The trickiest part of strength training, for most endurance athletes, is getting started. There are plenty of good reasons to do it, both for health and for performance. But there鈥檚 an important wrinkle that doesn鈥檛 get much attention: when should you stop?
The practice of tapering鈥攁 short-term reduction of training before an important competition鈥攊s common practice. A big concluded that the best approach is a two-week period during which you gradually reduce training volume by 40 to 60 percent without altering the frequency or intensity of your workouts. More recently, researchers have suggested that a 鈥,鈥 avoiding stressful or mentally fatiguing activities before a big race, could be useful. But how and when do you taper your strength training routine?
in the journal Sports offers a little bit of data, from a team led by Nicolas Berryman of the Universit茅 du Qu茅bec 谩 Montr茅al. It鈥檚 a belated follow-up to that looked at the effects of strength training on running economy, which is a measure of how much energy you require to sustain a given pace. That study, like many similar ones, found that adding certain types of strength training did indeed make runners more efficient. The new study reanalyzes data on a subset of the original subjects who underwent further testing four weeks after they鈥檇 stopped the strength training intervention.
It鈥檚 worth recapping a few details of the original study, which involved just one training session a week for eight weeks. One group did a 鈥渄ynamic weight training鈥 routine of concentric semi-squats using a squat rack, exploding upward as quickly as possible. The other group did plyometric training, performing drop jumps by stepping down from a box and immediately bouncing as high as possible. The box height was 20, 40, or 60 centimeters, chosen based on what height produced the highest jump for each subject. In both groups, they started the first week with three sets of eight repetitions, with three minutes of rest between sets, and eventually progressed to a maximum of six sets.
In the original study, the plyometric group improved their running economy by an average of seven percent, the dynamic weight training group improved by four percent, and a control group that didn鈥檛 do either saw no change in their running economy. That鈥檚 consistent with other studies, which have found a range of two to eight percent improvement in running economy from various forms of strength training. For context, recall that Nike鈥檚 Vaporfly 4% shoes upended the running world because they offered an average running economy improvement of four percent. Strength training is legit, at least among the recreational athletes in this study.
So what happens four weeks after the subjects stop their strength training? Only eight subjects completed this follow-up (four from the plyometric group, four from the dynamic group), so they鈥檙e all lumped together for this analysis. These subjects maintained their newly improved running economy, and lowered their 3,000-meter race time even further.
Here鈥檚 a graph of some of the key outcomes. The triangles show changes from baseline after eight weeks of strength training; the circles show changes from baseline after the additional four weeks with no strength training.

Running economy (shown here as 鈥渆nergy cost of running鈥) was essentially unchanged by the four-week taper. Aerobic capacity (shown here as 鈥淰O2peak,鈥 which is basically the same as VO2 max) actually seemed to regress a bit during the taper, which is surprising and may just be a fluke. In contrast, 3,000-meter race performance ends up 2.4 percent better after strength training and 4.4 percent better after the strength training taper鈥攚hich is precisely the kind of extra bump you hope to get from a taper.
(We鈥檒l ignore 鈥渁erobic endurance鈥 in the graph above. It鈥檚 defined as the ratio of peak treadmill speed in the VO2 max test to 3,000-meter race speed. I鈥檓 not clear what its significance is, but it didn鈥檛 change during the taper anyway.)
The authors go out of their way to emphasize all the caveats here, particularly the small sample size of eight subjects. We also don鈥檛 really know how things were changing during the four-week taper. Maybe the best performance of all was actually one or two weeks after the cessation of strength training. Still, the results suggest that the running economy boost you get from strength training鈥攚hich is widely considered to be the main performance benefit for endurance athletes鈥攕ticks around for at least four weeks without any additional strength training. If nothing else, this suggests you can err on the side of caution in backing off your strength routine fairly early.
The question that鈥檚 left hanging is whether the extra boost in 3,000-meter performance after the taper (despite running economy staying the same and VO2 max getting worse) is just a statistical quirk, or whether it鈥檚 something real. There鈥檚 simply not enough data here to draw conclusions, but there are that there might be an 鈥渙vershoot鈥 effect that supercharges your fast-twitch muscle fibers a week or two after you stop your strength training routine. That鈥檚 fodder for future research鈥攂ut even without an overshoot effect, these results add support to the idea that you can and probably should taper your strength training at least a week before a big race.
For more Sweat Science, join me on and , sign up for the , and check out my book .