To start this article, I should check my bias. I’m a fan of carbohydrates. I want the athletes I coach to fuel early and fuel often. The best man at my wedding was a deep-dish pizza. I am in the pocket of Big Cookie.

But there is some nuance to my bias. The one disclaimer I always provide is that for some male athletes, doing non-excessive amounts of carbohydrate-restricted endurance training may improve fat oxidation, with possible benefits for performance in long events.

A new study may indicate that I don’t need the disclaimer—time to let my bias flow freely like the syrup that engulfs my pancakes.

RELATED: A New Study Challenges Conventional Wisdom on Protein Consumption

Study Overview

The study in the journal Nutrients asked a devilishly direct question: how does periodization of carbohydrate intake impact performance, substrate utilization, and body composition in well-trained athletes over five weeks? The design was the epitome of provocative science–take advanced male athletes who are already training hard, train them harder, and manipulate only the carbohydrate variable in ways backed by previous research theories.

Seventeen male cyclists participated in the study, all training 15 to 20 hours per week and competing at the under-23 age level before the study, with VO2 max levels averaging 70.8 mL/kg/min. Nine of these young beasts were assigned to the carbohydrate periodization group, and eight to the high carbohydrate group. If I were in this study, I would sell two souls (my soul and a soul to be named later) to be in the high-carb group.

Both groups completed performance tests before and after the intervention, with the same training plan across the five weeks. And that training plan was a doozy. They did a few hard workouts every week, high training volume over six days with one rest day, and two gym sessions. The training plan in Table 1 is so good that I’m thinking about copying it, putting it behind a paywall, and creating a media site that slowly loses revenue due to shifting media consumption patterns, until we announce mass layoffs and a pivot to video.

The only difference between the groups was the fueling strategy. The high-carb group fueled every session. The periodized carb group performed 13 low-carb availability training sessions on easy days, involving restriction of carbs from lunch the day before until after the easy ride the next morning, with meals focused on protein and fat. That group also did “medium” carb refueling on some days where lunch, snacks, and dinner had lower carbs, with the exact breakdown depending on the next training session.

RELATED: 50 to 100 Mile Training Plan For Lower Mileage Athletes

Carbohydrate Study Findings

Now is a good time to take a step back so we can all check our biases. What do you think would happen with this type of study design? I would predict that the group practicing carbohydrate restriction would have improved fat oxidation, at the very least, and possibly body composition changes. Do you agree?

Well, if you agreed, you’d be wrong. Frankly, that’s on you. I’m George Washington, leading his brave troops across the Delaware River and into an abandoned chainsaw warehouse.

The big finding is that there was no major difference between the groups. The high carb and periodized carb approaches both led to improvements in performance and output at maximal lactate steady state (when lactate production equals lactate clearance, corresponding to a moderately hard effort), with no significant differences between the groups in carb or fat oxidation at max lactate steady state, or between-group changes in body composition. In other words, the benefits were due to the training plan, not due to the fueling plan.

This study represents a major leap forward in the literature because it aimed to isolate training effects from fueling effects, a confounding variable in past studies, and it was conducted on advanced athletes over a longer time horizon.

RELATED: The Wild, Uncertain Science of Post-Exercise Ketones

Study Context

Three primary training approaches limit carbohydrate availability. The first is “sleep-low,” where athletes do a glycogen-depleting training session, do not refuel with carbohydrates before bed, and then do a moderate training session with low carbs the next morning. With that approach, I would sleep two hours and I would have dirty dreams about ice cream.

The second approach is fasted exercise, involving training after not eating overnight or in the morning, without carbs during the training session. Fasted training has been the wind that has propelled 10,000 bro science ships directly into icebergs.

The third is twice-a-day training, when the second session includes low glycogen availability. I actually like doubles for athletes (without intentional depletion), so I won’t make a disparaging joke about this one. Objectivity doesn’t drive ratings.

Past research identified possible benefits for adaptation and performance. For example, a 2010 study in Medicine and Science in Sports and Exercise had one group of cyclists alternate high-intensity sessions one day with low-intensity sessions the next day, all with full carbohydrate fueling. The other group completed low-intensity followed by high-intensity on the same day with a day of rest between, so that their high intensity was with low glycogen stores. The low glycogen group put out less power in the high-intensity workouts but had increased fat oxidation, which could theoretically improve performance where glycogen availability is a primary limiter, like marathons and above.

A 2016 study in the same journal used the “sleep-low” approach, where the intervention group did an evening interval session to deplete glycogen, followed by sleeping with low glycogen (ice cream sex dreams), followed by a morning training session without carbohydrate fueling. The control group did the same training structure, but with high carb availability. The sleep-low group had better cycling economy and lost body fat after the intervention, in addition to having 300 percent more microwaved pints with holes cut in the bottom.

Those studies (plus another from 2016 in Nutrients) helped support a general thought process that carbs were essential for fueling high performance, but periodizing their availability could help some male athletes become more economical and adapt better. A 2018 article in Cell Metabolism put it all together, noting that training with low glycogen could enhance the adaptation response via cell signaling and protein expression changes, along with spurring the fat oxidation benefits that had long been theorized.

However, there is a list of other studies that didn’t find any performance benefit from similar interventions. In fact, a 2021 meta-analysis in the Journal of the International Society of Sports Nutrition found “no overall effect of CHO periodization on endurance performance compared to control endurance training with normal (high) CHO availability.”

RELATED: The Science of Hemoglobin Mass and Endurance Performance

Study Implications For Runners

The authors of today’s study felt that most of the past studies did not have a long enough time horizon, usually one to three weeks. Plus, the studies should measure substrate utilization based on lactate threshold, rather than substrate utilization at a certain percentage of peak power output, since peak power output percentages correlate worse with actual physiological zones. Reading the literature together with their findings, the authors make a powerful conclusion: “Periodized CHO intake does not show superior metabolism adaptations compared to a high-CHO diet and may also impair training sessions and following training improvements due to lower glucose oxidation and carbohydrate utilization, which may affect the duration and quality of training.”

“Alright,” a skeptical reader might say, “The studies are mixed, and today’s study finds no difference. What’s the big deal?” My take is that these findings are so important because of how restrictive practices may risk causing long-term health issues that none of these studies can see due to the practical limitations of shorter study interventions. As the authors state, “It should be considered that low energy availability during prolonged periods may cause adverse effects that will eventually compromise sports performance and health.”

One of the fascinating parts of these studies is how they control overall energy availability across days, so athletes theoretically consume the same total calories relative to their expenditures. But metabolism and the endocrine system are more complicated than that. Within-day glycogen deficits may be one of the main drivers of Relative Energy Deficiency in Sport (RED-S). For example, a 2017 study in the Scandinavian Journal of Medicine & Science in Sports on female athletes classified deficits as any one-hour period with more than a 300-calorie deficit. Even when athletes ate the same amount in a given day, the athletes with more within-day deficits had higher rates of menstrual cycle disturbances, suppressed resting metabolic rates, lower estrogen levels, and higher cortisol levels. Similarly, a 2018 study in the International Journal of Sport Nutrition and Exercise Metabolism looked at the question for male athletes, finding that the more 400+ calorie deficits athletes had within a single day, the lower their metabolic rate, the higher their cortisol, and the lower their testosterone: cortisol ratio.

It’s not just about getting enough fuel every 24 hours but keeping the fire burning hot throughout the day. As outlined by a 2019 review study, low energy availability within or across days can harm the endocrine system in relatively short periods for both male and female athletes. As stated by a 2023 review in Nutrients, the harms faced by female endurance athletes from low carbohydrate availability may be independent of low energy availability. “It is proposed that [low carbohydrate availability] may not only contribute to [low energy availability] but also have independent health and performance consequences in athletes.” Thank god these studies rarely include female athletes, who generally operate with a narrower margin for error for the endocrine system. Injury rate could skyrocket too, particularly for runners who deal with more impact than cyclists.

That’s layered on top of the practical difficulty of nailing these carbohydrate restriction approaches in a science-driven way. Given that the risks of within-day deficits and low energy availability are so high, there better be a really good justification for playing a game of craps with long-term health. And as today’s study shows, there may not be a good justification. Perhaps, if we measured the carbohydrate periodization group a few months later, the neutral results would turn into a crapstorm of high cortisol, low testosterone, and torpedoed performance. If the study included female athletes, we’d probably see higher rates of menstrual cycle disturbances, lower estrogen, and worse bone density.

Or, to put it another way, athletes who are training well already have plenty of stress to adapt without introducing risky, potentially disastrous metabolic stresses. It’s fine to do doubles, mix up carb intakes, and not always practice extreme fueling approaches of 90+ grams of carbs per hour. You can even go out in the morning without a big breakfast every once in a while. All of this has nuance and given that every study doesn’t have the same results, different daily approaches can likely work for different athletes, at least for a short period of time.

But across training cycles, make sure you fuel the work you’re doing, and don’t mess with intentional carb restriction unless an evidence-based practitioner who recognizes the nuance is helping you on that journey. The low-carb guy lecturing you at the gym or in the comments about fasting or restriction can keep his opinions where they belong: to himself.

Based on the trajectory of the research, I wouldn’t be surprised if we have a new place for him to store his opinions with five to 10 more years of studies on endurance athletes: in the trash.

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 answer training questions in a bonus podcast and newsletter on their Patreon page starting at $5 a month.