How Many Times Should a Runner Run?

Declan Connolly, Ph.D
Posted June 6th, 2008

How many times a week should I exercise? How many days should I rest? How many times should I exercise each day? While the average exerciser will benefit from a standard exercise prescription of 5 days a week, that is not so for the competitive athlete. These are important questions with unclear answers for the competitive athlete. Given the volume of information that exists, we have come to the point where we pick up bits and pieces of information from the different disciplines. For example, if you’re just interested in basic health and fitness, we’ve been told to aerobically exercise 3-5 times a week, do resistance work 1-2 times a week, do interval work once a week. So what is the physiology behind all this? And more importantly, what does the science tell us about the success of different approaches?
The first thing we need to do is sort through the information and ultimately distinguish between health, fitness, and performance. General exercise guidelines, such as those published by the American College of Sports Medicine, are written for the large majority of the population who are inactive, not competitive, and who, for the most part, don’t exert themselves too much. These guidelines are overly conservative both in frequency and intensity to the degree that people still don’t get enough exercise. These basic guidelines prescribe 3 to 5 times a week for 30 minutes. So while these levels of exercise will accomplish basic health and fitness for the average Joe, they will do little for the competitive athlete.
The competitive athlete has several body systems to consider and thus really needs to aim for between 8 and 10 exercise sessions per week. This includes aerobic, anaerobic, and resistance work. The question then arises about how to divide up these sessions. When VO2 max is a primary concern, science suggests that a training frequency of 5 to 6 times per week tends to yield the optimum results. Six or more aerobic sessions appears to promote overtraining.
This relationship between frequency of training and performance is grounded in recovery. For many years we have known that muscles require at least 24 hours to recover from strenuous workouts, and that positive adaptation requires the muscles to have recovered or “healed” before optimal response from a subsequent bout is realized. This is especially true with resistance training. However, some recent evidence suggests that a controlled overload of the muscle forcing it to work in an un-recovered state may be more beneficial. Recent work from Scandinavia by Hansen and colleagues in the Journal of Applied Physiology (2005) indicates improved performance in exercised limbs forced to exercise twice a day versus once per day. In this study, the investigators had subjects exercise six times in a seven-day period. In one group, subjects exercised two times a day with a rest day in between and then two rest days after day five. In the other group, subjects exercised five days in a row and then again on day eight (both groups ultimately performed the same amount of work in the same time). The idea was to have one muscle group exercise a second time under conditions of low muscle glycogen versus being recovered 24 hours later. Incidentally, in this study, the different groups comprised the two different legs of the same people. So the bottom line here was that muscle groups either exercised six times in three days or six times in eight days. I found the results interesting and surprising. Prior to the program, both legs were similar in maximum power (watts), time to exhaustion at 90 percent of max power (minutes), and in total work performed (kj). Following the 10 weeks of training where each session lasted one hour the summary data were as follows:
Pre-Training Post-Training
1x/day 2x/day 1x/day 2x/day
Max Power (Watt) 77 74 106 107
Time Exhaustion (min) 5.6 5.0 11.9 19.7
Total Work (KJ) 25 22 69 114
So one can see the improvement in performance variables as measured by total work were much better in the leg that trained two times/day for three days, even though max power stayed similar, indicating potential advantage in endurance performance. Cyclists, runners, swimmers, cross-country skiers, and rowers should take note! One other point worth mentioning is that ending lactate levels were also greater in the twice-a-day trained leg, suggesting greater tolerance of absolutely higher-end work. What is interesting about endurance athletes is that they often become obsessed with end-point numbers, such as VO2 max. In reality, it is power output at sub-maximal levels that is arguably more important. If top-end values were the bottom line, then the guy with the highest VO2 max would always win and this is not the case—often it’s the guy with a high VO2 max who also has a high anaerobic threshold. This is one reason why I often discuss with athletes their testing results and whether the focus should be improving anaerobic threshold power output or max power output (which is at VO2 max). The law of diminishing returns comes into play a lot in this arena.
So, how do we interpret this data? The authors of the study suggest that the low glycogen environment created by the first exercise bout actually stimulates greater adaptation, and certainly the data supports this. The authors further suggest that this approach induces an increase in the transcription rate of the genes that induce the muscle adaptation. So, should we train two times a day? We must be careful, as repeatedly training with low glycogen has also been shown to cause overtraining. To generically adapt this training methodology without more information could be detrimental. However, this study did permit a lengthy recovery period in that a day off is given following one day of two workouts, thereby, allowing a degree of recovery. So, a valid question is how much of a role did the recovery day every other day contribute in the end. I suspect it did play an important role. In the big picture, this data is refreshing in that it provides some support for twice-a-day workouts, which are a common practice for many athletes at various times of the year.
We need more work to support these suggestions, but for now we have some anecdotal support for real-life practices of more accomplished athletes. However, as with all situations, rest and recovery are crucial for optimum performance, and constant evaluation and assessment are necessary when emphasizing a higher frequency of training volume regime.
Declan Connolly is a professor of exercise physiology and kinesiology at the University of Vermont, a fellow of the American College of Sports Medicine, a certified strength and conditioning specialist, past president of the New England American College of Sports Medicine, and consultant to the NHL, NFL, IOC, and many others. Check out for more information, training tips, articles, etc.