar than glucose alone allows through.
- The classic glucose:fructose ratio was 2:1; recent studies prefer near-equal blends, which perform better and cause fewer problems.
- Gut problems affect many athletes, and the gut can be trained.
- Professionals’ intake figures are reported, not measured. Beyond the current ceiling, the science is not yet agreed.
The ceiling was set by the gut, a narrow door through which only a certain amount of sugar passes at a time.
In between lie some twenty years of studies on the gut, on sugars and on how much riders really eat.
The old rules: little, and the same for everyone
For years there was a single rule. The American College of Sports Medicine guidelines put race intake at 30-60 g of carbohydrate an hour and gave it to everyone: they made no distinction by type of activity, duration or the athlete’s level.
Manuals for people in training were even more cautious. Joe Friel, in Fast After 50, spoke of long sessions with roughly 100 to 400 kcal an hour, or about 25-100 g, mostly from drinks. Under an hour, if you had eaten beforehand, water was enough.
The first step was to realise that it depends on how long the effort lasts: according to Jeukendrup, in 2014, more precise advice can be given by duration, not one figure for all. The review by Vitale and Getzin, in 2019, repeats it in bands: nothing needed under an hour, 30-60 g an hour between one and two and a half hours, 60-70 and up to 90 g an hour beyond two and a half hours.
The gut gate
Until 2004 it was thought that carbohydrate ingested during a race could not be burned beyond a certain rate, whatever the sugar. That rate was 1 g a minute, and the reason lay not in the muscles but in absorption. Glucose passes from the gut into the blood through a transporter called SGLT1, and that transporter becomes saturated.
It is a narrow door. That is why the limit is absorption: eating more does not help, because hunger has nothing to do with it, and neither do the legs.
Grams a minute have to be read as grams an hour to see what happens in a race. Just multiply by sixty.
The sums
- 1 g a minute for 60 minutes is 60 g an hour.
- 1.5 g a minute, the peak observed with blends, is 90 g an hour.
- 2 g a minute would be 120 g an hour.
Raising the ceiling by 30 g an hour means getting half a gram more through every minute.
Fructose opens a second door
A second transporter burns more sugar
Fructose does not use the same gateway as glucose: it passes through another transporter, GLUT5. With both sugars in the bottle or the gel, two transporters are used. Glucose at 60 g an hour plus fructose: oxidation exceeded the old limit of 1 g a minute.
Blends using multiple transporters have produced oxidation up to 75% higher than glucose alone. The effect holds for drinks, gels and bars low in fat, protein and fibre.
Power rises, especially in long efforts
A concrete example comes from a study with cyclists: two hours at 54% of VO2max, then a time trial of about an hour. A glucose drink (1.8 g/min) raised average power by 9%, from 231 to 254 W. Adding fructose brought another 8%: 275 W against the 254 W of the glucose-only drink.
The effects on performance, however, are not always seen. In the trials gathered by Jeukendrup in 2014 they show up mainly in efforts of two and a half hours or more, from the third hour onwards.
From 2:1 to 1:0.8, the changing ratio
For a long time the standard ratio of glucose to fructose was two parts to one. Podlogar and Wallis, in 2022, lined up the most recent studies: a near-equal blend performs better, meaning a larger share of the ingested sugars is actually burned, and causes fewer gut problems. The advantage appears even when glucose stays below the old ceiling.
The study by Podlogar and colleagues sets the two approaches against each other. At 120 g an hour, with fructose and maltodextrin in a 0.8:1 ratio, oxidation of ingested carbohydrate was about 17% higher than at 90 g an hour in a 1:2 ratio.
The figure that changes the tone is a different one: the use of stores within the body did not move, 2.15 against 2.20 g/min. The body burned more sugar taken from the bottle, but did not spare what it holds inside. Whether this translates into better performance, the study did not show.
How much do professionals really eat
The numbers are reported, and they do not rise in steps
Intake figures observed in racing need a caveat: they are reported, not measured. They are reported by the athletes and the teams.
A single study on the 1984 Tour de France, on 5 riders, reports 94 g an hour: an isolated figure and perhaps inflated, because it was based on diaries. For the rest, the series starts from low values and reaches high ones, but does not climb in regular steps: across the whole dataset the correlation between year and amount is not significant, with r = 0.12.
For women’s cycling there is only one data point, from a single athlete: too little for a comparison with the men.
| Race | Year | Athletes | g per hour |
|---|---|---|---|
| Vuelta | 1994 | 10 men | 30 ± 8 |
| Vuelta and Dauphiné | 2009 | 15 men | 64 ± 20 |
| Vuelta | 2015 | 9 men | 91 ± 15 |
| Spring Classics | 2018 | 6 men | 51 ± 9 |
| Tour de France Femmes | 2023 | one woman | 84 ± 13 |
The Froome case at the 2018 Giro d’Italia
At the 2018 Giro d’Italia, on stage 19, about 600 g of carbohydrate in total were estimated for Chris Froome, about 100-110 g an hour.
The guidelines climb in steps
The ceiling was not shifted in one go. The 2016 American College of Sports Medicine advice went up to 90 g an hour, but only for efforts beyond 2.5-3 hours and only with blends that use multiple transporters for sugars: glucose and fructose together.
The next step is Jeukendrup’s, in 2017: there is no single number, two are needed. About 60 g an hour up to two hours, about 90 beyond, always with multiple-transporter blends.
The last step is a proposal, not yet a rule. Morton and colleagues look at three things: the new blends, which get more sugar oxidised than before; the studies, which show higher oxidation even above the old ceiling; and the athletes, who in the field were already beyond it. Hence the idea of raising the limit from 90 to 120 g an hour, at least for the trained.
A gut that trains
Absorption is not a factory constant. In a study on 16 cyclists, 28 days of a diet richer in carbohydrate raised the oxidation of ingested carbohydrate. In animal models transporters increase with diet: moving from 40 to 70% carbohydrate, SGLT1 doubles in two weeks.
Then there is the awkward part. Gut problems regularly affect 30-50% of athletes and get worse with heat, dehydration, fibre, fat and over-concentrated solutions. Those on low-carbohydrate diets risk, conversely, absorbing less when it matters.
A field example comes from Rowlands and colleagues, who followed mountain bike races averaging 141 minutes, with carbohydrate at about 95 g an hour (1.2 g/kg an hour): with maltodextrin and fructose, race time fell by 1.8% and abdominal cramps dropped by 8.1 points out of 100.
Grams an hour, not grams per kilo
Carbohydrate in races is counted in grams an hour, not per kilo of body weight: the limit is absorption, and there is no correlation between body weight and oxidation of ingested carbohydrate.
What matters instead is absolute intensity. At low paces little carbohydrate is burned and amounts should be reduced: for a five-hour marathon runner, 90 g an hour would approach or exceed total carbohydrate use at that effort.
In racing up to 1000 kcal an hour are spent for 3-6 hours, and 20-25 g an hour more in a race is worth about 300-400 kcal a day.
Where the science disagrees
Above the current ceiling the data remain uncertain. The doses of 120-200 g an hour seen in the field have no evidence behind them yet, and for Podlogar and Wallis neither a performance improvement nor a saving of stores has been shown.
Performance does not always rise with the dose
In the multicentre study by Smith and colleagues, fifty-one cyclists and triathletes tried 12 doses from 10 to 120 g an hour (glucose:fructose 2:1). The improvement grew with the dose and peaked between 60 and 80 g an hour.
King and colleagues (2018) compared 60, 75, 90 and 112.5 g an hour: the highest power came at 90. In 2019 the time trial gave 219, 228 and 212 W with 80, 90 and 100 g an hour. The direct data show no advantage beyond 90.
Stores and recovery remain a hypothesis
Wilson calls high intake that of 100 g an hour and above. The studies that compare it with 60-90 g an hour do not imitate stage races, which last for days and weeks.
Among professionals, intake in racing tracks the daily total closely: r = 0.89 at the 2015 Vuelta (9 riders), r = 0.95 at the 2009 Tour of Andalusia (6). The hypothesis is that high doses help glycogen recovery in the days that follow. Some data suggest benefits for recovery and for markers of muscle damage, but in some cases very aggressive fuelling increased the use of muscle glycogen.
Everyone absorbs in their own way
A personalised dose of 65 ± 10 g an hour gave, compared with 90, a similar peak oxidation (0.91 against 0.90 g/min) and less gastric fullness. Weight, height and absolute power partly explain the differences between individuals: it does not contradict grams an hour, because the limit is the gut, but guts come in different sizes.
Story written with the help of Mnemosyne.
Sources
- Valenzuela P.L., Leo P., Mateo-March M., Gallo G., Seiler S., Mujika I., “UCI Sports Nutrition Project: The Science of Successful Cycling Performance”, International Journal of Sport Nutrition and Exercise Metabolism, 2025, PMID 41265433, doi:10.1123/ijsnem.2025-0157.
- Jeukendrup A., “A step towards personalized sports nutrition: carbohydrate intake during exercise”, Sports Medicine, 2014, PMID 24791914, doi:10.1007/s40279-014-0148-z.
- Podlogar T., Wallis G.A., “New Horizons in Carbohydrate Research and Application for Endurance Athletes”, Sports Medicine, 2022, PMID 36173597, doi:10.1007/s40279-022-01757-1.
- Podlogar T., Bokal Š., Cirnski S., Wallis G.A., “Increased exogenous but unaltered endogenous carbohydrate oxidation with combined fructose-maltodextrin ingested at 120 g h-1 versus 90 g h-1 at different ratios”, European Journal of Applied Physiology, 2022, PMID 35951130, doi:10.1007/s00421-022-05019-w.
- Wilson P.B., “A Narrative Review of the High-Carbohydrate Fueling Revolution (≥ 100 g/h) in the Professional Peloton”, Sports Medicine, 2026, PMID 41343040, doi:10.1007/s40279-025-02372-6.
- Vitale K., Getzin A., “Nutrition and Supplement Update for the Endurance Athlete: Review and Recommendations”, Nutrients, 2019, PMID 31181616, doi:10.3390/nu11061289.
- Friel J., “Fast After 50”, 1st ed., VeloPress, 2015.
- Jeukendrup A.E., “Training the Gut for Athletes”, Sports Medicine, 2017, PMID 28332114, doi:10.1007/s40279-017-0690-6.
- Morton J.P., Fell J.M., Gonzalez J.T., Hearris M.A., Podlogar T., Pugh J.N., Wallis G.A., “From Metabolism to Medals: Contemporary Perspectives and Revisiting Carbohydrate Guidelines for Fueling Endurance Athletes during Exercise”, The Journal of Nutrition, 2026, PMID 41759826, doi:10.1016/j.tjnut.2026.101442.
- Photo: Fukumoto, CC BY-SA 4.0, via Wikimedia Commons; adaptation: la fagIAnata, under the same licence.
