Train low, compete high


As of September 2017, new Sweat Science columns are being published at Check out my bestselling new book on the science of endurance, ENDURE: Mind, Body, and the Curiously Elastic Limits of Human Performance, published in February 2018 with a foreword by Malcolm Gladwell.

- Alex Hutchinson (@sweatscience)


One of the most interesting developments in sports nutrition over the last few years is the “train low, compete high” concept — the idea that purposely doing some of your training when your glycogen stores (the main form in which your body stores carbohydrate for exercise) are low can boost your performance when you eventually compete fully loaded. An initial study in 2005 found that subjects doing half their training in the “low” state ended up with higher glycogen levels and longer time-to-exhaustion. But there were some questions about how well those results would translate to real-life — for example, whether trained athletes would experience same effects as the untrained volunteers in that experiment.

A new study [LINK FIXED] from Asker Jeukendrup’s group at the University of Birmingham, just posted online at Medicine & Science in Sports & Exercise, tackles some of these questions, with interesting results. The set-up was two groups of seven cyclists, training six days a week for three weeks. Both groups alternated 90-minute aerobic sessions with intense interval sessions of 8 x 5:00 with 1:00 rest. The control group did the sessions on alternate days, while the “low” group trained just three days a week, starting with the aerobic session to deplete glycogen, then doing the intervals an hour later without refuelling.

As hypothesized, the “low” group learned to burn more fat instead of carbohydrate — a physiological strategy that some experts think might allow your body to last longer before running out of glycogen. On the other hand, there was no difference in time-trial performance at the end of the study. You might think this means that the strategy was ineffective, but there’s an added wrinkle: not surprisingly, the “low” group managed a much feebler effort in their interval sessions (since they were so depleted), but still managed to improve by the same amount on the time trial. This means that the key variable in a training session isn’t how fast you go, but rather what stimulus signal you’re sending to convince your body to adapt.

In this regard [the authors write], we suggest that the additional “stress” of training with low glycogen compensates for a slight reduction in physical performance during training.

There was also a penalty to pay for the increased fat burning: the “low” group didn’t increase their carbohydrate-burning abilities as much as the control group:

This also suggests that training with low muscle glycogen may be counterproductive for athletes who compete in high intensity events where CHO oxidation plays a significant role in performance, and that this type of training may be more suited to preparation for ultra-endurance activities.

For now, it’s a case of “more studies needed” — although there’s no doubt that coaches and athletes are already experimenting with these ideas. It may be that trial and error will sort out some useful approaches before we really understand why they work.

[An unrelated note: I’m heading out on a hiking trip tomorrow morning, so I’m unlikely to be able to update the blog for the coming week. Next weekend, I’ll be watching the Sydney triathlon — the first stop on this year’s world championships series, and the first race on this course since Simon Whitfield’s immortal gold-medal performance in 2000. Can’t wait!]

6 Replies to “Train low, compete high”

  1. It seems to me there are too many other variables in this study. Training twice a day, three days a week and taking the other days off is a significant change in training frequency. You get a bigger training push and you get a bigger recovery. So that may also have something to do with it. Would it not have made more sense to control the diet of the participants, rather than the training load? Like have them do their intervals without eating or something, and make sure they don’t make up for it after the aerobic session? Too much noise, I think, for this study to be worthwhile.

  2. Yes, there are certainly some potential confounders. That being said, they were trying to test the model that athletes have actually been using (i.e. doing a depletion run followed directly by an intense interval session), so they had to have a difference either in training frequency or training volume. They couldn’t hold EVERYTHING constant, because the two approaches are quite different. The main strength of this study was the extremely careful direct measurement of variables like glycogen stores (using invasive muscle biopsies) and fat metabolism (using stable isotope tracers). So unlike previous studies, they weren’t just looking “do the subjects get faster?”, they were able to answer “why or why not?”

    The other approach I’ve heard athletes and coaches discuss is training first thing in the morning without refuelling (which I think means you start with depleted liver glycogen but not necessarily muscle glycogen?). That seems like a less disruptive and perhaps lower-risk procedure.

  3. That is a very cool result! But how significant is it that there was no change in performance? One would like to know the length of the time trial (I cannot access the paper through the U Sydney link.)

    The time trial format is perhaps necessary for experimental measurement, but “real” road races last much longer than any time trial – yesterday’s Tour of Flanders was almost 6 1/2 hours of hard riding. Perhaps glycogen sparing would be more beneficial in that context. On the other hand, “real” training stimulus is quite different too – 90 minutes is a pretty short aerobic session for a cyclist. Would “starved” interval training be a more efficient substitute for the usual long training block? Or are there physical parameters beyond glycogen use that can only be modified by sustained training stimulus?

  4. Do you have information on how to training using a low heart rate during the training? I’ve recently talked to friends that are using this technique. Long duration … low heart rate.

    How do you do it? What are the benefits?

  5. Those are good questions, Phil — and the only one I can answer satisfactorily is that the performance consisted of 60 minutes of steady-state riding (at about 70% of VO2max) followed by a time trial that was designed to last roughly 60 minutes.

    The authors do indeed speculate that the technique may be more relevant to longer events. But there are problems with that too. As Louise Burke of the Australian Institute of Sport wrote in a paper on the train-low-compete-high concept back in 2007:

    “Many people see endurance and ultra-endurance sports as events involving only submaximal exercise; hence, the interest in increasing fat utilization and conserving limited endogenous carbohydrate stores. However, the strategic activities that occur in such sports—the breakaway, the surge during an uphill stage, or the sprint to the finish line—are all dependent on an athlete’s ability to work at high intensities that are carbohydrate-dependent.”

    Ces, I’m not really sure what you mean by training with a low heart rate. Do you mean keeping your effort low, or using some other way of keeping your heart rate low even when your effort is high?

Comments are closed.