Get access to everything we publish when you sign up for Outside+.
My co-coach/wife Megan and I recently did an exercise. We went back through all of the best longer-distance performances from our team in the last few months, and we chatted about each athlete. After 11.5 years together, our romance still burns hot.
Athlete attributes varied. Some were former track prodigies, others came from a ski background. Many trained hard for years, others were newer to running. Most worked on speed a bunch, but a few barely did any workouts. It seemed like a semi-random scatterplot, except for one variable. Every athlete was really good at fueling.
That revelation confirmed what Megan and I have learned in coaching. Fitness is important, but in races over 90 minutes, the only way to express that fitness is through consistent energy availability and replenishment.
Here’s why carbs are important as endurance fuel.
Almost all coaches would agree with that statement, but what constitutes proper energy availability and replenishment? How could you even start to test that question, given that fueling is just one variable of many that would be tough to isolate from other athlete attributes?
Announcer voice: Aaaannnnd now, coming to us from Scotland, an all-star study! THE CROWD GOES WILD.
A wonderful November 2021 study in Frontiers of Physiology by Emma Kinrade and Stuart Galloway tested the energy hypothesis with real-word ultra performance data from a 24-hour trail race in Scotland. I absolutely love this study specifically because it did not make an intervention, with the researchers telling participants to behave as they normally would. Instead, the authors asked: What do athletes actually do before and during an ultra event, and how does that correlate with outcomes?
That approach helps get around a problem with nutrition: the gut and brain adapt over time, so short-term interventions have limited utility. Show nutrition professionals and coaches what successful athletes do in an uncontrolled setting, rather than the lab or in theory, and then they can hopefully help direct people toward those long-term adaptations in training.
Before getting to the study details, let’s review the principles of the energy equation.
Note: Always work with a trained expert in nutrition topics. This article is a review of studies that simplifies some complex topics for a general audience, rather than a guide of what any individual athlete should do in practice.
During endurance events, a certain amount of energy comes from fat oxidation and glycogen metabolism. The proportion of fat and glycogen depends on intensity, training, background nutrition, and genetics–very low intensity in an efficient athlete will use more fat, higher intensity will use more carbs. But carbs are important for peak performance at all effort levels.
Athletes ideally start with a full glycogen tank from pre-race fueling patterns, with the 2021 study saying: “Intakes of 8 to 12 grams per kilogram carbs per 24 hours are recommended in the 36 to 48 hours leading up to a prolonged endurance event to ensure well stocked muscle glycogen, with a further 1 to 4 grams per kilogram in a pre-race meal during the final 1 to 4 hours recommended to top up liver glycogen stores.” (For reference, each gram of carbohydrates is 4 calories.) A great overview of the context for these recommendations is this 2016 position statement from the American College of Sports Medicine, which indicates that high daily needs from an endurance training program require 6 to 10 grams per kilogram per day.
Eating plenty before an event? Sign me up! But for many athletes, the hardest part is the next variable: continuing to top off the tank with mid-race fueling, balancing consumption against intensity.
During-event fueling overview
As outlined by a seminal 2014 review article in Sports Medicine, carbohydrate intake based on individual needs/attributes is important for peak performance to balance the energy equation. While that amount varies, a 2011 review in the Journal of Sports Sciences indicated that in endurance events longer than 2.5 hours, some athletes will perform better with up to 90 grams of carbohydrates per hour (360 calories), coming from multiple transportable carbs with different pathways for intestinal absorption.
If your first response to the 90 grams number is THAT’S A LOT, most runners would agree with you. Much of the research has been conducted on cyclists, who don’t have to deal with the pesky impact forces that can cause runner stomachs to be like one of those papier-mâché volcanoes from elementary school. For me, the only difference between a long bike ride and a Cracker Barrel buffet trip is that Cracker Barrel doesn’t appreciate when I stick a pastry down my shorts for later. But running fueling doesn’t come naturally to me in the same way.
In running, individual tolerance plays a far bigger role, as outlined in this 2019 study in Research in Sports Medicine. I love this quote from that study: “Considering gastrointestinal symptoms are common in ultra-endurance activities, effective dietary prevention and management strategies may provide functional, histological, systemic, and symptomatic benefits.” Very pretty words that prevent very gross outcomes. A 2022 study in Frontiers of Physiology found higher rates of gut distress were seen in athletes taking in 90 grams per hour relative to those taking in 76 grams per hour.
An incredibly fun 2020 study from the journal Nutrients did an extreme protocol involving athletes consuming either 120 grams of carbs per hour, 90 grams per hour, or 60 grams per hour during a mountain marathon. Yes, you read that right. 480 calories per hour of carbs! I’d personally need to wash that down with a hydration flask full of Pepto Bismol. Fascinatingly, the athletes in the 120-gram group had lower evidence of muscle damage and fatigue post-race.
The 2021 study we are reviewing today provided a great overview of the literature on fueling and race performance. “Faster/elite runners have been shown to consume more hourly CHO than slower/amateur runners, and finishers reported to consume more than non-finishers … but ultra-runners typically consume lower amounts than recommended, and less than competitors in other ultra-endurance disciplines.”
Faster runners fuel more, whether the event is a road marathon or a trail ultra. It’s an adaptation of the classic song. “You wonder why I’m going slow, and I think I know. My carbs got low, my carbs got low, my carbs got lowwwww.”
I am not sure if that reference is a sign that I am trying too hard, or that I am not trying at all. You can be the judge.
The 2021 study included 7 women and 11 men competing in the G24 24-Hour Trail Race in Scotland, which consists of 4-mile loops with 270 feet of climbing per lap. The weather on race day was a pleasant 60 degrees Fahrenheit. All participants had completed at least one other ultra. In the 48 hours pre-event, all food was weighed meticulously, and for food consumed away from home, participants took photos and sent it to the researchers. Maybe all of this time, people taking photos of their meals for Instagram were just doing it FOR SCIENCE.
During the event, dietary intake was monitored by a recorder assigned to each participant. Participants wore a heart rate monitor or GPS watch, along with a continuous glucose monitor.
The 3 categories of nutrition variables reported are:
- Mean 24-hour intake (pre-race nutrition)
- Intake for 1 to 4 hours pre-race (pre-race meal)
- Analysis of in-race nutrition, including macronutrients, carb distribution, sodium, and caffeine
The study compared the numbers to research-based recommendations, and also correlated those nutrition variables with race performance. They hypothesized that there would be a similar pattern to past research, with athletes undershooting carbohydrate targets relative to those recommended in the literature.
Now is the time that we take a step back. What do you think is going to be the association between nutrition strategies and performance? The results are absolutely striking.
Pre-race total carbohydrate intake was strongly correlated with race distance (r = 0.78), as was carbohydrate intake per kilogram (r = 0.70). There were moderate correlations with total energy intake (including fat and protein) and race distance (r = 0.57). There were no associations between pre-race fluid intake and performance.
I am obsessed with what the researchers did next. They did a sub-analysis comparing athletes that consumed more than 5 grams per kilogram of carbs with those that consumed less. The athletes over 5 grams per kilogram covered an average of 98.5 miles. Those under 5 grams covered an average of 78.0 miles.
A moderate positive correlation was also seen between total carb intake in the pre-race meal and race distance (r = 0.68).
During the race, there were moderate positive correlations between race performance and carb intake (r = 0.65) and carbs per kilogram intake (r = 0.64). Again, the secondary analysis is fascinating. Athletes consuming more than 40 grams of carbs per hour covered an average of 92.2 miles, and those under 40 grams per hour covered 74.7 miles.
Perhaps coolest of all, the researchers fit a regression equation to the data and found that both pre-race and in-race carb intake were helpful in predicting race performance, with an r-squared value of 0.653. WOW! R-squared represents the proportion of the variance of the dependent variable (race performance) that is explained by the independent variable (carb intake). When other variables remained constant:
- Race distance increased by 6.6 miles for every 1 gram per kilogram carbs increase in pre-race nutrition
- Race distance increased by 1.5 miles for every 1 gram per kilogram carbs increase in nutrition during the race
Let’s rapid-fire a few more cool findings.
Participants consumed carbs under the saturation rate of gut transporters that would require multiple carb sources, so the exact distribution of carbs was likely less important.
Athletes wearing heart rate monitors averaged 68% of their max heart rate, corresponding to low to moderate effort. Even at lower efforts, glycogen recovery via carb intake is key for performance.
Glucose profiles showed no association with race performance or carb intake. That’s weird! I love weird things!
Finally, there was no correlation between fluid intake and race performance. That’s surprising! Would that change in hotter races? Perhaps that indicates that individual fluid-loss rates on a 60-degree F day vary more than individual carb needs. We tell our athletes to adjust hydration for personal needs, but always to fuel lots.
The researchers knocked that study out of the park!
Or, since baseball isn’t as common in Scotland, maybe they kicked the ball into the net. GOOOOOAAALLLL!
Like all studies, there were still a few limitations. Nutrition studies are especially vulnerable to confounding variables. Maybe willingness to consume plenty of carbs overlaps with athletes that are more prepared for the event generally, or with approaches to food that are indicative of overall health/endocrine status. In addition, it could be inadvertently measuring genetic variables related to gastric emptying/distress, or even some unexpected association like muscle fiber typology, glycogen storage capacities, and race intensity. However, when read together with the field of literature, it seems pretty unequivocal: carb intake alone isn’t sufficient to drive performance (you still need fitness), but adequate carb intake is necessary to express fitness.
There are three major takeaways for endurance athletes.
One: Make sure you’re eating plenty in the 48 hours before races.
Athletes in this study were way under the 8 to 12 grams carbs per kilogram recommended in the days leading up to the race, with researchers finding that the cohort of athletes over 5 grams per kilogram greatly outperformed those under that threshold. Other studies find similar correlations. For example, a 2011 study in the International Journal of Sports Medicine found that sub-elite marathon performance was significantly improved in athletes that consumed greater than 7 grams per kilogram of carbs in the day before racing.
I think this study will push me as a coach to be more proactive in specifically encouraging athletes to eat more carbs in the days before their races. I always assumed that ultra athletes would fall into appropriate ranges without too much urging, but I think I may have been missing a huge opportunity. Pizza and pasta are not just meal choices, they are lifestyles. Make sure you experiment with your approach prior to long runs to prevent GI system distress.
Two: Practice higher carb consumption during training and races, ideally from multiple sources.
Based on the studies, many athletes grossly undershoot optimal mid-race carb intake. Athletes in this study consumed a mean of 33 grams per hour, compared to the 60-90+ grams per hour usually recommended in the literature, and they underperformed as a result. Another example comes from a 2011 study completed at Western States and published in the Journal of the American College of Nutrition, which found that finishers consumed 66 grams of carbs per hour, while non-finishers only consumed 42 grams of carbs per hour.
The gut adapts over time, and the logistics of taking in fuel can be difficult, so practice increasing carb intake in long runs and training races. My usual recommendation in higher intensity races is a high-carb gel every 20-30 minutes (depending on race distance and intensity), supplemented by sports drink, aiming for 250-350 calories per hour, pushing that total to 400 calories per hour if it works in training.
In longer events, you can get more creative based on your individual preferences and tolerances. I love when athletes are able to practice with complex carb sources, like energy bars, rice balls, potato chips, and yes… even pizza (something I learned from Trail Runner Mag Editor-in-Chief and all-around rockstar Zoë Rom). The same caloric targets are ideal, but the range can be averaged over 2-4 hours to account for a mid-race slice.
Three: Experiment with very high carb intakes if your GI system is reliable during races.
There are massive gains from increasing carb intake into the range suggested by the research. But what about the emerging research finding that very high intakes of 90+ grams per hour may reduce muscle damage? The jury is out–I think most athletes would struggle to hold that much down unless they are very talented eaters and it isn’t hot.
But, I need to check my bias. Maybe those “talented eaters” are just “athletes that strategically practice fueling to be able to consume more fuel over time.” This is what Coach Megan and I see. For example, Zoë can consume up to 400 calories per hour, and she is one of the most talented ultra runners I know. After a blazing fast performance at the North Face 50 Miler in 2017, I calculated Clare Gallagher’s energy intake at just over 400 calories per hour.
So consider trying it. Just make sure you are taking in carbs from multiple sources to avoid overwhelming gut transporters (see this 2015 review in the Journal of Strength and Conditioning Research). And bring toilet paper.
I am not a nutritionist. My co-coach Megan is a doctor but not a nutrition expert. However, from reading the research and seeing tons of athletes’ trajectories over time, we have come to a simple piece of nutrition guidance.
Eat enough, always. Eat too much, sometimes. Eat too little, never.
The blossoming research on carb intake before and during races might require an addendum.
And when it’s time to compete, eat a metric crapton of carbs.
David Roche partners with runners of all abilities through his coaching service, Some Work, All Play. With Megan Roche, M.D., he hosts the Some Work, All Play podcast on running (and other things), and they wrote a book called The Happy Runner.