The Pacing Puzzle – the most interesting mystery

The running world pays a lot of attention to aerobic and anaerobic physiology, using terms like VO2max, aerobic threshold, lactate threshold, aerobic base, and max heart rate.  All of this is done in an effort to explain endurance performance, to explain what sets and limits the speed at which you can run, cycle, or row.  As interesting as those terms may be, the reality is that scientific research offers little in the way of explaining performance.  For example, the theory that lactate levels set an upper level of endurance performance has been disproven in recent years.  Similarly, VO2max has also been shown to be a poor predictor of performance.  In short, all those physiology terms may seem really neat and make the person saying them sound smart but they really don’t provide much in the way of useful information for explaining endurance performance. Unfortunately, exercise physiology still has a long way to go before it will be able to accurately explain what makes one person a world champion and another person an average runner.

That being said, there is something that actually does a great job of predicting performance and you are likely already familiar with it – the running calculator or race time predictor.  This device enables you to predict your performance at one distance based on your performance at other distances.  For example, using the calculator Greg McMillan has posted on his website I was able to calculate that if your best performance in a 5K race was 20 minutes then you can expect to race a 10K in 41:31.  And Greg’s calculator is not unique – other people publish similar calculators.

The reason we have race time predictors is that there is a known relationship between performance at varying distances. There is a relationship between how fast you can run the 1/2 marathon and the full marathon.  Or between your 10K performance and your 10 mile performance.  The various running calculators use that relationship in order to predict performance at any distance from a know performance at one distance.

Which brings us to the puzzle of pacing.  Think about it for a moment – when you run any distance other than the shortest of sprints (less than 100 meters) you do not run at the fastest pace you are capable of running.  For example, you don’t race a 5K at your all-out sprint pace do you? No. You race the 5K at a slower pace than your all-out sprint pace.  Similarly, your 10K pace is slower than your 5K pace.  And your 1/2 marathon pace is slower than your 10K pace.  In fact, any time you increase the distance you are running, your average pace per mile slows.  And the variance between your 5K pace and your 10K pace, or between all distances, is quite stable and predictable.


Experience teaches us that we have to slow our pace in order to run a further distance.  There is a maximum speed you can run for any distance and that maximum speed varies in a predictable manner based on the distance to be run.  Therefore, in order to maximize performance at any distance runners naturally learn to pace themselves.  New runners quickly learn their 5K pace, their 10K pace, indeed, their pace at any distance.

What happens if you try to run 10K at your 5K pace?  Most any runner can tell you – you won’t make it.  At a short distance beyond 5K your fatigue levels will rise dramatically and your pace will slow precipitously.

This is the pacing puzzle.  Prof. Tim Noakes says, “pacing is the really interesting athletic phenomenon” and he’s right. If you think about it for a few moments you will realize that pacing is a fascinating mystery. What, physiologically speaking, forces you to run slower at longer distances?  Or said another way, why can’t you run 10K or 10 miles at your 5K pace?  Something within your body obviously prevents us from doing so, but what?

Whoever solves the pacing puzzle will likely hold the key to improved performance.  Unfortunately, traditional endurance physiology has mostly ignored pacing because it’s not easy to test in the laboratory. Sure we can easily observe that pacing occurs, but how do you go about testing the cause or causes on a treadmill in the laboratory?  Instead, exercise physiologists have traditionally conducted tests in the lab that have no resemblance to how athletes actually perform during competition.  For example, a VO2max test – which for many years was the gold standard of endurance physiological tests – is performed on a continually inclining treadmill at a continually increasing pace.  The runner begins the test on a treadmill at a slow pace with little incline. The researcher then regularly increases both the speed and the incline until the runner drops from exhaustion (usually reached in 7 – 12 minutes or so). But no one races like that in real life.  Instead, we all naturally learn to pace ourselves in order to ensure our best performance.  Yet, despite the obvious difference between how a VO2max test is performed and the way runners actually run races, VO2max was touted as the primary factor explaining endurance performance.  But the fact of the matter is that you don’t race at VO2max pace. How could VO2max limit performance if you aren’t running at a pace that elicits VO2max?

Perhaps you are in the camp that claims that lactate threshold, not VO2max, exerts the most influence on performance.  Okay, how is lactate threshold measured?  The same was as VO2max.  The runner begins running on a treadmill in the lab at an easy pace.  At regular intervals the researcher increases the speed or resistance (incline) while taking blood samples from the athlete.  As you can see, there is no relationship between how lactate threshold is measured and how runners actually race. (All this talk of lactate threshold is actually a mute point since we now know there is NO lactate threshold and lactate does not cause fatigue.  Some people still cling to this outdated belief which is why I bring it up here.)  And since you don’t race most distances at lactate threshold how can lactate threshold limit performance?  Wouldn’t you have to run at lactate threshold for lactate threshold to limit performance?  What about races, like the marathon, that are run at less than lactate threshold – how could lactate threshold be limiting performance at those distances?

The bottom line is that traditional endurance physiology does not account for how runners actually race in order to maximize performance.  It doesn’t account for the phenomenon of pacing.

Since traditional endurance physiology has no explanation for pacing, where does this leave us?  Lucky for us, there are some in the physiology world who have been thinking about and studying pacing. I will talk about their work in a future post.


The Pacing Puzzle – the most interesting mystery — 18 Comments

  1. First thing that came to mind as i am reading about pacing is the issue of volume sets bodybuilders style training. I just came off such protocol after about 12 weeks and now I am experimenting with the opposite, very low volume sets relatively higher reps but more frequently during the week sometimes twice a day, session lasting 20-30 minutes. Within a week my strength shot straight up and it still going. When you are doing one or two sets per body part you sprinting right from the start versus pacing yourself doing 20 sets per body part. At 60 years of age I concluded that strength is an important factor. Time will tell how I will hold up.

  2. There are several items in your article I take issue with but I’ll limit this particular post to your opening paragraph and give some clarity on a point that you made regarding lactate threshold.

    “For example, the theory that lactate levels set an upper level of endurance performance”

    I’m sure you can cite samples of statements similar to this but it’s actually a misrepresentation of what lactate threshold is. Lactate threshold is a calculative value derived from a testing protocol that looks at blood (plasma) lactate levels and how they change during exercise. While there are several studies and methodologies around the testing protocols, collectively they identify a relationship between the derived value of ‘lactate threshold’ and exercise performance. This is basically what it is.

    Admittedly, even published studies speculate that there may be an actual cause/effect relationship between blood lactate levels and performance, but this is different than assertion of such. Nonetheless, it would be better if you could formulate your blog entries around accurate presentations of concepts instead of interpretations and speculation.

    In a comment I made to your article you link to in your paragraph above, I gave link to one of the Coyle/Costill papers that discuss LT. Here’s a 2007 clinical analysis of collective information regarding LT that gives some good links and history, it’s discovery and it’s relationship to exercise performance as a predictive metric.


    • Vinnie,

      Thank you for the thoughtful reply. My objective with my website is to discuss current scientific information and research about physiology and exercise in a way that informs the layperson. As such, I tend to focus on popular training and physiological wisdom and when it is at odds with science I try to point out the discrepancy. In the case of lactate threshold the popular myth has long been that lactic acid caused fatigue and as such runners should train at or below lactate threshold. As Dr Jack Daniels phrased it, “Runners must be able to work increasingly close to their maximum oxygen consumption without suffering from high accumulation of lactic acid in the blood.” Pete Pfitzinger stated the theory this way, “Your lactate threshold determines how fast you can race.” I contend that most runners still believe there is a causal relationship between lactate and performance and that a primary goal of training should be to reduce lactate levels at any particular pace. Pfitzinger claims that “…lactate threshold training is the most important type of training for distance runners…” while Dr Daniels created cruise intervals and tempo runs “for the same purpose (raising lactate threshold)…” It is these types of popular misconceptions about lactate that I was attempting to address.

      • Thanks. IMO think you were shooting for more just addressing misconceptions, particularly when you stated ‘All this talk of lactate threshold is actually a mute point since we now know there is NO lactate threshold”. It seems you’re inclined to dismiss it entirely.

        I think this link gives a nice summary of it all, including differences, misconceptions and why some, like yourself, would prefer to dismiss it rather than look further into it:

        • Vinnie,

          It has been interesting to watch the evolution of the term “lactate threshold” over the past 20 years. Way back then the “lactate threshold” was described as a point where lactate suddenly, precipitously, began to accumulate. It was held that lactate levels remained relatively low as exercise intensity increased from a starting low level but at a particular point, the “threshold” point, further increases in exercise intensity caused lactate levels to spike upwards, hence the use of the term “threshold”. Today we know from additional research that lactate does not increase in a “threshold” manner, that the “threshold” was an error of measurement.

          Today, the link you provide describes it like this, “However, there is wide controversy as of what lactate threshold really means as well as to what is the exercise intensity that elicits it. Lactate threshold is commonly known as the exercise intensity or blood lactate concentration at the one we can only sustain a high intensity effort for a specific period of time.”

          Assuming this is the currently held popular notion of lactate threshold indicates that the original “threshold” concept has been abandoned (which is good since we know it is wrong) but rather than abandoning the (known inaccurate) term a new definition is now being applied. Unfortunately, the new definition is the essence of vague. At least with the original idea of “threshold” it was at least a somewhat measurable point. Now the term means “you can run this long at this pace” where “long” can be any arbitrary distance or time. How useful is that? Clearly, not.

          As the article linked summarizes, ” I believe simply that after several decades of discussion and controversy, it is time for the “lactate threshold” concept to evolve and be named and defined differently so athletes and coaches can use it in a more meaningful and understandable manner in order to describe that “magic” exercise intensity that can only be sustained for a specific amount of time which is crucial for performance and success.”

          I suggest abandoning the term altogether as an unuseful relic of a past error that has long since been disproven. Further, there is no “magic” exercise intensity crucial for performance and success no more than there is a “magic” food or pill one can take. Performance and success are built on a wide range of factors, not one of which, as far as I know, reaches the standard for being termed “magic”.

          • “I suggest abandoning the term altogether as an unuseful relic of a past error that has long since been disproven”

            It hasn’t been disproven at all.

          • Let me follow up with something similar. Have you ever had a blood PSA test? It’s a screening test for prostate cancer. The relationship between blood PSA levels, changes in blood PSA levels and possible presence of prostate cancer are well established and capabilities and limitations of the tests are well documented. But PSAs don’t cause cancer, so does this mean the tests have no value?

  3. Another thing I want to address is a couple of peculiar statements you are making in your blog entry.

    “Unfortunately, traditional endurance physiology has mostly ignored pacing because it’s not easy to test in the laboratory. Sure we can easily observe that pacing occurs, but how do you go about testing the cause or causes on a treadmill in the laboratory?”

    The peculiarity in this statement is your reference to treadmill testing in relation to pacing. Ironic to your post is that the treadmill is actually the ideal mechanism for pace related testing because pacing can be stabilized on a treadmill. The mechanics of running on a treadmill are identical to running outside, it’s only peripheral issues that differ, such as changes in direction, variations in wind, these types of things. For steady-state running, it’s as good as being outside. The treadmill also grants greater latitude in the use of testing equipment since it is a much more controlled and stable environment. There’s no easy way to test a runner outside and it can be nearly impossible to hook up a moving runner to testing equipment.

    At a more fundamental level that has nothing to do with laboratories or treadmills, there are challenges in determining precisely what’s going on within a runner even with the best equipment available today, it’s mainly due the state of technology and limitations in intrusion with live subjects. Even if you could drive right into what actively happens within cell membranes and across boundaries and could actively observe complex interconnected chemical reactions, which is unlikely to ever happen in a practical sense, none of this means that it would lead to some magic bullet for training, which leads up to this statement:

    “Whoever solves the pacing puzzle will likely hold the key to improved performance.”

    I doubt this would true. For example, gravity will always be a puzzle in the world of physics from a truly fundamental standpoint, but we don’t and have never needed to understand it to that level since it’s properties in our physical world are all we really need to know to work with it effectively. Fully understanding gravity doesn’t mean you can build an antigravity machine and doesn’t change how a lever works. Landing a tiny craft on a small rock millions of miles away are challenges of logistics and technology, not one of our understanding of gravity.

    With running, there’s really only so many ways to train and so many ways to augment training. Certainly there’s unlimited ways of combining the many facets of training and in many cases identifying what actually constitutes training, but new ways of training are little more than reformation of old ways of training, or just recognition of aspects of behavior and lifestyle that were not included in the equations in the past. Improved knowledge might lead to better understanding of why something has the effect it does, however it’s generally more than sufficient to simply identify and study the relationship between cause and effect and adjust accordingly, which is exactly what runners and coaches have done since training ever began. Log books actually reveal far more than laboratories and studies do since they are documentation of a valid and highly useful form a research called ‘black box testing’, learning about something by observing what comes out in relation to what goes in.

    • Vinnie,

      Pacing on a treadmill is a great way to precisely set a desired speed. What it doesn’t do is explain in any way why a runner selects a particular pace, which is what I am discussing in my pacing post. What, physiologically speaking, forces you to run at a slower pace for longer distances? Why can’t you run 10K or 10 miles at your 5K pace? Something within our bodies obviously prevents us from doing so, but what? I submit that when (or if) that riddle – a riddle I call the pacing puzzle – gets solved it will likely provide additional refinement to training methods. Whether those refinements will be large or small is a matter of speculation, but my bias is that they will likely provide a significant change.

  4. Hi Rich,
    I believe you may have already hinted at a possible hypothesis that solves the pacing puzzle: muscle fatigue.

    I notice that when I run at a constant pace for an extended time, my heart race will slowly rise. It could be that muscle fatigue begins to set in, so that the muscle fibres become increasingly less efficient until complete fatigue an failure.

    • Marson,

      I believe that my muscle factor theory explains most of pacing but is not a complete explanation. I agree with you – during a run, muscle fibers fatigue, requiring more fibers to be recruited, driving up oxygen consumption.

    • Unfortunately ‘muscle fatigue’ is not much better defined than ‘fatigue’ is, so it doesn’t help much. It’s a given that muscles are the actuators of motion.

      How to stave off fatigue isn’t much of a mystery, it’s already well known. From a physiological standpoint you want high capillary density to facilitate the development of mitochondria (which forms adjacent to capillary walls) and high enzyme development. The translation is lots of miles at ~%70 MHR to build capillary density and work at higher %s to for mitochondria and enzyme production. Lots and lots of easy miles, spike it with some quality from time to time.

      • “From a physiological standpoint you want high capillary density to facilitate the development of mitochondria (which forms adjacent to capillary walls) and high enzyme development. The translation is lots of miles at ~%70 MHR to build capillary density and work at higher %s to for mitochondria and enzyme production. Lots and lots of easy miles, spike it with some quality from time to time.”

        That is certainly one of several theories. To date research hasn’t been able to prove or disprove the theory, so the question of what factors influence/determine fatigue remains mostly conjecture.

        • I’m not sure what you mean here.

          The relationship of angiogenesis and endurance training is well studied and documented, as is the formation of mitochondria in relation to capillaries. Are you referring to my referencing of running at ~%70 MHR? That wasn’t proposed as a theory but more of a best practice. You can induce a higher rate of angiogenesis at higher effort levels but there comes a level of tradeoff in terms of how much training can be performed at higher levels over a given period. Too low an effort level and the reverse happens, you’re not getting sufficient development regardless of the volume.

          Keep in mind that this is a process that happens over time and so somewhere around %70 is that sweet spot between volume and development that gives the best yields. This is somewhat subjective.

  5. There is no simple answer; no single cause or reason will ever be found. It is fun to think about how the human body ( or the universe ) works in simple to quantify terms. Me thinks that each individual is likened to a long chain that always has a weakest link; test your body and that weakest link will show itself. That is what makes it all so interesting. Don’t you just love it?.

  6. Hi Rich
    The pieces of ‘The Pacing Puzzle’ that you have discussed mostly apply to runners in their most fit years
    Runners like me ( age 87 ) face a whole different set of problems. or so it seems to me
    A whole new universe for you to explore – go for it ( there are lots of us old guys out there )

    Best, Fred

  7. I have enjoyed the comments section much more than the article itself, especially Vinnie’s input. I agree with him that there is a relationship between the leves of lactate, fitness and fatigue and understanding that relationship can lead to better training strategies.

    Good job Rich.

Leave a Reply

Your email address will not be published. Required fields are marked *