The WattPlan Podcast · Episode 6 · 10 min
Should you eat 100g of carbs an hour?
Read the transcript
James. Hello, and welcome to the WattPlan Podcast. I'm James. Somewhere between family life, the day job, the odd podium and building WattPlan, I read a frankly unreasonable amount of sports science. This week it's fuelling. Two new trials say a hundred grams of carbs an hour beats the old sixty on a long ride. So should the rest of us be eating like the pros? Lyra, our AI research voice, has been through the papers, and every study is linked in the show notes.
Lyra. Ha, the big fuelling question. Straight answer, on long, hard rides about eighty to ninety grams an hour is the target we'd give you. The newest trials point higher, to a hundred or so, but that's something to practise in training first, not a number to try on race day. And nothing in them gives a reason to go past a hundred and twenty.
James. Okay, so not a free-for-all. Before the new stuff, where does the old sixty grams even come from?
Lyra. Right, so the reference point is a 2014 review in Sports Medicine by Asker Jeukendrup. It set out guidance by how long you're going. Around an hour, a mouth rinse or a small amount of carbohydrate is enough. For two to three hours, about sixty grams an hour. For ultra-endurance, about ninety. And the reason sixty comes up so often is that a single type of sugar can only be burned at up to about sixty grams an hour. [5]
James. Which is why everyone goes on about mixing sugars.
Lyra. Exactly that. At the higher rates, the review says you need more than one type, glucose with fructose for instance, so you can burn more and so carbohydrate doesn't build up in the gut. Every high dose trial in this episode mixed them.
James. Right, the new ones then. What did the Danish team actually do?
Lyra. This is the first new trial, by Oscar Mazza, Kasper Gejl and colleagues, published this year in Medicine and Science in Sports and Exercise. Fifteen trained male cyclists and triathletes. Each did a fifteen minute time trial fresh, then rode for 240 minutes at 60 percent of that time trial power, with hard intervals mixed in, then did the fifteen minute test again. Once on sixty grams an hour, once on a hundred, and double-blind, so nobody knew which was which. [1]
James. And?
Lyra. Both rides cost them power, but the bigger dose cost less. On sixty grams, their time trial power fell 9.3 percent below fresh. On a hundred, it fell 6.4 percent. That worked out at 9 watts more in the final test, after four hours of riding.
James. Nine watts at the end of four hours. Hang on, is that because the extra carbs saved glycogen in the legs?
Lyra. That's the surprising bit. No. They took muscle biopsies from the thigh before and after, and glycogen fell by a similar amount on both doses. What held up better on a hundred grams was how much carbohydrate the whole body was burning late in the ride.
James. So if I'm hearing you right, the tank in the muscle empties the same either way, but the extra sugar keeps the fire going at the end.
Lyra. That's a lovely way to put it, and close enough. The extra fuel kept carbohydrate available to burn late on, without saving any muscle glycogen. Blood sugar stayed steady on both doses, so it wasn't that the sixty gram ride ran dry in the blood either.
James. Right. So if a hundred beats sixty, surely a hundred and fifty beats a hundred?
Lyra. The obvious next question, and the second new trial asked it. Henry Martyn, James Morton and colleagues, led from Liverpool, also this year in the same journal. Eight trained men rode for three hours and then did a test to exhaustion, on ninety, a hundred and twenty and a hundred and fifty grams an hour, in random order. There was also an earlier ride on just ten grams, which doubled as familiarisation. [2]
James. And the hundred and fifty?
Lyra. The drinks carried a tracer, so they could tell the drink's carbohydrate from the body's own. As the dose went up, riders burned more of the drink. But total carbohydrate burning stopped rising at a hundred and twenty. It was the same at a hundred and fifty. And how long they lasted didn't differ between ninety, a hundred and twenty and a hundred and fifty. Their best estimate for exercise capacity was 107 grams an hour, somewhere between 92 and 122.
James. So past a hundred and twenty you're pouring in sugar for nothing.
Lyra. In that trial, yes, for total carbohydrate use and for how long they lasted. One more thing. The spread between riders was wide. Some burned far more of the drink than others at the same dose. The authors read that as a case for personalising the dose rather than one number for everyone.
James. Okay, but here's my problem. Fifteen blokes in a lab in Denmark and eight in Liverpool. Does any of this really apply to someone doing a sportive on a Sunday?
Lyra. Fair challenge, and the honest answer is partly. Both trials were small and both were all men, so we can't say how women respond. The Danish one held riders at a fixed 60 percent of their time trial power, not the surges of a real ride. And it compared only sixty and a hundred, not eighty or ninety, which is where most guidance already sits. So it shows a hundred kept more late power than sixty over four hours. It doesn't show a hundred beats eighty-five.
James. And the stomach side? Because that's where this falls apart for most of us.
Lyra. That's the other gap. Neither abstract reports stomach symptoms, and that's what matters most at these rates. So we know what happened in the lab, and we don't yet know how many riders could eat this much comfortably out on the road. Nobody has tested that properly yet.
James. Right. Let's split it by who's listening. If you're new to this, where do you even start?
Lyra. If you're new to structured training, start by counting. On your next ride of three hours or more, log every bottle, gel and bar, and divide by the hours. That's your real number, not the one you think you eat. And for a one hour turbo session, a small amount is plenty. The review puts a mouth rinse or a little carbohydrate as enough around the hour mark.
James. And if you train with power?
Lyra. If you train with power, you've got a ready-made test. End a long ride with fifteen to twenty hard minutes and note the power. Then do the same ride at a higher intake and compare. It's a small version of the Danish trial on your own legs. And WattPlan's Durability score is built partly from the same thing, comparing the second half of your rides of two hours or more with the first half.
James. And if you race, or you've got a big four hour event coming up?
Lyra. If you race, or you've got a long sportive or gravel race that finishes hard, aim for eighty to ninety grams an hour and practise it in training. The newest trials point to a hundred or more. If you want to try that, rehearse it over weeks of long rides first, never on race morning. And neither trial gives a reason to go past a hundred and twenty.
James. How fast do you build up, though? I can't see myself going from sixty to ninety in one go.
Lyra. No, and you don't need to. Add ten to twenty grams an hour every couple of weeks on long training rides. Mix your sugars, glucose or maltodextrin with fructose, because a single sugar tops out at about sixty grams an hour. And if you plan the ride in WattPlan's Route analysis, in Premium, it marks a feed stop about every forty-five minutes, and the course file for your bike computer says sixty grams at each. That's about eighty an hour, a good target to build towards.
James. Brilliant. So what's the homework this week?
Lyra. Three things. One, count what you actually eat on your next long ride. Two, if it's below eighty grams an hour, add ten to twenty grams on the next one, with mixed sugars. Three, finish one long ride with fifteen hard minutes, so you've got a number to compare against when you try more.
James. Okay, in my words. Find out what I really eat, nudge it up slowly with mixed sugars, and test the finish rather than guess. Eighty to ninety on long, hard rides, practise anything more in training first, and nothing past a hundred and twenty.
Lyra. That's it. And if you only remember one line, the new trials show more fuel helps a long, hard finish, not that more is always better.
Lyra. The full article and every study we mentioned are on the WattPlan blog, linked in the show notes.
James. And if you want a plan that's built around the time you've actually got, that's what WattPlan does. Thanks for listening.
Voices produced with ElevenLabs. James's voice is a clone of his own, Lyra is an AI voice, and James reviews every script before it is published.
Two trials published on 7 October 2026 back a high carbohydrate dose on long rides. In Denmark, eating 100 g an hour instead of 60 g left 15 riders 9 W stronger at the end of a 4 hour ride. In Liverpool, 150 g gave no gain over 120 g. Rehearse about 100 g before a hard 4 hour event.
How the Odense trial worked
Oscar Mazza, Kasper Gejl and colleagues at the University of Southern Denmark ran a randomised, double-blind crossover trial, published in Medicine and Science in Sports and Exercise. The 15 riders were trained male cyclists and triathletes. Each did a 15 minute time trial fresh, then after a day on a high-carbohydrate diet rode 240 minutes at 60% of that fresh time-trial power, broken up with high-intensity intervals, and repeated the 15 minute test straight afterwards.
They rode the 4 hours twice, once on 60 g of carbohydrate an hour and once on 100 g, from drinks and gels mixed at 1:0.8 glucose to fructose. Thigh muscle biopsies before and after measured glycogen.
What changed at 100 g an hour?
Both rides cost the riders power, and the higher dose cost less. Time-trial power fell 9.3% below fresh on 60 g an hour and 6.4% on 100 g, and average power in the final test was 9 W higher on the bigger dose (P = 0.028). The abstract gives no confidence interval and no count of riders who failed to improve.
Glycogen fell by a similar amount in both rides, finishing at 192 and 209 mmol per kg of dry muscle (P = 0.56), and blood glucose stayed steady in both. Whole-body carbohydrate burning held up better on 100 g an hour, so the extra fuel kept carbohydrate available to burn late in the ride without saving any muscle glycogen.
Does 120 or 150 g an hour help more?
A second trial in the same journal, published online the same day, tested the doses above 100 g. Martyn, Gonzalez, Stellingwerff, Burke, Morton and colleagues, led from Liverpool John Moores University, had eight trained men ride for 3 hours on 90, 120 or 150 g of carbohydrate an hour in random order, after a first ride on 10 g that doubled as familiarisation. The drinks mixed maltodextrin and fructose at 1:0.8 and carried a carbon tracer, so the team could tell the drink's carbohydrate from the body's own, and each ride ended with a test to exhaustion at 150% of lactate threshold.
The riders burned more of the drink as the dose rose, 1.23, 1.48 and 1.62 g a minute at 90, 120 and 150 g an hour, with wide differences between riders. Total carbohydrate burning stopped rising at 120 g, at 2.6 g a minute on both 120 and 150 g against 2.3 g on 90 g. Total exercise duration, the ride and the test together, averaged 2 h 41 min on the 10 g familiarisation ride, short of the planned 3 hours, so some riders stopped before the ride was over. It averaged about 3 h 5 min on 90 and 120 g, a significant gap, and 3 h 1 min on 150 g, no different from the other high doses. The authors put the best dose for exercise capacity at 107 g an hour, with a 95% confidence interval of 92 to 122 g.
So going from 120 to 150 g an hour raised how much of the drink was burned, but not total carbohydrate use or how long the riders lasted.
| Study | Riders | Doses compared | Main finding |
|---|---|---|---|
| Mazza and colleagues, 2026 | 15 trained men | 60 vs 100 g/h over 240 min | Final 15 min test 9 W higher on 100 g/h, power fell 6.4% against 9.3% |
| Martyn and colleagues, 2026 | 8 trained men | 10, 90, 120 and 150 g/h over 3 h | No more carbohydrate burned in total at 150 than 120 g/h, best capacity at about 107 g/h |
| Smith and colleagues, 2013 | 51 across 4 labs | 10 to 120 g/h, about 160 min in total | Best 20 km time trial at about 78 g/h |
| Podlogar and colleagues, 2022 | 11 trained men | 90 vs 120 g/h over 3 h | More of the drink burned at 120 g/h, no extra glycogen saved, performance not tested |
| Jeukendrup, 2014 review | Review | By event length | About 60 g/h for 2 to 3 h, about 90 g/h for ultra-endurance |
Should you eat 100 g of carbs an hour on a long ride?
Start with the pooled guidance. Asker Jeukendrup's 2014 review puts the target at about 60 g an hour for rides of 2 to 3 hours and about 90 g for ultra-endurance, using mixed sugars at the higher rates. A dose-response study by Smith and colleagues across four labs found the best 20 km time trial at about 78 g an hour, at the end of about 160 minutes of riding.
The Odense trial compared 100 g with 60 g and not with 80 or 90 g, so on its own it shows that the bottom of the usual range costs late power over 4 hours. The Liverpool trial found no measurable difference in exercise capacity between 90, 120 and 150 g after 3 hours, and its best estimate, 107 g, came with a range of 92 to 122 g. The 2022 study of 120 against 90 g didn't measure performance at all.
Our view at WattPlan is that 80 to 90 g an hour stays the right target for most riders on rides over 3 hours, that 100 g is worth rehearsing for a 4 hour event that finishes hard, once your gut is used to it, and that neither new trial gives a reason to go past 120 g. The Odense trial speaks most to riders still eating 60 g an hour on a long sportive or gravel race.
Fuelling is one reason power fades in the last hour, the fade that WattPlan's Durability score reads from your rides of 2 hours or more. A one hour turbo session sits in the range where Jeukendrup's review finds a mouth rinse or a small amount of carbohydrate is enough.
How to move from 60 g to 100 g an hour
- Count what you eat now. Log every bottle, gel and bar on your next ride of 3 hours or more and divide by the hours.
- Build up on long training rides, 10 to 20 g an hour more every couple of weeks, as in our carbs per hour guide.
- Mix your sugars. All three high-dose trials here paired glucose or maltodextrin with fructose, and Jeukendrup's review puts a single sugar at about 60 g an hour.
- Rehearse the finish. End a long ride with 15 to 20 hard minutes and compare the power with the same ride at your usual rate.
- Treat the reminders as a floor. WattPlan's workout player prompts 60 g every 45 minutes, 80 g an hour, on sessions of 60 minutes or more. The amount is fixed, so if you are working towards 100 g, plan an extra gel or bottle between prompts.
What we don't know yet
The Odense trial had 15 men and no women, rode in a laboratory at a set 60% of time-trial power, and compared only two doses. Double-blinding is its strength. The Liverpool trial had eight men, its 10 g ride came first rather than in random order, and its test measured how long riders lasted, not how fast they covered a set distance.
Neither abstract reports stomach symptoms or who funded the work and supplied the drinks. Stomach tolerance matters most at these rates, and Jeukendrup's review gives carbohydrate building up in the gut as the reason to mix sugars when intake goes high.
Sources
- Carbohydrate Intake of 100 vs. 60 g·h-1 During Prolonged Cycling Improves Durability Without Altering Muscle Glycogen Utilization in Male Endurance Athletes, Medicine and Science in Sports and Exercise
- Effects of 10, 90, 120 and 150 g·h-1 CHO on Substrate Metabolism, Exogenous Oxidation and Exercise Capacity: Implications for CHO Personalization, Medicine and Science in Sports and Exercise
- Curvilinear dose-response relationship of carbohydrate (0-120 g·h(-1)) and performance, Medicine and Science in Sports and Exercise
- 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
- A step towards personalized sports nutrition: carbohydrate intake during exercise, Sports Medicine