What Predicts Endurance Performance? The Physiology Behind Running Faster

Understanding what drives endurance performance can help us make better training decisions, whether the goal is to run a faster 5K, improve marathon performance, or simply become a more capable and resilient endurance runner.

VO₂ max, lactate threshold, running economy and training history all play a role, but endurance performance is not determined by any one of these factors in isolation; the more interesting question is how they work together and, importantly, which of them we can influence through training.

A recreational runner and an elite marathoner are separated by enormous differences in aerobic capacity, for example. But once you start comparing highly trained athletes, the differences become much smaller, and factors such as economy, durability, biomechanics, and race execution can become increasingly important.

A recent study involving 888 endurance athletes gives us a useful opportunity to revisit this question. Rather than using the study as a recipe for predicting race times, I think it is more valuable to use it to challenge some of the assumptions we have developed around endurance physiology.

The problem with looking for the “best” endurance variable

Running performance is often discussed as though physiological characteristics exist in isolation. We talk about having a high VO₂ max, a strong threshold or good running economy as if each were a separate quality that can be ranked. But a runner doesn’t race with their VO₂ max. They race with the output produced from that aerobic capacity.

Imagine two runners with very different aerobic capacities. One has an impressive VO₂ max but is relatively inefficient. The other has a slightly smaller aerobic engine but can produce a given pace at a much lower oxygen cost. Which one is the better runner? You cannot answer that from VO₂ max alone.

Now imagine two runners with almost identical VO₂ max and running economy. One can maintain those characteristics deep into a marathon, while the other becomes progressively less economical as fatigue accumulates. Again, the laboratory numbers taken while fresh may not tell you who is going to run the faster race. This is why endurance performance is best understood as a system rather than a single metric.

A bigger aerobic engine still gives you a major advantage

If there is one finding from the new research that is particularly difficult to argue with, it is that aerobic capacity matters. Across this broad group of runners and cyclists, VO₂ max was the strongest contributor to the performance measures studied. In runners, it accounted for roughly three quarters of the variation in speed at lactate threshold and lactate turnpoint. That is a substantial effect.  But we need to understand that VO₂ max does not perfectly predict race performance.

If you compare athletes ranging from recreational to world-class, aerobic capacity explains a large part of why some people can produce considerably more endurance performance than others. But here’s the catch: the smaller the differences in the athletes’ metrics become, the less useful VO₂ max is as a sole explanation. Once everyone in the group already has a very good aerobic engine, having another few millilitres of VO₂ max may not be what separates them on race day. At that point, relatively small differences in other characteristics can become much more meaningful.

This is something that is easy to miss when we look at elite athletes and try to apply their physiology to everyone. For a developing runner, increasing aerobic capacity may be hugely valuable. For an elite runner with an already exceptional VO₂ max, improving how effectively they use that capacity may be the more productive target.

The engine is only useful if you can turn it into speed

This brings us to running economy. Running economy is essentially the amount of oxygen you need to run at a particular speed. Two runners can have similar aerobic capacity but require different amounts of oxygen to maintain the same pace. The more economical runner has an advantage.

In the study, economy made a much larger contribution to performance than fractional utilisation did, accounting for approximately 20–24% of the performance prediction across the running and cycling measures. For runners, this is one of the most practically interesting findings. It means there is more than one route to becoming faster. You can increase the size of the engine, or you can become better at using the engine you already have.

This is also why a runner can make significant progress without seeing a dramatic increase in VO₂ max. If running at 4:00/km becomes physiologically cheaper, that pace becomes less demanding even if the maximum amount of oxygen you can consume has barely changed. And that kind of development is often slow. Months and years of consistent running can gradually change the way you move and produce force. Faster running, hills, strides and strength training may all contribute to the development of a more effective running system, although the exact response will vary between individuals.

Stop asking only how high your threshold is

Perhaps the most interesting part of the research is what it tells us about threshold. Threshold training is a huge part of endurance running culture. We talk about threshold pace, threshold heart rate and threshold as a percentage of VO₂ max as though these numbers provide a precise description of an athlete’s endurance ability. They are useful concepts, but there is a potential trap in focusing too heavily on the percentage.

In this study, fractional utilisation, the percentage of VO₂ max associated with lactate threshold or lactate turnpoint, contributed relatively little to explaining differences between athletes. That does not mean threshold is unimportant. It means that saying “this runner reaches threshold at 88% of VO₂ max” doesn’t tell us as much as we might assume.

And when we care about running performance, absolute sustainable output is what eventually must become running speed. This is why I think threshold is better viewed as an outcome than a number to chase for its own sake. You don’t ultimately win a race or make a new PB because your threshold represents a particularly impressive percentage of VO₂ max. You perform better because you can run fast at a sustainable physiological cost.

What laboratory numbers can’t tell us?

There is another reason I would be cautious about turning these findings into a universal formula for race performance. The study did not measure marathon finishing times or 5K results. It used speed at lactate threshold and lactate turnpoint as performance proxies. The relationships with the physiological variables were extremely strong, but a laboratory performance proxy is still not the same thing as racing.

And racing introduces a whole collection of variables that become increasingly important as the distance gets longer.

  • Can you pace yourself properly?
  • Can you fuel effectively?
  • Can you maintain your mechanics when your legs are tired?
  • Can you tolerate the muscular damage associated with prolonged running?
  • Can you continue producing force when the effort starts to feel uncomfortable?
  • Can you cope with heat, hills or poor conditions?
  • Can you execute the race rather than simply possessing the fitness to run it?

None of these questions can be answered by looking at VO₂ max alone. This is where I think the concept of durability deserves much more attention.

What does this mean for your training?

This is where the research becomes much more useful than simply knowing which variable had the biggest statistical contribution. You don’t need to spend your training trying to optimise three laboratory numbers independently.

Build the aerobic engine

  • VO₂ max matters. That doesn’t mean every runner should obsess over testing it, but it does mean that the fundamentals of aerobic development remain extremely important. Consistent training over months and years gives you the opportunity to develop the physiological capacity required for endurance performance. And there is a useful distinction here: You don’t necessarily need to know your VO₂ max to train your aerobic system effectively. A laboratory number can be interesting, but the adaptation is what matters.

Don’t neglect the easy running

  • If economy and aerobic capacity both contribute substantially to endurance performance, there is a strong argument for thinking about training as a long-term accumulation of adaptations rather than a collection of heroic workouts. Easy running allows you to accumulate the volume needed to develop endurance without constantly generating excessive fatigue. That matters because the goal isn’t to produce one spectacular training session. The goal is to become a better runner six months from now. Keep some faster running in the programme.

Economy is not simply about being relaxed on an easy run.

  • Running at faster speeds exposes you to different mechanical demands and gives you opportunities to practise producing force quickly and coordinating movement at higher velocities. That’s why strides, hills, and faster running complement a versatile training program alongside more traditional endurance training. You don’t need to turn every run into a test. In fact, doing so would probably undermine the consistency required for long-term development.

Train your threshold, but don’t worship the percentage

  • Threshold training can be highly effective. But the goal should be to improve your sustainable running performance, not to achieve an impressive threshold percentage simply because the number looks good. If your threshold pace improves, that is meaningful. If your threshold heart rate changes slightly, it may or may not matter. The practical outcome matters more than the label.

Think about what happens when you get tired

  • This may be particularly important if you are training for longer races. A runner who is excellent for the first hour is not necessarily a good marathoner. Long runs and race-specific sessions can provide an opportunity to practise maintaining pace and mechanics while fatigue accumulates. They also allow you to rehearse fuelling and pacing, two things that become increasingly important as race duration increases. This doesn’t mean every long run needs to be brutally hard. Quite the opposite. The objective is to develop the ability to keep producing useful performance as fatigue rises, while managing the overall training load well enough that you can continue training consistently.

So, what predicts endurance performance?

There probably isn’t a single answer that can explain every runner. But we can make the picture clearer.

  • VO₂ max gives you the aerobic capacity to produce endurance performance.
  • Running economy determines how expensive that performance is.
  • Threshold determines how much sustainable intensity you can turn that capacity into

And then there is the part that is harder to measure: How well can you maintain all of it when the race gets long, and fatigue starts to accumulate? So, rather than fixating on specific numbers and values, the actual training process reveals a great deal about your future performance. Focusing on the balanced development of various aspects of your training—and being able to identify and work with your strengths and weaknesses—will drive your running progress forward. Instead of predicting a result from a chart based on VO₂ max, your own test runs and race simulations provide a true measure of your ability to sustain a given pace.

Research: Mougin, L., Bailey, S.J., Jones, A.M. et al. 35 Years of Joyner’s Endurance Performance Model: Assessing the Contribution of Physiological Determinants of Performance Proxies in 888 Individuals from Recreational to World Class. Sports Medicine, 56, 2047–2065 (2026).

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.