Training before breakfast has become one of the most talked about strategies in endurance sport. The promise is appealing. Skip the porridge, head out the door, and supposedly teach your body to burn fat better, build more mitochondria and become a more efficient athlete. A survey of nearly 2,000 endurance athletes found that 63% perform at least some sessions in the overnight fasted state, most often citing fat burning adaptations, gut comfort or simple convenience as the reason.

But anecdotes aren’t evidence. So what does the evidence actually say about fasted training and endurance performance? And do the metabolic adaptations that fasted sessions promote, particularly increased fat oxidation, translate into faster race times?

What we mean by “fasted training”

In the research literature, fasted almost always means an overnight fast of at least 8 to 10 hours, with training performed before any food (water and sometimes black coffee permitted). It is not a multi day fast, and it is not the same as a low carbohydrate diet. The most common scenario is a morning session done before breakfast.

Fasted training and acute performance: the fed state usually wins for longer sessions

The majority of literature in this area relates to acute, single session performance. The most comprehensive meta analysis to date pooled 46 studies and found that eating before exercise improved continuous aerobic performance lasting longer than 60 minutes. For shorter aerobic sessions (under 60 minutes), there was no overall difference between fasted and fed conditions. 

This pattern fits what we know about substrate use. As exercise extends past about an hour, glycogen availability becomes increasingly important, and starting topped up with carbohydrate helps maintain pace. Several studies show meaningful improvements in time to exhaustion when athletes consumed carbohydrate before prolonged cycling or running compared with the fasted state.

For shorter, higher intensity work, the picture is more mixed. A recent trial in trained cyclists found no difference in average power during high intensity intervals when the same session was performed fasted, after a meal rich in carbohydrate, or after a meal rich in protein. So if the session is short and sharp, going without breakfast probably won’t tank your numbers, although hunger and perceived effort can vary between individuals.

Fasted training and chronic adaptations: less of a clear cut win 

The case for fasted training is built less on acute performance (which it doesn’t help) and more on the idea that it produces superior long term adaptations. The mechanism makes sense in principle. Training with low carbohydrate availability amplifies signalling through molecules called AMPK and PGC-1α. These act as cellular fuel sensors, switching on when energy is running low and telling muscle to build more mitochondria (the tiny power plants inside muscle cells), burn more fat, and use fuel more efficiently.

The most cited study supporting fasted training is a six week training intervention. Both the fasted group and the carbohydrate fed group improved VO₂max (around 9%) and 60 minute time trial performance (around 8%) by the same amount. The fasted group did show greater increases in two important mitochondrial enzymes that help muscle process fat and carbohydrate; citrate synthase (up 47%) and β-HAD (up 34%), along with more use of fat stored inside the muscle during exercise.

But things don’t quite add up. Amplified molecular signals, identical performance gains? A separate six week interval training study in overweight women found no difference in body composition or muscle oxidative capacity between fasted and fed training. And a meta analysis specifically of periodised carbohydrate restriction protocols in trained endurance athletes concluded there was no overall performance benefit compared with normal high carbohydrate training.

In short, fasted training reliably amplifies the cellular signals associated with endurance adaptation, but those amplified signals don’t appear to produce a measurably better athlete on the bike, the road or the trails.

Fasted training and fat oxidation: does the adaptation translate?

This is where things get interesting, but as always, there is nuance to be considered.

Fasted exercise definitely increases fat oxidation acutely. A meta analysis of 27 studies confirmed that aerobic exercise performed in the fasted state burns significantly more fat than the same exercise performed fed. And cross sectional data show that peak fat oxidation tracks closely with endurance performance, for example with 30 minute time trial power in trained male cyclists.

So higher fat oxidation looks like a marker of a good endurance athlete. But here is the catch. Correlation isn’t causation. The athletes who are fastest also tend to have the highest fat oxidation rates, the biggest mitochondria and the highest VO₂max, all because they train a lot. Increasing fat oxidation in isolation doesn’t appear to be the lever that improves race times.

The more important fuel for endurance race pace efforts is still carbohydrate. Research on low carbohydrate, high fat (LCHF) diets has shown that even when fat oxidation rates almost double, performance can be impaired because the muscle’s ability to use carbohydrate at high intensities is downregulated. For endurance events that involve any meaningful intensity, such as climbs, breakaways, race winning surges or the final 10K of a marathon, carbohydrate availability still matters more than fat burning capacity.

Put simply, a higher fat oxidation rate is a useful adaptation that lets you save more glycogen for when it counts. But on its own it doesn’t make you faster, and chasing it at the expense of session quality or daily fuelling can backfire.

The risks of fasted training

Fasted training isn’t risk free, especially when it becomes a default rather than a tool that is carefully integrated into a training programme.

Session quality. Hard intervals and long aerobic sessions performed underfuelled tend to feel harder, produce lower power outputs and reduce the training stimulus. If you trade a high quality fed session for a moderate quality fasted one, you are probably losing more than you gain.

Energy intake and recovery. A network meta analysis showed that while fasted exercise can reduce 24 hour energy intake, it also significantly increases hunger and reduces total energy expenditure. For endurance athletes that combination can be a problem rather than a benefit.

Bone health and REDs. A recent study in elite endurance athletes reported a 38% prevalence of fasted training and a 44% incidence of bone injury, with a possible association between the two. Female athletes appear particularly sensitive. Low energy availability disrupts hormonal function, bone metabolism and recovery, and habitual fasted training is one of the patterns flagged in Relative Energy Deficiency in Sport (REDs) frameworks.

So should endurance athletes use fasted training?

The most defensible position the current evidence supports is this. Fasted training is a tool, not a philosophy.

If you are an endurance athlete and you want to use it, here is a sensible framework:

• Reserve fasted sessions for easy, low intensity aerobic work lasting up to about 60 to 90 minutes. This is where the performance cost is minimal and any signalling benefit is most plausible.
• Always fuel for hard sessions, long sessions and key race specific work. The acute evidence is clear that prolonged or high quality work is supported by carbohydrates eaten before exercise.
• Don’t train fasted more than two or three times per week, and never on consecutive hard days.
• If you are female, an adolescent or masters athlete, returning from injury, or have any history of underfuelling, low energy availability or disordered eating, fasted training is probably not the right tool for you. The downside risk is too high for a marginal, and unproven, performance upside.
• Watch the markers that matter: session quality, recovery, sleep, mood, body composition trends, training data and (for female athletes) menstrual cycle regularity. If any of those trend negatively, the approach isn’t working for you.

The summary is that fasted training reliably changes some biochemistry, sometimes changes acute performance for the worse, and rarely if ever changes long term performance for the better. For most endurance athletes most of the time, the bigger performance gains sit in better daily fuelling, well judged carbohydrate availability around hard sessions, and getting their fuelling right on race day. Fasted training can sit somewhere in that mix. It just shouldn’t be the headline act.