I posted last month about the “train low, compete high” concept, in which you do a depletion workout to empty your glycogen stores, then do a hard workout while running on empty. The idea is basically the nutritional equivalent of running with a weighted vest — it makes training harder, and perhaps forces your body to adapt to the tougher conditions so that when you do fuel up properly, you get an extra boost. But results so far have been ambiguous: your body does respond in several ways, including learning to burn more fat instead of carbohydrate, but there’s no good evidence that it actually improves performance. And it’s very hard and makes you feel like crap.
A kinder, gentler version of “training low” is doing your workout before breakfast, without restocking the glycogen that you’ve burned overnight (your liver glycogen stores drop by about 50% while you sleep). There have been a few studies on this regime, and the most recent is now in press at the Journal of Science and Medicine in Sport. Researchers in New Zealand had 14 cyclists perform a four-week training program, five mornings a week, starting with 25 minutes per session at 65% of VO2max and building up to 100 minutes per session. Half of them had a “standard cereal breakfast” an hour before working out, while the other half had the same breakfast shortly after working out.
The results were pretty striking. The fasted group increased the amount of glycogen stored in their muscles by 54.7%, while the breakfast group increased by just 2.9%. The fasted group also increased their VO2max by 9.7%, compared to 2.5% in the breakfast group. The other interesting finding is that men and women responded differently to some of the training adaptations: men seemed to benefit more from the fasted training, while women seemed to benefit more from the fed training.
Some caveats: First, it’s a small study, so the margins of error are high. Second, the subjects were untrained — athletes who are already training definitely won’t boost their glycogen stores by 50% or their VO2max by 10%, so it’s possible that the differences may disappear entirely. Third, actual exercise performance wasn’t measured as an experimental outcome. The literature is full of bright ideas that change some parameter in lab tests but don’t actually make you faster when the chips are down. Finally, the training program isn’t one that anyone would implement in the real world. It’s highly unlikely that doing ALL your training in a fasted state is the optimal approach.
Still, the results are interesting. They raise the possibility that mixing in an occasional fasted workout, perhaps once or twice a week, could provide your body with a different sort of stimulus that might prompt some useful adaptations. And this no-breakfast approach seems a lot more realistic to me than the idea of doing a one-hour depletion run before a hard workout.
(Thanks to Steve Magness for pointing out the study.)