The instinct in endurance training is straightforward: to get faster, run faster more often. A 2007 study by Jonathan Esteve-Lanao, Carl Foster, Stephen Seiler, and Alejandro Lucia, published in the Journal of Strength and Conditioning Research, tested that instinct directly against an alternative. The researchers took 12 sub-elite Spanish runners and randomly split them into two training groups. Both groups did the same small amount of true high-intensity work, sessions run above the second ventilatory threshold, at roughly 8 percent of total training. What differed was how the rest of the training was split. One group ran about 81 percent of their training at an easy, conversational pace, with only around 12 percent at a harder, sustained “threshold” effort. The second group ran less of their training easy, about 67 percent, and more of it at that harder threshold pace, around 25 percent.
After the training period, both groups improved their 10.4-kilometer race times, but by different amounts. The group that ran mostly slow improved by an average of 157 seconds. The group that spent more time at moderate-hard effort improved by 122 seconds, a smaller gain, and the difference between the two groups was statistically significant. Running more of their training slowly, and correspondingly less of it at a demanding-but-not-quite-hard pace, left the first group faster on race day than the group that spent more time working moderately hard.
The finding fits a pattern researchers had already noticed by watching what elite endurance athletes do on their own, rather than what they’re told to do in an experiment. In 2006, Seiler and Glenn Øvereng Kjerland published an analysis in the Scandinavian Journal of Medicine & Science in Sports of the training logs of nationally competitive junior cross-country skiers. Without being assigned any particular training plan, these skiers had naturally organized their training into what researchers call a polarized distribution: roughly 75 percent of sessions run clearly easy, another 15 to 20 percent run hard, and comparatively little time spent in the moderate zone in between. The skiers weren’t following an 80/20 rule because a coach had assigned it as an experiment. They’d arrived at something close to it on their own, which is part of why the pattern drew researchers’ attention in the first place.
The proposed mechanism, though still an active area of research rather than settled fact, centers on what a lot of easy running does to the body over time that hard running doesn’t. Sustained low-intensity effort is understood to build mitochondrial density and improve the body’s ability to burn fat as fuel, developing what exercise physiologists call an aerobic base, without accumulating the fatigue and injury risk that comes from spending large amounts of time at a harder, more demanding pace. Running hard remains part of the picture in these training distributions. The point isn’t that hard running does nothing. It’s that padding out the rest of the week with moderate, not-quite-hard effort appears to cost more in fatigue than it returns in fitness, compared with keeping that same time easy instead.
It’s worth being honest about how firm this evidence actually is. The 2007 study’s finding, while statistically significant, came from only 12 runners, a small sample by the standards of exercise science. A larger 2014 follow-up in the International Journal of Sports Physiology and Performance, led by Iker Muñoz, Seiler, and colleagues including Esteve-Lanao, randomly assigned 30 recreational runners to the same two approaches over ten weeks. Both groups improved their 10K times, by 5.0 percent in the more-polarized group and 3.6 percent in the more moderate-intensity group, but that overall difference between the two full groups didn’t reach statistical significance. It was only when the researchers looked specifically at the runners in each group who most closely stuck to their assigned training distribution that a significant advantage for the polarized, slow-heavy approach reappeared. The pattern holds up across multiple studies and lines up with what elite athletes do unprompted, but the size of the benefit, and how strictly a runner needs to follow the 80/20 split to get it, is still being worked out rather than fully settled.
What the research supports, cautiously, is a specific and slightly counterintuitive claim: not that easy running is good in some general sense, most runners already assume that, but that running easy more often, at the direct expense of moderate effort rather than hard effort, has outperformed the alternative in more than one controlled comparison. The slow miles aren’t filler between the workouts that matter. In these studies, they were part of what made the hard efforts pay off.