The Mileage Fallacy in Running

I am neither a running coach nor endurance-training specialist. I am simply a curious Sport Science generalist that is fumbling about trying to figure this stuff out. I am also far more interested in what I don’t know than what I do. As such, if you have specific critiques of my points then I genuinely do wish to be educated and course corrected.

Let us begin.

Goldilocks Happenstance

Here is a little hypothetical to kick things off.

In this scenario, there are 3 different people who can all do 10 repetitions of a strict Pull-Up technique if they go to absolute failure in a single all out set. All 3 of them want to get better at Pull-Ups from a maximum repetition perspective in 8 weeks.

A coach decides to put all 3 of them on the same 3 days per week (3/W) program for 8 weeks. The program starts with a conservative opening week of 3 sets of 5 reps with 3 minutes break between sets in each of the 3 days. The program then involves adding a single rep to each set of each day each week meaning they do 3 sets of 6 reps on week 2, eventually aiming to do 3 sets of 12 reps on week 8.

WEEK Frequency x Sets x Reps WEEK Frequency x Sets x Reps
1 3/W x 3 sets x 5reps 5 3/W x 3 sets x 9reps
2 3/W x 3 sets x 6reps 6 3/W x 3 sets x 10reps
3 3/W x 3 sets x 7reps 7 3/W x 3 sets x 11reps
4 3/W x 3 sets x 8reps 8 3/W x 3 sets x 12reps

Sounds good to be able to go from a 10 repetition maximum for one all out set to being able to get 3 sets of 12 reps 3 times in a week, doesn’t it? Except, remember, each of the 3 individuals are different people. What if the built-in rate of progression in the program is not conducive to getting each of the three participants better at a rate suitable to the individual?

Here is what happened to our 3 program participants during the 8 weeks:

PERSON 1: This person adapted to the training faster than the built-in progressions in the program made things harder. Essentially, they got better at Pull-Ups at a faster rate than the pre-determined weekly repetition increase could keep up with. As such, they stopped being properly challenged in the program just a few weeks in and finished the 8 weeks better at Pull-Ups than they started with but not as good as they could have been had the program been harder. For Person 1, the challenge from the program was what we could call “sub-adaptive training” (SAT) as it wasn’t sufficiently stressful to drive further adaptation after a certain point and they could have actually finished the 8 weeks much better at Pull-Ups than they did if they had a more challenging program.

PERSON 2: This person adapted to training at a slower rate than the built-in progressions in the program made things harder. Essentially, the program got harder than they could keep up with, and just a few weeks in they either (a) hit a plateau unable to hit the weekly reps in accordance with the program or (b) actually got worse at Pull-Ups due to fatigue or (c) got injured. For Person 2, the program was what we could call “maladaptive training” (MAT) as the challenge presented by the built-in progressions resulted in the person not adjusting adequately or appropriately to the training.

PERSON 3: By pure luck, this person adapted to training at the exact same rate that the program got increasingly difficult. The program challenged them optimally. It wasn’t too easy and SAT, and it wasn’t too challenging and MAT, it was just right. For Person 3, the program was what we could call “Goldilocks adaptive training” (GAT).

In following the program Person 1 had SAT and could have been challenged more, Person 2 had MAT and needed to be challenged less, then by absolute happenstance Person 3 ended up with GAT and the perfect bowl of porridge.

This hypothetical scenario was not done to illustrate a specific program or the ideas of 3 days per week training, or weekly increases in reps. In fact, the specifics of the program are largely irrelevant. It is also important to note that I am not suggesting that 1 in 3 people will do well get into a GAT Zone in a program like this—the 3 hypothetical people are purely for demonstration sake. I presented that hypothetical scenario in the narrative format I did to illustrate problems with Top Down Training Planning (TDTP) in general using a purposefully simplified example.

Top Down Training Planning (TDTP) is essentially forcing a rigid plan down onto a person and is to physical preparation what Marxism is to economics. This is because Marxism is a utopian pipedream based around the naïve idea that one can take all relevant factors into account and effectively centrally plan for the needs of society from the top down. TDTP, like Marxism, places the information we do know on a higher pedestal than the information we do not know (epistemic arrogance) and represents the attempted rigid imposition of expectation.

I used Pull-Ups to illustrate the problem with TDTP in a piece about distance running because:

(1) Familiarity: I hazard a bet most people reading this will know how hard Pull Ups are to perform.

(2) Easy To Quantify: The beautiful thing with strict Pull-Ups is their binary nature in that one either pulls oneself up to that bar or one does not. Miles, however, can often be run even if one is quite tired or even in a MAT scenario—the miles will just potentially be run at a slower pace and/or at a higher effort level or arousal state than ideal.

While they are inherently different movements with different physiological demands, I struggle to see how Pull-Ups and distance running are different from a training adaptation perspective. Pull-Ups are more strength and distance running is more endurance, but getting better at both involves achieving certain physical adaptations as a result of applying appropriate GAT Zone stressors to ones body over time. The stressors and the potential resultant adaptations may be different, but the fundamental principles underpinning how training in general works are not.

The Mileage Fallacy

“What’s your mileage?” is to a runner what “how much ya bench?” is to a meathead goon like myself.

I have interacted with the running community for many years on many different levels. I have runner friends. I have assisted runners with physical preparation work as a Sports Scientist. And, I have also consumed a ton of content in the form of books, podcasts, research articles etc. in order to build my knowledge about how endurance training works directly from the specialists so as to aid my own combat sports exploits as well as my work in coaching athletes.

There is, in my experience of the endurance community, quite the widespread idea that “mileage” drives performance.

A plethora of book chapters, magazine articles, and blogs etc. have revolved around the topic of what running “mileage” particular high-level runners were doing, are doing, or plan to do.

Many Sunday morning endorphin-fueled “long run” conversations and post-session “cooldowns” have also included discussions around what mileage people themselves are currently doing, did do in the past, or hope to be doing in the future.

A misconception in the running world is that hitting certain weekly mileage is what allows people to run well. The reason I think this to be a misconception is that mileage does not drive performance—performance drives mileage. This misconception is what I call the Mileage Fallacy.

The Mileage Fallacy relates to the idea that miles ran OUGHT to drive performance whereas the inverse IS the case.

(*NB: An important caveat here, of course, is that the Mileage Fallacy assumes people are trying to be in the GAT Zone as often as possible and are thereby trying to avoid both SAT and MAT Zones as much as possible. Hence, the Mileage Fallacy assumes that performance improvement is the primary goal for running rather than enjoyment/escapism/Caloric expenditure/etc.)

A 100 Mile Bowl of Porridge

Let’s suppose a hypothetical marathoner ran on average 100 miles a week in a particular preparation period that preceded a race in which their performance improved greatly.

This is not a perfect science by any means, but the fact that their performance improved greatly suggests that they were likely not in an SAT Zone, and they didn’t hit a stale plateau or get subjectively overcooked or injured which suggests they were likely not in a MAT Zone.

One can be forgiven for thinking that the fact that they run 100 miles a week is what allowed them to be so fit. However, when one acknowledges the reality of the Mileage Fallacy, one sees that they did not get so fit because they ran this volume of miles, they were so fit that they needed to run that volume of miles in order to get into their own unique GAT Zone and drive further adaptation. You see, they did not just pick 100 miles out of thin air as a number to hit because they thought they should do so, but found that this average mileage was the sweet spot that emerged from weeks based on their performance trending upward over time and that when this was exceeded undesired occurrences resulted. Let’s put some numbers on this just for example’s sake:

  1. Less mileage than that, say <90 miles a week on average may have been SAT in that it may not have provided sufficient stress to drive further positive adaptation to the same extent and so may have undercooked them.
  2. More mileage than that, say >110 miles a week on average may have been MAT in that it may have been more stress than they could deal with and resulted in overcooking them.
SAT Zone Average GAT Zone Average MAT Zone Average
<90 miles / week 100 miles / week (emergent property) >110 miles / week

Viewing training planning from the individual upward like this might be called Bottom Up Training Planning (BUTP).

Bottom Up Training Planning (BUTP) is essentially designing training that comes up through a person aiming to meet that person where they actually are and is to physical preparation what market forces are to economics. This is because market forces are based around the idea that one is unable to take all relevant factors into account and effectively centrally plan for the needs of society from the top down. BUTP, like market forces, places the information we do not know on a higher pedestal than the information we do know (epistemic humility) and represents the attempted systematic navigation of uncertainty.

GAT Zone Factors of Influence

I may be ignorant here, but there is not, to my knowledge, an odometer or GPS in the human body that records how many miles one runs and then upon reaching a certain number gives one the go-ahead to run a certain race distance in a certain time. A mile is an arbitrary man-made unit of distance measurement, not a determinant of endurance running performance capability.

For a certain runner, at a certain time point, there is a GAT zone of mileage that will be conducive to inducing an appropriate training stimulus.

What this GAT zone may be for a specific individual depends on a whole host of factors including but not necessarily limited to the following:

  • Athletes beliefs and expectations
  • Allostatic load considerations outside of training (work stress/study stress/relationship stress etc.)
  • Fitness levels
  • Training history
  • Injury history/legacy
  • Event specialization regarding distance, terrain etc.
  • Nutritional considerations (Calorie balance, macronutrient intakes, micronutrient intakes, etc.)
  • The individual genetics/epigenetics that determine adaptation rates to certain stimuli
  • The percentage of their miles that are “quality” (race pace efforts, tempo work, threshold work, hill repeats etc. etc.) as the stress that these “quality” miles may leave on the body is different than that of more “easy” runs
  • Sleep quality and duration (I have personally known and also read about high-level athletes easily hitting 9-11 hours of sleep per 24 hour period which is a far sight more than the “average” person gets)
  • Temporal relativity to races (particularly in relation to “peaking” practices)

As one can see, even with a non-exhaustive list like this there are evidently quite a few moving parts that can influence what an athlete’s GAT Zone for mileage may be. Taking all of these factors into account as a coach is likely impractical and potentially impossible, even if you realize the flaws with a TDTP approach. Therefore, we need to plan for the fact that there are factors influencing the performance of an athlete that are unknown and even unknowable.

Closing Summary

I am writing this to do my bit to highlight what I consider to be the widespread Mileage Fallacy and all of the overtraining, illness, injury, stress fractures, and poor performance that comes with forcing miles into a week rather than letting miles emerge from a week based on where that person’s performance allows their GAT zone to be. In my eyes, training planning, regardless of the sport, needs to come from the athlete upward rather than hope downward. Rigid TDTP is absurd, especially when it comes to a sport like endurance running where it is so easy to overdo things.

To further elaborate and reinforce a point I made above—there is not, that we know about, a law written into the fabric of the physical universe that allows one to run certain race times just because one ran certain mileage in training.

When it comes to training planning, what we do not know is far more important than what we do know. As such, efficient planning needs to involve the bottom up systematic navigation of uncertainty and not the top down rigid imposition of expectation.

And where does one start when trying to do this? In my opinion, by honestly paying attention to what IS actually going on with an athlete, not to what one thinks OUGHT to be going on.

Thank you to Danny Lennon for help with the editing process.

After re-reading this piece prior to publication, there are 8 figures of note from diverse fields whose work likely most influenced my thinking in the general area of uncertainty navigation that may be of interest to the reader to look up. They are as follows: Nassim Nicholas Taleb, Karl Popper, Percy Cerutty, John Kiely, Israel Halperin, Mike Tuchscherer, Eric Helms, and Marduk.

Related Articles

Self-Experiment in Frequent Training

I decided to try out frequent training (upper/lower 6x/week) with using higher volume (lots of sets) of low reps with submax intensity and effort to explode on every rep as fast as possible. Here is the Excel file with all the data and program outline. If someone wants to perform certain analysis be my guest – you can…


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

Cancel Membership

Please note that your subscription and membership will be canceled within 24h once we receive your request.