People always notice my stride.
—Lucas Verzbicas
WHEN IT WAS ANNOUNCED in August 2005 that Kenenisa Bekele would attempt to break his own 10,000 m world record in Brussels, Belgium, and that video of the attempt would be streamed live on the Internet, I made sure to tune in. Not only was I interested in seeing history made; I also simply love watching Bekele run. Some people would find the monotony of a man running around a 400 m oval 25 times unbearable. Not me. Few things are more beautiful than the stride of a great runner. Each runner’s stride is unique in its nuances, but in the stride of every great runner we see power and grace commingled in some fashion.
In the same way that a great song, novel, or film can resonate with the masses, I believe that everyone can recognize a beautiful stride. Suppose you are watching a couple dozen runners of widely ranging ability levels running individually at a fixed, moderate pace for 20 seconds. If you were then asked to rate the stride of each for pure aesthetics, your scoring would probably correspond closely to the results of a performance test, such as a 5,000 m time trial, of the same runners. In other words, the runner with the most beautiful stride would also be the fastest, and so on down the line.
In 2010 Lucas Verzbicas became the first sophomore to win the Foot Locker high school boys cross-country championship in the event’s 31-year history. This young man, who had already set a national high school record for 5,000 m indoors (14:18.22) as a freshman, not only won the race, but also utterly destroyed the other 39 boys in the race, all older than he was. Verzbicas stood out in another way, though. His stride was beautiful—noticeably lovelier than those of the slower kids. My astounded media friends and I were all struck by it. And Verzbicas himself was not unaware of it. “That is what people always notice and acknowledge about me is my stride,” Verzbicas told me after the race, answering my compliment in the aw-shucks manner of a Midwestern prom queen praised for her beauty, albeit with a slight Lithuanian accent (his family immigrated to the United States when he was 8).
Running ability is plainly visible in the form of the stride. In fact, I daresay that it is possible not only to identify the better runner in almost any pair of runners by observing their respective strides but also to accurately assess the fitness level of any single runner whose stride is well known by seeing how it looks today. American mile record holder Alan Webb has had an up-and-down career. I never know which Alan Webb is going to show up for the next race. But within the first 100 m, I always know which Alan Webb is racing, because his stride tells me. I could never put into words exactly what looks different about Webb at his worst versus Webb at his best, but it could not be more apparent to my eyes.
The stride is everything in running. Ability, fitness, fatigue, motivation, and all other factors that affect running performance are wholly mediated through the stride, mostly in visible ways. There is simply nothing else going on in running but the stride. This fact may seem too obvious to bear mentioning, and yet it has been oddly overlooked in exercise science over the last several decades. The dominant theoretical paradigm of running performance views body form and biomechanics as a transparent receptacle within which lies all the stuff that really matters—mainly, the machinery of oxygen transport. To be sure, a large aerobic capacity, or VO2max, and all of the physiology underlying this capacity (high muscle capillary density, high muscle mitochondria concentration, and so forth) are critical to running performance, but only inasmuch as they influence the stride. While it may be neat to have your VO2max measured in a laboratory and to learn that number, there is absolutely nothing useful you can do with this knowledge. The whole VO2max concept has done nothing but overcomplicate the pursuit of better running performance. Looking beyond the running stride at VO2max for the sake of running better is like looking beyond the cookie at its sugar content for the sake of finding out how it tastes.
Exercise scientists understandably got pretty excited about the relationship between oxygen consumption and exercise performance when they discovered it a century ago. But what happens so often in science is that scientists overestimate the importance of what they can measure and underestimate the importance of what they cannot measure. When exercise scientists developed the ability to measure oxygen consumption, they soon came to believe that oxygen consumption was everything to exercise performance. Even though it may be obvious that the stride action itself is no small piece of the running performance puzzle, scientists have had difficulty quantifying its characteristics, and so the stride has been marginalized as an object of study and wrongly considered a peripheral factor in running performance.
This situation has begun to change. Some clever scientists have lately developed new ways of distinguishing better strides from worse strides, and such lines of research may help create a new paradigm of running performance that correctly identifies the stride itself as the thing that really matters.
One of these researchers is Stephen McGregor, an exercise physiologist at Eastern Michigan University with whom I had the good fortune to coauthor a book entitled The Runner’s Edge, which is largely based on his work (and which might at first reading seem a total refutation of the thesis of this book). McGregor comes from a cycling background, and in the early 2000s he became interested in developing new tools to quantify training loads in cyclists. These tools relied on data from power meters, which are a cycling-specific instrument. Thus, when McGregor set his mind to transferring these new conceptual tools to running, he needed new sources of data and hence new instruments. There are two types of devices able to capture distance, pace, and other such data in runners: GPS (global positioning system) devices and accelerometers.
McGregor saw more potential in accelerometers, because unlike GPS devices, accelerometers can capture not just speed and distance information but also much higher-resolution information about stride characteristics. In his work in cycling, McGregor was not interested in technique, because he understood that the importance of pedaling technique in cycling is small (because the fixedly rotating pedals basically force every cyclist’s legs to move the same way). But he rightly recognized that stride technique is a major factor in running, so he decided to use accelerometers instead of GPS devices in his research with runners, which began in 2005.
An accelerometer is a fairly simple mechanism that essentially measures the speed and direction of its own movement. In his work with runners, McGregor used triaxial accelerometers, which measure accelerations in all three planes of movement: forward-backward, up-down, and side-to-side. He and his colleagues and students performed a few interesting studies using these devices, and they have several others planned. McGregor’s first study demonstrated that data from running accelerometers could be used to accurately estimate the oxygen cost of running and thus to quantify a runner’s training load over time.1 Other studies used accelerometers to identify signature differences in the stride characteristics of trained and untrained runners and to identify changes in stride characteristics associated with fatigue.2 Some of the findings confirmed truths about running biomechanics that had been previously discovered through other forms of measurement, whereas other findings taught us things about running biomechanics that had not been previously known.
As expected, McGregor found that trained runners accelerated less in all three planes of motion. At any given speed, the members of the nationally ranked Eastern Michigan University cross-country team whom McGregor used as subjects bounced up and down less, moved less from side to side, and lost less forward momentum between strides than did the non-runner students in the study. But there were also differences within the pool of trained runners, with the fastest runners tending to accelerate least in all three planes. This finding suggested that some runners were gifted to be more economical than others, but McGregor’s studies also produced compelling evidence that training enhanced stride efficiency in a highly specific manner. Fascinatingly, McGregor found that trained runners were rather uneconomical walkers and were even somewhat inefficient at running speeds that were slower than their habitual training speeds. One in particular, who had the highest VO2max on the team, was among the least economical members of his team at a slow jog. McGregor speculated that this runner wasted a lot of energy at these speeds because, thanks to his huge aerobic engine, he could afford to. Because he never ran slowly to the point of exhaustion and hence never subjected his body to the crisis required to force a more efficient stride to delay exhaustion at those specific speeds, his body seemed not to have bothered adapting to the slow running.
Indeed, one of the general conclusions that can be drawn from McGregor’s research is that running becomes efficient only inasmuch as it has to. Another runner on the Eastern Michigan team is a case in point. We’ll call him John (as the subjects in proper scientific studies are anonymous, I will keep him anonymous here). John was the highest performer on the squad despite having one of the lowest VO2max measurements. How did he do it? He was the most economical at faster speeds. McGregor told me he wondered if John had not become extremely efficient as a consequence of forcing himself to keep up with more aerobically powerful runners in training and races. If so, this case is evidence that training in a group environment with some slightly faster runners may be an effective way to develop a more beautiful—I mean efficient—stride.
The result may not be automatic, however. After testing John, McGregor became quite curious about him and asked the team’s coach if he could account for the test results based on real-world experience in working with John as a runner. “He’s not the most talented guy, but he can really suffer,” said the coach. When he heard this remark, McGregor was reminded of the VO2max testing he had done with the team members. In a VO2max test, the subject runs on a treadmill at incrementally increasing speeds, each of which is sustained for one to two minutes, until he can go no faster, while breathing into a tube connected to a machine that calculates oxygen consumption. Most runners reach their highest level of oxygen consumption at their fastest running speed in the VO2max test, but some are able to survive one or even two more belt speed increases after reaching VO2max, and John was one of them. So perhaps training with naturally faster runners is a good way to develop a more economical stride—but only if such a runner is willing and able to suffer more than most other runners.
Another interesting finding of Stephen McGregor’s work with accelerometers was that vertical acceleration accounted for a greater proportion of total oscillation in all three planes for trained runners than for nonrunners. In other words, when changes in forward velocity, side-to-side movement, and up-and-down movement during running at any given speed were totaled, up-and-down movement accounted for a greater portion of the total in trained runners than in nonrunners. According to McGregor, there are two possible reasons training reduces up-and-down movement less than movement in the other two planes. First, while training reduces acceleration relative to speed in all three planes of movement, vertical oscillation can be reduced only so much, because runners have to be able to take big strides to run at high speeds and they have to get airborne to take big strides, as all runners fall toward the earth at the same rate. Second, a certain amount of vertical acceleration (bouncing) is required to maximize the “free energy” that the legs can capture from impact with the ground and then send back into the ground to propel forward motion. In any case, whereas vertical oscillation tends to increase steadily with increasing speed in nonrunners, vertical oscillation increases only to a certain point with increasing speed in trained runners, then plateaus and actually begins to fall at very high submaximal running speeds. This phenomenon is undoubtedly a training adaptation that enables runners to “put everything” into moving forward as they approach their physiological speed limit.
It bears noting that all of the training adaptations that distinguish the trained from the untrained stride are almost entirely the results of unconscious, automatic operations. Not only did the runners in McGregor’s studies not train consciously to minimize their side-to-side and forward-backward accelerations more than their vertical accelerations, but they were also not even aware that their strides had changed in this way through years of training. Although all of the trained runners involved in McGregor’s research thus far have also been naturally gifted runners, it’s safe to assume that the improvements in running performance that occur in runners of all natural ability levels through training are largely the results of automatic improvements in stride efficiency as well.
McGregor’s most intriguing study so far is one in which he measured and compared entropy levels in the strides of trained and untrained runners. Entropy is essentially a measure of the predictability of the behavior of a system. In the case of the running stride, entropy has to do with how much each individual stride looks like the one before and the one after. McGregor speculated that entropy would decrease as runners approached exhaustion—that reduced entropy would prove to be a signal of fatigue. He anticipated finding that the stride would become more predictable and less varied near exhaustion because such predictability in any system would indicate that system was “constrained.”
In the case of running, it stands to reason that the stride becomes constrained when some component of the stride runs up against a performance limit. To oversimplify, suppose that after sustaining a certain speed for a certain period of time, a runner’s left soleus muscle begins to lose contractile force. This limit becomes a constraint on the whole system of the runner’s stride, making each stride look more like every other stride than when the runner is not fatigued and his stride is not thus constrained, so that it has a little more “play.” Research has shown that certain individual muscles do fatigue faster than others during running, and that the brain responds to local muscle fatigue by tweaking the entire stride to ensure that it remains properly coordinated within the limit set by that one tired muscle.3 It’s sort of like putting the most tired sled dog at the head of the team to keep the whole unit working together.
In his study, McGregor did indeed find that stride entropy decreased as runners approached the point of fatigue. Among the Division I college runners used in the study, entropy stayed high at speeds up to 20 km per hour (kph, or 4:50 per mile). But when speeds above 20 kph were sustained, entropy decreased, anticipating the moment of exhaustion. This finding suggested that decreasing stride entropy was a manifestation of fatigue. An important question raised but not answered by these results is that of the specific nature of the constraints that caused entropy and fatigue. The two main categories of constraint sources are metabolic and biomechanical. An example of a specific metabolic constraint is the inability of the cardiorespiratory system to provide enough oxygen to sustain a given running speed. An example of a biomechanical constraint is joint decoupling, which is a fancy name for a loss of the precisely coordinated rhythm of movements required for efficient running (which is a bit like juggling in that any small, localized loss of rhythm causes the whole show to fall apart). Such a slippage in coordination could be caused by local muscle fatigue, a loss of elasticity (or springiness) in certain muscles, or changes in central drive (will) from the brain. McGregor did not favor any particular hypothesis in his study, but in conversation he told me that he thought the primary constraint on the stride at exhaustion was biomechanical.
Whatever the source of fatigue may be, McGregor’s work demonstrates that fatigue ultimately manifests as changes in the stride. Indeed, fatigue does not even exist except inasmuch as the stride changes. This idea stands in stark contrast to the conventional idea of running fatigue, which sees the source of fatigue as always entirely metabolic and ignores the stride per se. My most recent conversation with McGregor about his work got me to thinking: If fatigue exists only to the extent that it changes the stride, then a sort of fatigue profile for each individual runner could be created and then used to precisely assess the runner’s present fatigue level at any time thereafter. For example, suppose an accelerometer was used to measure stride characteristics in a given runner as she ran at 7:00/mile, beginning in a fully rested state and ending at exhaustion. The gradual changes in stride characteristics observed during that span of time would become a fatigue spectrum for that runner (at that pace). And so, if the same runner were to show up at McGregor’s lab the day after running a marathon that McGregor did not know about, strap on the accelerometer, and start running 7:00 miles, the runner could not fool the scientist. McGregor would see that after only a minute of running at this pace, the runner’s stride characteristics looked like they normally did after three hours of running at that pace, and McGregor might say, “Did you by chance run a marathon yesterday?”
In support of the accelerometer data, direct visual observation would undoubtedly also reveal to McGregor, and anyone else who knew the runner, a telltale ugliness in the runner’s stride. Indeed, because fatigue is stride deterioration, and stride deterioration is almost always visible, runners can seldom fool a good coach about their fatigue level even if that coach does not have fancy equipment with which to measure biomechanical changes.
Steve McGregor has no immediate plans to pursue my hypothesis. Instead, he has plans for even more ambitious research that could profoundly validate the stride-based philosophy of running performance. McGregor’s idea is to track physiological and biomechanical changes (using accelerometers) in college runners over the course of four years—their entire collegiate running careers—and correlate these changes with changes in their competitive performances during that time. The objective will be to determine whether it is primarily physiology or biomechanics that accounts for the improvement of those runners who improve the most over four years. McGregor is far too good a scientist to prejudice the results of his work with his own expectations, but based on what he has learned from his research so far, he says he does expect to discover that, at this level, improved performance is caused mainly by changes in the stride—that is, by neuromuscular, not cardiovascular, adaptations to training.
The teaching of running technique has become popular lately. The top-selling running book of the last several years prior to the publication of Christopher McDougall’s Born to Run was Chi Running by Danny Dreyer, which taught a quasi-yoga-based style of running purported to reduce injury risk.4 Dreyer has made a thriving business of Chi Running, with videos, clinics, and even a certification program that trains new instructors in the technique. The Chi Running method is very similar to the Pose running method, created by Nicholas Romanov, which has been around for many years but has really taken off only within the last decade. Once all but ignored, running technique is now the topic of countless magazine and Web site articles, is taught by a growing number of running coaches, and is intensively discussed on Internet chat forums and actual training runs. Underlying all of this discussion is a gradually spreading consensus that running technique can in fact be effectively taught—that there is an identifiably correct way to run that every runner can learn and use to run faster and with fewer injuries. (Most of the popular running technique systems, which, in addition to the Pose Method and Chi Running, include Evolution Running, are indeed similar to one another. Each is, at its core, a way of correcting the common error of overstriding. The various methods don’t peddle wildly different notions of the right way to run.) This belief that there is a single right way to run represents quite a departure from the old-school view of running technique from previous decades, which held that good running technique was essentially something that a runner was either born with or not, and that the only way to improve running technique was to simply run and let the process happen naturally.
There are some running experts who still believe that this is the case. Among these experts is Ross Tucker, an exercise physiologist at the University of Cape Town, South Africa, whose work I discussed in Chapter 1. Tucker is not persuaded that there can be a single right way for every runner to run. In an article on his Web site, Tucker explained: “My personal opinion is that if there [were] a way to run faster and with fewer injuries that was guaranteed to work in all people … then it would be discovered by default. It’s difficult to fathom that millions of people, with different body shapes and sizes and leg lengths and [centers] of gravity and joint angles could fit into one single pattern or technique. Rather, the passage of time would filter out any flaws for each person.” Tucker believes that individual runners naturally develop the stride pattern that works best for them in the normal course of training, but that this pattern is not transferable—in other words, past a certain point, what works for me is unlikely to work for you.
Scientific research on the teaching of running technique tends to support Tucker’s view. For example, a 2005 study published in the Journal of Sports Sciences reported that the running economy of 16 high-level tri-athletes was actually reduced (meaning the athletes became less efficient) after 12 weeks of practicing the Pose running method.5 In fact, to my knowledge no study has ever demonstrated an improvement in running economy or performance resulting from technique training.
Consciously meddling with your stride may indeed make it less efficient instead of more efficient. Research has shown that there is less activity in the brains of skilled performers of all manner of coordinated movements when performing those movements than in the brains of the unskilled. The more you think about something while you do it, the less efficiently you do it. This is as true of running as of any other activity, as shown in a recent study from the University of Munster, Germany.6 Trained runners were asked to run for 10 minutes at a designated pace on three occasions: once while thinking about their breathing, once while thinking about their stride, and once while thinking about the environment around them. Oxygen consumption was monitored during all three runs. And guess what? The runners consumed the least amount of oxygen—that is, they were most economical—while not thinking about their bodies as they ran.
As you probably know from experience, as running speed and fatigue levels increase, it becomes increasingly difficult to concentrate on the external environment and more and more necessary to consciously will that next stride to happen. This act of willing, called “executive brain function,” costs a lot of energy and thus actually hastens fatigue even as it is called upon to resist fatigue. While all runners have to concentrate on the movement of their limbs when running near the limits of their speed and endurance, it appears that better runners don’t have to think as much about running while they run, and indeed the very unconsciousness of their running is a major aspect of their superior efficiency. Much as an expert knitter can carry on a conversation while knitting a sweater, seemingly paying no attention whatsoever to the workings of her fingers, whereas the beginning knitter develops a headache after 20 minutes of totally focused knitting, so ferociously must she concentrate on the needles, the highly trained runner can mentally “get out of the way” of his stride and let the brain’s unconscious motor centers control it with no wasted energy, whereas the beginner must force his legs to obey the command to keep moving.
Where running differs from skills such as swinging a golf club and knitting is that golfers and knitters have to begin the learning process with conscious imitation of demonstrated techniques, but runners do not. In all skilled movements, technique becomes more unconscious as it becomes better, but the learning of running technique can be (and usually is) done through blind trial and error from the very beginning. The question that Chi Running and other running methods challenge us with is whether runners would be better off learning running the same way golfers learn golf and knitters learn knitting, even though running is clearly innate.
While there is still much more that we need to learn about how the running stride improves, a preponderance of existing evidence indicates that conscious manipulation of the stride is not the best way to run better. Conscious stride manipulation forces the runner to think about his stride, and as we have seen, thinking is the enemy of movement efficiency. Defenders of Chi Running, the Pose method, and the rest will argue that thinking is required only until the new technique becomes “second nature,” but there are other problems. Chief among them is that conscious stride manipulation involves making gross motor changes in movement patterns that are usually plainly visible to the outside observer. However, the real improvements in running economy resulting from long-term training that Stephen McGregor was able to measure are fine motor adjustments that cannot be consciously controlled. In the real world, the stride improves as the unconscious brain figures out how to sustain desired speeds with less activation of fewer motor units, not by changing where the arms and legs go. Such gains in efficiency are visible only as a general increase in the beauty of the stride.
The only common running technique flaw that exists at the level of gross motor coordination is that of overstriding, which is caused by the wearing of shoes and is best corrected primarily by addressing footwear, not by learning an entirely new way to run. Indeed, I believe that if all runners ran barefoot, the various running technique systems would not exist. Yet another problem with the technique systems is that they force every runner to try running the same way, whereas it is rather obvious—as Ross Tucker pointed out—that our very different bodies do not allow us to run the same way. A glance at the lead pack of runners in any major marathon will reveal all kinds of variety even among the best of the best. Some runners have a pronounced forward lean, while others are perfectly upright. Some carry their arms high; others, low. Some runners are forefoot strikers, while others are midfoot strikers. Some have loose, loping strides, while others exhibit compact strides with very high turnover rates. Each runner, through years of practice, has “solved” the “problem” of running fast over long distances by working out the optimal stride for his or her unique body. Yes, there are many characteristics that are common to the strides of all elite runners, but the rest of us cannot become elite runners—or even measurably better runners than we are today—by consciously aping these characteristics.
For example, faster runners typically have higher natural stride rates than slower runners. If two runners of different ability levels run together at the same pace, the more gifted of the two will take smaller, more frequent strides than the other. When I ran with Haile Gebrselassie, I observed that he took approximately 9 strides for every 8 I took. The natural stride rate of the typical elite runner is 90 strides per minute. The average stride rate of the typical midpack runner is closer to 80 strides per minute. Now, you might think that consciously increasing your stride rate from 80 to 90 strides per minute would be an effective way to gain a more efficient, elitelike stride. However, research has shown that runners become less efficient, not more efficient, when they force themselves to run at any stride rate other than their natural one.7 Indeed, I felt extremely awkward when I tried to match Geb’s stride rate while running with him.
It appears that with respect to most aspects of running technique, the unconscious brain knows better than the conscious mind what is most efficient. Each runner naturally adopts the stride rate that is most efficient given the totality of his or her biomechanics and body structure. My natural stride rate is lower than Gebrselassie’s because of largely unchangeable differences in the ways our bodies are put together. Trying to run more like him in this or just about any other way cannot possibly do me any good. This is not to say that my stride rate could not increase in a manner that would enhance my running economy, but if it did, it would have to do so through an unconscious evolution of my overall stride. There is, in fact, anecdotal evidence from runners who train with speed and distance devices with cadence-monitoring capability that stride rate increases naturally and unconsciously as fitness increases.
Not only is it difficult to identify stride changes that are actually helpful, but it is also extremely difficult to make most specific stride changes (whether potentially helpful or not) stick. On a visit with the Mammoth Lakes Track Club, I watched three-time Olympian Jen Rhines work consciously to increase her knee lift while running 100 m sprints under the watchful eye of her coach and husband, Terrence Mahon. “I’ve always had a bit of a shuffle, so that’s our goal this year, is to get rid of the shuffle,” she told me. Frankly, I expect to see Rhines shuffling as always the next time I watch her race, and if I do, I am confident it will be for the best. Rhines shuffles for a reason. I think it is highly likely that her body is designed in a way that makes low knee lift and a low back kick more efficient for her than the high knee lift and high back kick seen in some other elite runners, such as Kenenisa Bekele. Forcing a change probably will not help her running, and because her unconscious brain knows this, it will cause her to revert to her efficient shuffle in the heat of competition.
Jen Rhines is not alone among elite runners in fiddling with her stride. Nowadays many elite runners in the West do the same. Alberto Salazar told me: “Biomechanics are vital. The old idea that your form really doesn’t matter is so outdated. We know without a doubt that the very best runners that have the longest careers are the ones that are most biomechanically sound. So that’s something I really stress with my runners. I feel it’s as important as how many miles you run and the pace that you run your intervals.”
Conscious stride manipulation might make more sense for elite runners than for the rest of us. Stride technique improvement methods are like nutritional supplements. A balanced, high-volume, progressive running program is like food. Nutritional supplements are not intended to provide the foundation for optimal nourishment. Health is best supported when a person uses supplements minimally to augment a nutritious diet. Elite runners have a nearly perfect “diet” in the sense that their bodies are exceptionally well designed for efficient running and they already do what is known to improve running efficiency: hard, varied, high-volume running. So there is nothing left for these athletes to do with respect to enhancing the “beauty” of their running but to supplement their training with conscious stride fiddling (which they never do in the form of learning a universal technique system like the Pose method but instead do by identifying and attacking specific limiters in their individual strides, as in the case of Jen Rhines’s shuffle). But the rest of us have much more to gain from improving our “diets” by developing strength, mobility, and power to make our bodies structurally better suited to efficient running and by increasing the repetition, variation, and exposure to fatigue in our training.
The value of conscious stride manipulation is debatable even for the elites. The very best runners in the world, the East Africans, do not widely practice stride technique manipulation. (Nor do they take nutritional supplements, for that matter.) I do not, however, rule out the possibility that stride technique manipulation can sometimes yield performance gains. There are a few celebrated cases of successful stride manipulation, such as that of Derek Clayton, an Australian who had an unspectacular career as a track runner before deciding to move up to the marathon. When he made this transition, Clayton consciously replaced his bouncy track stride with a lower-impact marathon shuffle, which he credited with helping him break the marathon world record twice in three years between 1967 and 1969.
For every success story like this one, however, there are probably dozens of unknown stories of consciously made stride changes that produced negative results. So while it can work, conscious stride manipulation is the last place I would advise you to seek improvements in the beauty of your running. The first place you should seek it is in a running program that is properly designed to serve as “stride practice.”
Some scientists I know (but cannot identify here because they currently wish to keep their work secret) are working to develop a tool that will enable runners to continually measure their running economy in real time. With it, any runner could get instant feedback on the effect of any particular stride manipulation. Once this tool is in widespread use, we may discover that some particular stride tweaks do immediately boost efficiency for some runners. However, I doubt we will find many.
The conventional, energy-based model of running performance encourages runners to view training as a means to increase fitness—to change the physiology inside the blank vessel of body form and biomechanics. The new, stride-based model that is taking shape in the hands of the likes of Stephen McGregor suggests that training is something else entirely: It is stride practice. Every step of every run is a step in the direction of a more beautiful (that is, efficient and powerful) stride. Yes, physiology changes in the process, but such changes are not ends in themselves. They simply support the stride changes that enable a higher level of performance. For example, an increase in the muscles’ capacity to burn lactate as fuel is a physiological change that enables a runner to sustain a faster speed longer before her stride loses entropy and fatigue sets in. In other words, it is a change that lifts a fatigue-inducing constraint on the stride.
Practically speaking, what does it mean to approach training as stride practice? I suggest that it is rather different from compartmentalized “technique training” as it has come to be known. The best way to pursue improvement in running form is not to think about how you run but rather to simply facilitate and hasten the unconscious process that produces specific stride refinements through communication between the brain’s motor centers and the muscles. There are three obvious ways to hasten this process, alluded to earlier: repetition, variation, and exposure to fatigue.
The key difference between trained and untrained runners is, of course, that trained runners have done a lot of running and untrained runners have not. Thus, the research of Steve McGregor and others that demonstrates superior efficiency in the strides of trained runners is a rationale for high-mileage training, above all else. The more you run, the more time your brain and your muscles spend in collaborative communication about the problem of running efficiently, and the faster the fruits of this communication will accumulate. It takes years of training experience for any runner to develop the most beautiful stride she will ever have, but high-mileage training will accelerate the evolution. Likewise, a runner of any given level of experience will run most efficiently at a high training volume than at a low one.
High mileage and low mileage are relative phenomena, of course. As we have seen, not all runners benefit equally from equal amounts of running. Some runners thrive best on relatively low-volume training, either because they are injury prone, they excel with a heavy emphasis on high-intensity running (which necessarily limits volume), or there are psychological factors at work. But even low-volume runners have to run a heck of a lot more miles than zero to get their best results and will run best when training at a volume level that is close to their personal limit. Bear in mind that low volume at the elite level among marathoners is 100 miles per week, and there’s a reason for that.
Many runners speak of something magical that happens to their stride when they raise their mileage above a certain threshold. I had this experience when I started running consistently more than 80 miles per week for the first time. My stride became effortless in a way it never had before. It was a great feeling.
The process by which the brain and muscles learn to communicate in new ways that produce a more beautiful stride is similar to the process of biological evolution. Both processes adhere to the maxim “Necessity is the mother of invention.” Species living in a stable ecosystem do not evolve rapidly because there is little pressure to evolve. But a changing ecosystem creates a necessity for evolution by taking away one or more of the conditions that its species depend on for survival. And when evolution becomes necessary for survival, it happens quickly. Similarly, if you run more or less the same way every time you run, there is little need for your brain and muscles to come up with new and better ways to put your stride together. But when you vary your running by running fast some days and slow other days, flat some days and hilly other days, on roads some days and trails other days, your neuromuscular system is constantly stimulated to adapt to the new challenges imposed on your stride. Some of these adaptations will benefit your stride generally. For example, suppose you have never run uphill before, and one day you give it a try. To meet this challenge, your brain will have to fire your leg muscles with patterns it has never used before. Having practiced these new movement patterns, your brain may call on some of them again the next time you run on flat ground and discover that one or more of them enable you to run more efficiently.
There is no formula for optimal variation in training. As a general rule, I suggest that each week you do some running at paces ranging from a slow jog to a full sprint, at least one hilly run, and at least one off-road run (if you normally run on the roads).
The most potent stimulus for improvements in running biomechanics is mostly likely running in a fatigued state. As we have seen, fatigue manifests as a deterioration of running form that can be measured as a reduction in entropy. When you become tired, you become unable to run the way you normally do. Your stride turns ugly. Resisting fatigue is largely a matter of trying to continue to run normally despite factors such as joint decoupling and muscle fuel scarcity that pull your stride apart. Through this effort to keep your form together, your neuromuscular system learns new patterns that increase your resistance to stride deterioration and increase your running efficiency in a rested state.
The idea that running in a fatigued state is something to be sought out in training for its performance benefits is unusual. The conventional, energy-based model of running performance views the work that makes you tired as beneficial. For example, exposure to VO2max in training increases VO2max. But what I propose is that the fatigue itself—or more particularly, the effort to resist it—is the point.
I am not suggesting that the more fatigued running you do, the better. You could very easily run tired all the time by overtraining. There is a difference between quality fatigued running and nonproductive fatigued running. Fatigued running is quality when you have some capacity to resist it. When it goes past a certain point, fatigue wrecks your stride and there is nothing you can do about it. Any running you do in this state is unproductive. So the objective is to subject yourself to judicious doses of fatigued running in training. As a general rule, to maximize the rate of stride improvement, run in a fatigued state as much as you can without accumulating fatigue from day to day and eventually week to week.
There are three main ways to achieve this objective. First, your training program should include a few workouts per week that result in a high level of fatigue—typically a tempo run, a session of high-intensity intervals, and a long endurance run. Doing too few fatigue-inducing workouts will not produce sufficient fatigue exposure to maximize the neuromuscular adaptations you’re seeking. However, attempting to do too many hard workouts per week will cause you to carry too much residual fatigue between workouts, hampering your performance in them.
Second, instead of trying to do more than three hard workouts per week, you can increase your exposure to fatigue in a more productive way by adding short, easy recovery workouts to your schedule. Such workouts are gentle enough so that they will not hinder your recovery from previous hard training, but because you start these runs in a prefatigued state (within 24 hours after completing a hard run), they provide extra exposure to fatigue despite their brevity and slow pace.
Third, engage in interval workouts. The recovery periods that occur between high-intensity running intervals enable you to spend more total time running at high intensity than would be possible with a single, sustained, high-intensity effort to exhaustion. With rare exceptions, anytime you train above anaerobic threshold intensity, your workout should have an interval format.
The most important pace at which to experience fatigue is race pace. As Stephen McGregor’s research has shown, the stride becomes more efficient only to the degree that it has to. It will not become more efficient at paces you seldom run, and it will become only marginally more efficient at running paces at which you seldom experience fatigue. This is why McGregor’s accelerometer data show that very good runners tend to be rather uneconomical at slower paces. You must run at race pace to the point of entropy—that is, to the point where some specific constraint in your stride limits your performance—to stimulate the neuromuscular adjustments that will make you more efficient at your race pace. Therefore race-pace running needs to be a regular part of your training regimen.
Again, the rationale for running in a fatigued state is that it forces the neuromuscular system to confront the primary constraint that limits performance and thereby creates opportunities for the neuromuscular system to experiment with new stride patterns, one or more of which may alleviate that specific constraint. But exposure to fatigue is not the only way to stimulate this process. Runners are also limited by structural factors including muscle power, joint mobility, and leg stiffness (or the capacity to quickly tense the right motor units to the right degree in the instant before impact to maximize the bouncing effect). Research has shown that specific training to enhance these structural characteristics alters the stride in ways that boost performance.8 Aware of these effects, many elite runners incorporate large amounts of strength training, explosive jumping exercises, mobility drills, and dynamic warm-up activities into their training. In fact, this huge commitment to cross-training for stride improvement is the greatest difference between the training of today’s top runners and that of past generations. One of the best young middle-distance runners in America, Anna Pierce, who has a 1,500 m PR of 3:59.38, told me that at times she spends more time each week lifting weights, tossing medicine balls, bounding, and so forth than she does running.
We have seen that there is little scientific support for the practice of consciously manipulating the stride to conform to some universal ideal of good running form, and that new research evidence provides strong support for the idea that the best way to run more beautifully is to just run—and more specifically to run a lot, run with variation, and run fatigued. Nevertheless, there is other evidence, from real-world training and clinical environments, that supports limited use of conscious stride manipulation. Specifically, I encourage runners to limit themselves to making two specific types of conscious stride changes: those that reduce injury risk and those that reverse the stride distortions imposed by shoes.
Changeable aspects of running form contribute to many running injuries. It is often possible to identify these technique flaws, consciously change them, and thereby reduce the risk of future injuries. One of the college runners Stephen McGregor studied was a talented young fellow who struggled to fulfill his potential because he got injured every time he tried to ramp up his training load. Once this runner was hooked up to an accelerometer, the reason for his dilemma became clear: The young man’s vertical accelerations were off the charts. He landed with a force equal to seven times his body weight at a slow pace of 12 kph (8:00/mile). Research on gait retraining in injured runners by Irene Davis, a professor of physical therapy at the University of Delaware, demonstrated that runners could, through conscious control, learn to run with less impact and thereby reduce their risk of injury.
Many running injuries are partly caused by muscle imbalances, such as those that are discussed in Chapter 10, most of which develop as consequences of excessive sitting. Stretching and strengthening exercises or alternatives such as yoga poses are required to correct these imbalances, but these activities will not automatically correct the stride. Davis’s work has shown that conscious learning of new stride patterns is often necessary as well. For example, weakness in the hip musculature causes the pelvis to collapse during running and leads to knee injuries. Strengthening the hip muscles gives the runner the wherewithal to run without a collapsing pelvis, but does not in fact suffice to stop the pelvis from collapsing. Davis teaches injured runners to consciously activate their hip muscles to correct this stride flaw. After a few weeks it becomes an ingrained motor pattern, and the runners are able to maintain their new form without thinking about it.
Running shoes are known to wreak havoc on running efficiency. A 2008 study by French researchers found that running shoes decreased running economy both by adding weight to the feet and by altering biomechanics in ways that reduced the ability of the legs to capture “free energy” from ground impact forces and reuse it to propel forward motion.9 On the biomechanical side, the core problem is that running shoes encourage runners to overstride, striking the ground heel first with the leg extended ahead of the body, instead of flat-footed with the foot underneath the hips. Overstriding exerts a strong braking effect—a pronounced heel strike even looks like pressing the brake of an automobile. No runner overstrides without shoes, because heel striking without the presence of cushioning material between the foot and the ground would be painful and injurious. Fully 80 percent of runners instantly become heel strikers when they put on a pair of shoes. It is not clear why four in five runners overstride in shoes but not in bare feet. There is some evidence that naturally gifted runners are more resistant to the stride-ruining effects of shoes, as the minority of runners that do not overstride in shoes also tend to be more efficient without shoes.
In any case, it is safe to say that overstriding is unnatural, because no runner does it in the natural, unshod state. For this reason, I believe that overstriding is one of the easier stride technique “errors” to correct. Again, though, correcting this error is best done primarily through means other than conscious control. Practicing running barefoot on grass, on sand, and/or on an at-home treadmill will get your neuromuscular system accustomed to making ground contact with a flat foot underneath the body’s center of gravity. Wearing the lightest, least-cushioned running shoes in which you are comfortable in your everyday training will help you transfer your barefoot running form over to your shod running. But you will probably need to exercise a conscious effort for a while to keep from reverting back to over-striding while wearing your shoes. I am living proof that this is one gross motor stride change that can be made fairly easily. I switched from traditional to minimalist running shoes and from a moderate heel strike to a midfoot landing to overcome a prolonged case of runner’s knee, and it worked.
Evolutionary biologists believe that human beings are “born to run” in the sense that many of the anthropometric features that we developed after splitting from the common ancestor we share with our closest genetic relative, the chimpanzee, specifically enhanced our long-distance running capacity. These features include rigid feet, upright posture, large buttocks, and copious sweat glands. However, it is patently obvious that not all humans are equally suited to running. In fact, there are many humans who are scarcely better suited to running than the average chimpanzee. I have known many people who, even as children, could barely run a step. So while our species is generally well designed for distance running, there is far greater variation in distance-running ability throughout the total human population than there is (for example) variation in sprinting ability in the population of cheetahs, a species that is generally designed for sprinting and all of whose individual members sprint exceedingly well. I believe that this is because distance running was never more than a specialty in early hominid and human populations. It was never the job of every member of a prehistoric clan to run, so there was never sufficient natural selection pressure to force the genes that support the highest level of running ability to overtake our entire species.
Today’s running technique instruction phenomenon loosely purveys the idea that each of us has the potential to run in the image, if not at the velocity, of the best of us. All that is required to realize this potential is to learn the “right” way to run and practice it. This idea is false. A clear-eyed look at reality reveals that we are not all born to run to the degree that Lukas Verzbicas and Kenenisa Bekele are. Aerobic capacity aside, such runners are gifted with body structure and innate neuromuscular coordination (research has shown that athletes who master sports skills quickly have more adaptive brain motor centers than slower learners) that enable them to run with an extremity of power and efficiency that the rest of us could never match with any amount of conscious emulation.10
There are many ways in which nonelite runners can emulate elite runners and benefit thereby. Beyond a very limited degree, copying their strides is not one of them. Indeed, our inability to copy the strides of the best runners is the very reason we are not as fast as they are. This is just the cold, hard fact of the matter.
That’s the bad news. The good news is that runners of all natural ability levels have a tremendous capacity to improve—or beautify—their strides. And the way to do it is to copy the methods that the best runners use to develop their own, naturally superior strides. These methods are mind-body methods in the sense that they function to refine communication between the unconscious brain and the muscles in ways that enable the neuromuscular system to generate more power with less energy. The methods that work best for this purpose include running a lot, running fast, and running far. These simple, but seldom properly exploited methods help each runner find a unique solution to the problem of overcoming running performance limits imposed by the particularities of body structure and neuromuscular coordination. To paraphrase Friedrich Nietzsche, “I wish that everyone would follow my example and find his own way to run.”