The Evolution of Running Training Theory

For the past 100 years, real-world and lab experiments interacted across continents to create our current understanding of training theory. We can learn a lot about our future by understanding our evolutionary past.

Photo: Getty Images

Heading out the door? Read this article on the new Outside+ app available now on iOS devices for members! Download the app.

I have a go-to response whenever I am asked about how to start running coaching: first, learn some of the history. 

It’s wildly fun to think about our coaching ancestors, working in a vacuum to optimize running performance with limited scientific knowledge compared to what we have today. I suggest starting with pedestrianism, the 19th-century sport where competitors would walk hundreds of miles—typically indoors in front of large crowds—for prize purses approaching $700,000 in today’s dollars. They would have slayed at Big’s Backyard.

But maybe a more apt place to dig in would be 1954, when Roger Bannister broke 4 minutes in a mile for the first time with a training plan that had brutal efficiency. He did low-volume training, with several weekly sessions of fast 400-meter intervals that got progressively faster as competitions approached—even to the point of all-out time trials at race pace mixed in during specific training. 

Bannister had a watch and a dream, largely working from intuition, empiricism, and basic science knowledge, with most applied physiology research decades away. As Athletics Weekly summed it up, “In terms of cumulative aerobic volume, in running three or four times a week, Sir Roger averaged less than 30 miles per week in the winter phase of periodization, regressing to just 15 miles per week during the competition phase of the macro-cycle, which seems staggering by today’s standards.”

Contrast that with the fastest male miler in the world right now, Jakob Ingebrigsten, who runs more than 15 miles almost every day of the week. He does high-volume training, with tons of moderate threshold-based intervals, and a smaller amount of high-intensity speed. The Norwegian system involves lactate monitors and constant physiology testing, with scientists working alongside athletes and coaches. An incredible 2023 article in the International Journal of Environmental Research and Public Health outlined an approach with over 100 miles per week, with 3-4 lower intensity interval sessions, and one higher intensity session. The Norwegians are most famous for double threshold days, with two moderate workouts in a single day, pioneered and popularized by Dr. Marius Bakken’s self-experiments in the 1990s.

RELATED: An Overview of the Norwegian Approach to Running Training

How did the world’s best miler go from a medical student building from the top-down to a Norwegian prodigy building from the bottom-up? There were a million questions asked and answered along the way. 

What works and what doesn’t? 

Most importantly: Why? 

Then along would come a rebel with a new approach, who would detonate conventional wisdom and force athletes to rethink their answers. This article attempts to sketch out a basic outline of that history of rebels and rethinking, an ode to some of the giants whose shoulders we are standing on. By necessity, it’s just a rough sketch, and I’ll miss thousands of important steps along the way. Let’s do this!

The Development of Training Theory

Already in 1954 when Bannister broke the 4-minute barrier, Emil Zatopek was running insanely high volumes. Here’s a fun session: 5 x 200 meters, 40 x 400 meters, 5 x 200 meters. The hardest part of that training approach may have been the counting. 

Zatopek did up to 150 miles per week, with tons of moderate-intensity intervals, sometimes with two workouts a day, on his way to winning four Olympic gold medals. It’s so cool to think about how close he got to modern science-driven approaches, without the benefit of the current field of applied exercise physiology. His most famous quote tells the story, I think: “Why should I practice running slow? I already know how to run slow. I want to learn to run fast.” Zatopek’s intuition and real-world experience, building on runners before him, led strikingly close to where the Norwegians are today.

The evolution of training theory is embodied by the Bannister and Zatopek extremes coexisting in the 1950s. In nature, there’s a principle of convergent evolution, where species adapt in similar ways but without a common ancestor, like how birds and bats separately evolved wings to fly. Running worked similarly. Someone would experiment with huge volume, like Gerry Lindgren’s 200+ miles per week in the 1960s, while someone else was trying something equally mind-boggling, and the cream would rise to the top internationally, influencing more athletes. In that way, humanity probed the limits of training theory and physiological potential, with hundreds of little laboratories of experimentation all around the world.

Some of the evolutionary experiments became dominant almost immediately upon their use, similar to how multicellular organisms took nearly 4 billion years to evolve, then proceeded to dominate the Earth’s surface over the last 600 million years. In the 1920s and the 1930s, Finnish professor Lauri Pikhala and German coach Woldmar Gerschler started a formal approach to interval training. Swedish coach Gösta Holmér instituted “fartlek” training, a semi-structured “speed-play” style with analogous structures as interval training. Zatopek built on this foundation, realizing that more intensity-controlled intervals could allow him to do higher volumes and improve his endurance performance.

But sometimes, evolutionary experiments didn’t succeed long-term. Coach Mihály Igloi’s system in the 1950s and 1960s relied on a truly bonkers number of intervals, with most training days involving dozens and dozens of differently-paced efforts, never over 400 meters, and very few continuous runs. He nailed principles of intensity control, shying away from using anaerobic processes. But the use of daily intervals faded away, or perhaps was folded into other approaches with lots of strides, like how 23&Me will report your levels of Neanderthal DNA. 

If we continue the Neanderthal analogy, the approach that would become Homo Sapiens was developed by coach Arthur Lydiard in New Zealand, building from aerobic approaches before him. With Lydiard, the modern era of running training took off.

Aerobic-Based Training Approaches

Lydiard’s system relied on an aerobic base with high volumes of low- to- moderate level continuous running, with three distinct phases: a preparation phase of building volume, a shorter hill phase of building strength on top of volume, and a competitive phase with specific training at race efforts. “Miles make the champions,” Lydiard said, underscoring the importance of long-term aerobic development. And in critiquing other approaches of his era, Lydiard had a stern rebuke. “No one will burn out doing aerobic running. It is too much anaerobic running, which the American scholastic athletic system tends to put young athletes through, that burns them out.”

RELATED: How to Achieve Your Running Goals? Become an Aerobic Monster.

The Lydiard evolutionary tree won out, for the most part, with aerobic-focused systems sprouting forth into new phylogenies. Perhaps the most important change from aerobically dogmatic approaches from Lydiard involved more year-round speed development and structured workouts in the preparation phase. Italian coach Renato Canova has directed many of the fastest marathoners in history, and his Fundamental Phase (analogous to the preparation phase of Lydiard) includes faster long runs, hill sprints, and interval training, plus plenty of doubles. Notably, Canova still wants the “Aerobic House” to be built with long-term high volume easy training, but that is taken as a given at the start of the Fundamental Phase. From there, Canova workouts get more specific to racing across the Special Phase and Specific Phase, with occasional “block” workouts including two hard sessions in a single day.

Canova block workouts absolutely fascinate me from an evolutionary perspective. These sessions can be brutally hard, sometimes consisting of 20 kilometers of tempo in the morning and afternoon. If you’re a training sicko like me, that probably reminds you of something else: those Norwegians doing their double-threshold days. 

The Norwegian approach is different, though, with controlled intervals that are much more sustainable and lower volume by design. Evolutionary, we’re probably seeing some sort of biological signal on the efficacy of double workouts. Which will win out and fill the ecological niche of Olympic medals in the coming decades? Maybe one or the other. Maybe both. Maybe neither, as competing approaches take hold. If there’s one thing we can learn from history, it’s not to get complacent with training dogma, since someone is out there experimenting with a new system that will take our lunch money. Coaches all over the world are working at the frontiers, on the track and road and trails, progressing methods at a breakneck pace.

The Role of Science

And I think that the rate of progress in training philosophies will only accelerate in the future. The reason? Scientists are in the room. A truism in training theory is that the new approaches almost always come from the field, where thousands of empirical experiments are underway all over the world. The lab helps find out why those field experiments work, nudging theory to lean in the right direction. But I think that’s changing. 

Scientists like Trent Stellingwerff, Iñigo San Millán, and Megan Roche are often driving both training theory and literature, while scientists are driving training approaches in places like Norway. That creates an exciting opportunity. It used to be that we started with the question, “What do athletes do?” Only then were we able to find out “Why do they do it?” Now, those questions are being answered simultaneously, skipping over years of guessing and testing, throwing eggs against a wall to find out which wouldn’t break.

The old problem is exemplified by “Tabata Intervals.” A study in the 1990s found that these 20-second hard efforts with short recovery led to big VO2 max gains, egging on a rush toward high-intensity interval training, particularly among beginner and intermediate athletes. However, there was a problem—the protocol was confined to a short intervention duration, ignoring the importance of aerobic development. Tabata intervals repeated mistakes that were made in some elite training systems 50 years earlier! In the U.S., the 1990s were characterized by underperformance internationally, with the common rationale being a similar mistake made in the collegiate system—pushing athletes all-out in intervals, prioritizing short-term performance over long-term aerobic growth. 

RELATED: The Surprising Connection Between Training Intensity And Long-Term Growth

I don’t think we’ll see that type of mistake again, and that’s because scientists have worked alongside coaches to optimize aerobic training approaches. As stated by the 2023 article on lactate-controlled training, lower intensity aerobic approaches work partially via “mitochondrial biogenesis and capillarization of Type I muscle fibers.” Those slow-twitch fibers power endurance performance. To simplify it immensely, there are two signaling pathways for mitochondrial proliferation: calcium signaling (responsive to high-volume training) and AMPK signaling (responsive to higher intensity training). The calcium signaling pathway has a much higher adaptive potential, hence systems that prioritize easy volume and controlled intervals. Simultaneously, excessively intense intervals can reduce adaptation over time via inflammatory responses, excessive recruitment of Type II faster-twitch muscle fibers, weaker lactate shuttling, worse metabolic efficiency, and reduced efficiency of many aerobic processes. 

Speed is essential, particularly for the mechanical adaptations needed to put out power. But overdoing the intensity or volume of speed training can undercut the aerobic system. Coach Lydiard knew it in the 1960s, and researchers figured out why. Now, researchers are helping drive the process to understand physiological optimization for athletes with different genetics and goals.

Putting Theory and Science into Practice 

I think it’s most helpful to summarize the cutting-edge understanding of fatigue through the lactate shuttle. A very basic description is that lactate is produced as our bodies use glucose to fuel ATP production during glycolysis. Lactate is a fuel source for cells, and it’s accompanied by a hydrogen ion that changes muscle pH and contributes to fatigue. A 2018 review in Cell Metabolism described the lactate shuttle where the cells use lactate for energy. If this shuttling mechanism is overstressed, lactate levels and fatigue rise and exercise becomes less sustainable. 

When lactate concentrations begin to rise, intensity switches from easy to moderate, an inflection point known as LT1, broadly overlapping with aerobic threshold where athletes go from primarily burning fat to primarily burning glycogen. And when lactate levels rise more steeply at higher intensities, intensity transitions from moderate to hard, an inflection point known as LT2, broadly overlapping with traditional lactate threshold (or critical velocity, depending on the method of calculation). 

RELATED: How to Set Your Heart Rate Zones

The lactate shuttle relies on the body’s ability to both process and clear fatigue byproducts, two processes which rely on mitochondria in the cells—mitochondria which are developed most via aerobic processes. Thus, even when lactate levels are very high, like in a mile race, we now know that the ability to optimize performance on the cellular level via the lactate shuttle likely relies on more aerobic training for most athletes. Specific training for race speeds plays a big role, but primarily in preparation for competition, and even that faster work relies on the long-term development of aerobic processes. Imagine how fast Bannister may have gone if he knew what Ingebrigsten knows now. 

The mechanistic understanding of training theory combines with more holistic (but still scientific) understandings of what athletes actually do. For example, a 2022 review study in Sports Medicine—Open looked at the training characteristics of world class distance runners, finding high weekly volumes with greater than 80% of that volume easy, below LT1. That data is supported by a 2019 study in the Journal of Strength and Conditioning Research finding that volume of easy runs correlated most strongly with world-class performance, along with short intervals to build speed and tempo runs still working the aerobic system around LT2. And all of this running science is not isolated–studies from nordic skiing and cycling show similar lower intensity approaches are ubiquitous at the top level of endurance training. 

Layer that in with studies on periodization, nutrition, female athlete health and performance, aging athletes, the endocrine and nervous systems, and every other training element you can think of. Journal articles didn’t create our advanced understanding of training theory, but they are moving forward the theory at ever-accelerating rates.

What’s Next?

Lydiard wasn’t able to look inside the cell like scientists can do today, but through trial-and-error—and learning from his evolutionary ancestors—he got incredibly close to understanding the training that would optimize cellular processes. Thousands of incredible coaches built on that knowledge. Now, coaches and scientists are working together, sharing information via studies, with new information coming out every single week! Where might training go from here?

I’m not sure exactly. There’s still room for revolutionary evolutionary experiments. But a study just came out to give us some hints. Published in the International Journal of Sports Physiology and Performance, the study discusses the evolution of world-class endurance training, including predictions for the coming decades by some of the most prominent researchers on the planet. Ahead, expect to see more advanced technology, like non-invasive sensors to determine lactate levels and muscle fiber typology, mixed with tons of other data sources, interpreted by multi-disciplinary teams, perhaps with the assistance of artificial intelligence. As big data, scientific knowledge, and training theories interact, we could be at an inflection point where progress rates skyrocket. 

RELATED: A Theory About Building Speed While Doing Long Races and Ultras

Instead of using the evolution of life as an analogy, perhaps we should use the evolution of computer chips. Moore’s Law states that the number of transistors on a computer chip doubles approximately every two years, leading to exponential increases in computing power. Much like when Moore’s Law was theorized in the 1970s for computing, I think we’re at the beginning of a technological revolution in training. I think we can expect non-linear growth in endurance training theory as the process of proving theories compresses due to improvements in information gathering and dissemination.

In trail and ultrarunning, we are still in the very early stages of understanding how endurance theory applies in a specific way. We may have room for more experiments than our road counterparts given the unique demands of events, but it’s key to remember that running is running (and endurance is endurance), and whatever we do is built on top of thousands of experiments to optimize training that came before us. If our approaches look that much different than what works on the track and roads, we’re probably trying to reinvent the wheel.

And one thing is for sure: just like all organic life is carbon-based, all advanced training in the next 100 years will be based on lots of aerobic training, some speed, and adequate recovery, grounded in the unique genetics of individual athletes. But how should those elements be mixed, particularly in new frontiers of trail and ultra training? 

I don’t know, but I have my guesses. I can’t wait to see what happens with the tests.

David Roche partners with runners of all abilities through his coaching service, Some Work, All Play. With Megan Roche, M.D., he hosts the Some Work, All Play podcast on running (and other things), and they answer training questions in a bonus podcast and newsletter on their Patreon page starting at $5 a month.

Want to Know What It Takes to Finish at Western States? Just Ask Hellah Sidibe.

Find out what happened when this six-year run streaker and HOKA Global Athlete Ambassador took on an iconic ultramarathon in California's Sierra Nevada