One of the reasons I recommend grad school, is not for the classes, but for the informal theorizing sessions you have with classmates and professors.  Even several years after being out of school, I still look back to some of those informal after class sessions and realize how those talks were when we would get closer to the cutting edge, applied science, then in the class room.  Often times, looking back, it’s amazing how our little grad school theorizing turned out to be proven a few years later.  For instance, I always got a kick out of myself and a fellow student, Matt Andre, almost experimenting with the idea that a scoop of protein before bed and then getting up in the middle of the night to have another snack might help with recovery via keeping protein synthesis high.  Sure enough, two years later there was some good research to back up the idea.

The point isn’t to think back to those times fondly, but instead tackle another problem that I have been ruminating on since those talks: Training to failure.

The simple question of how hard should we delve into the depths of fatigue is at the core of training, yet not many athletes or coaches consider it.  It usually divides into two groups, the “no pain, no gain” group and a more cautious one.  The idea of toughness and pain is engrained into our sport, so the “no pain, no gain” mentality fits well within it.  Within this group of coaches and athletes, you routinely see athletes going, as my HS coach Gerald Stewart would say, “to the well” at the end of every hard workout.  That means, full on exhaustion and collapse.  We’ve all seen it, the athletes completely spent and exhausted after 400m reps or whatever you like.  In groups with this motto, it’s the norm.  The idea is on pushing as far down as we can go, before resting recovering and coming back out of it.

If you know the history of training, you might be familiar with Arthur Lydiard’s old saying of finish the workout (or session…since I’ve been hanging around Australians lately!) feeling like you could have done one more repeat.  So you stop the workout feeling fatigued, but
not completely spent.  There’s more left in the tank essentially.  It was still tough, but you didn’t completely drain the tank.  In programs like these, sessions are usually a bit more calculated and the workout is set for the athlete to achieve the standards, and more than likely the athlete is held back at the end from busting the last rep which might normally happen.

So, what method is “right”?  Do we have to push the depths of fatigue to the deepest in order to harden the mind and body for what it will encounter in a race or do we stop just short of that and instead nudge the body  along in adapting to slightly higher levels of stress?

Not surprisingly, I think the answer is a bit muddled in between, with a time and place for both, but where I fall on the spectrum might surprise you.  Before that, let’s try and attack this from a scientific standpoint.

The Stress of life:

I’ve often wanted to study the effects of going to the well versus not in training, as there are no research studies, and I’ve spent time over the years theorizing, and looking at the endocrine, neural, and chemical stress of certain workouts and races to get a better grasp.  We can look at the mechanisms of fatigue and how long they last post workout.  We can look at cortisol and how much it spikes, for how long, or even a Testosterone
to Cortisol ratio to look at anabolic-catabolic ratio.  We can look at other hormones that peak post workout, or heart rate variability and how it changes.  The possibilities are endless, but they still don’t tell us the answers. To me it breaks down to this:

  1.  How far do we have to push the body out of homeostasis to adapt to a particular stress?
  2. How long does it take to recover and adapt to that stress?
  3. Is it better to stack a lot of small nudging pushes out of homeostasis or a few big pushes and go for broke adaptations?

There’s no perfect solution when theorizing, but one possibility emerged when I was pondering on how to attack the problem. Look at weight training to failure or not.

Lifting: Reps to Failure

It’s a common question in the world of strength and conditioning; whether to go to failure or not.  So unlike in running, there is actually some research on the subject. Of course, there are some key differences. But what lifting does is it gives us a look at local muscular failure coupled with a high CNS stress depending on the type of exercise, but it doesn’t give us is an almost overall systems failure like would occur after a very very tough running workout.  Regardless, it’s a good way to start the search for an answer.

In looking at weight training and what happens when training to failure vs. not it’s not unlike on the track.  It’s essentially classified as training to failure leads to a gradual decrease in power or load, while not to failure means you maintain the power and load throughout and stop before you get to that point where it starts to tail off.  In the track world that could be compared to running repeats until you start tieing up versus stopping before hitting that moment.

Gorostiaga and a group of researchers tried to tackle this issue in a recent article and discussion that centered on the idea of “no pain, more gain.”  In of the studies done by the group, they evaluated the effects by doing 5 sets of 10 reps to failure, or 10 sets of doing 5 reps at the same weight but obviously not to failure as there were less reps per set and more recovery.  From a metabolic point of view, even with lifting the same weights and volume, they looked completely different.  The to failure set saw large changes in the energetic state of the muscle.  There were big increases in lactate, anaerobic energy use, and changes in PCr and ATP in the muscle.  On the other hand, the not to failure set showed only a small budge in a few of the factors.

So there was an obvious different in stimulus.  But what about functional adaptations?  In combing through the research on training to failure or not, they found:

“Some studies have compared the effects of 8 to 16 weeks of dynamic high resistance training to failure vs. non-failure in the knee extensor movement in untrained recreational weight trainers and top level male subjects. Their results indicate that high-intensity resistance training not to failure enables a favorable environment for achieving greater enhancements in maximal strength or power and muscle power output of the knee extensors when compared to a training to failure approach. The lower degree of fatigue incurred and the higher average movement velocities developed may explain the superior adaptations in strength and power gains observed in the non-failure compared to the training to failure group. These results suggest that in order to improve performance in the majority of activities with great demands of muscle strength and power of the lower extremities, a program of dynamic resistance exercise which minimizes fatigue and discomfort and is characterized by not training to repetition failure, low metabolite accumulation and excellent energy balance and cellular homeostasis, may be a more effective, efficient and safe option compared with training designed to maximize fatigue or metabolite accumulation.”

But, research has gone beyond looking at metabolic stress and adaptation.  A study from the journal of applied physiology( took a look at the hormonal reaction to lifting to failure or not over an 11 week training block.  Overall, the gains were very similar for each group, but how each reacted to a subsequent taper differed.  For instance, the to failure group tended to improve the number of reps in the bench, while the not to failure tended to increase power in the lower legs.  While it’s just a mention in this one study, it would be interesting to know whether a larger taper would be needed to bring an athlete out of the hole who has been training to failure versus a smaller taper for the more moderate approach.

In investigating the hormones, what they found was that training to failure resulted in a reduction in Insulin like Growth factor 1 (IGF-1), while the not to failure group had increased levels of testosterone, and lower levels of cortisol.  While a generalization, the not to failure seemed to spark some recovery and anabolic benefits while the stress of the to failure reps might have compromised that a bit.

 Going back to theory:

While studies on lifting give us just a glimpse, it still provides some valuable information and guidance.  First, the stress of training, even when it’s simply local fatigue like in a bench press, is enough to put an enormous stress on the body.  That stress leads to endocrine changes that could effect recovery and overreaching vs. overtraining state.

Secondly, training to failure or not gives different stimuli.  This is obvious, but probably should be stated more often.  Just because we are both doing 400s at 60sec pace with 1min recovery, doesn’t mean it’s the same workout stimuli if you do them until you collapse and get 10, while I stop early at 5, take a break and then do 5 more and feel pretty good when I’m done.  The neural, endocrine, muscular stimuli have all likely changed.  So know what you are working towards in your training.

The conclusion then has to be that whether to go to failure or not depends on multiple items, including:

  1. What the training stimulus is
  2. Recovery into and out of the workout
  3. Recoverability of the athlete
  4. Time of year/season/periodization
  5. Psychological freshness
  6. Risk/Reward

To briefly explain, the type of stimulus determines how far one has to push to get adaptation.  For instance, if it’s strength or power, then fatigue might detract from the adaptation.  If the goal is strength endurance, or prolonging our ability to maintain strength, then we have to push a little into fatigue while not going so far that we start to see very large reductions in strength output (or speed/pace).

The next factor is of course the recovery.  In order to have a big stimulus, we need a comparable amount of recovery going into and out of a session.  So the bigger the workout, the more the recovery.  That’s why it makes sense to periodize your workout days, and time between big workouts during heavy intense training periods.  Or as I like to call it, when we are hitting big specific workouts, there’s more “space” in the program.  When we’re just doing a bunch of smaller medium workouts, the training schedule is more dense.

Along those lines, how each individual responds can be incredibly different.  I tend to look at people as hyper or hypo stress response athletes.  That means whether certain stresses really knock them for a loop or whether they bounce back off of it well.

One of my coaching mentors, Tom Tellez, was a fan of telling me that if a workout was too stressful not only physiologically but psychologically, we’d just fry the adrenal cortex.  What he essentially meant is the stress response would go haywire.  That means, if we are getting so psychologically amped and then depleted just to get through a workout, eventually it will take its toll and that constant up and down of our adrenal system will wreak havoc.  We might lose motivation, drive, or whatever you want to call it.

And last but not least, is the risk versus reward.  Going for big workouts all the time may have a big short term pay off, but it often leaves the athlete fried and pushed into chronic overtraining if it’s repeated enough.

There’s no clear answer at the moment, but hopefully this will shift some of the thoughts away from just researching and looking at training to failure in terms of the bench press or other weight training, and instead onto the track.  It would be fascinating to see some of the metrics of a study.

In the end, I think it’s wise to in general do the minimum required dose to get the adaptation you need.  If we go overboard now to get an adaptation, then that limits where we can go a year or two down the line.

So what?

My instinctual feeling is that both are needed to a degree but I am very much on the side of caution and gradually stressing the body.

With my athletes, we don’t go somewhere until we need to go somewhere.  So if we only need a slightly higher stress to get to the next level, then we only go slightly higher.  Is there a need to occasionally go for the big stress and adaptation? Yes, but in my opinion, you only “see god” on 2-3 big sessions a year/season depending on the athlete.

It’s just too easy to mess up the recovery and fry the athlete, and if you are in it for the long term, then it is a safer more consistent bet to gradually introduce higher stress loads.

Although I don’t have the research for it, my anecdotal evidence is that most of the “burnout” you see is from a combined physiological and psychological draining that occur.  People often say “burnout is all psychological” but what they fail to realize is that they are interconnected.  If I keep doing workouts that spike the adrenal system, give me crazy cortisol numbers, and eventually result in an abnormal stress response, then guess what? That will radically alter my psychological state of mind and my drive.  Hormones affect cognitive function and thinking.  It’s not that they were weak at all, it’s that their bodies were crying for help with a flood of hormones and when ignored for so long, they do the last ditch thing of shutting down drive and motivation so that the person stops subjecting their body to the torture…

If we look at the research, the idea of psychological drive or motivation not only has a genetic component (read David Epsteins fantastic new book The Sports Gene for more) but also an endocrine component.  There’s a reason why serotonin, dopamine, opiods, etc. are all linked to motivation and drive.  And guess what can alter all of those for the good or bad, training.  So, if we push the systems too far, constantly train to failure, then the decrease in motivation may not just be “psychological burn out” but instead it may be that the chemical change via training hard came first.

When I was a runner in HS, my HS coach Gerald Stewart would give us a handful of “see god” days throughout the year.  We knew what that meant, and for me it most likely meant lying on the track with my head spinning at the end.  It was fun, challenging, and a part where you pushed your limit.  But they were special days.  Not a normal occurrence.  He also said, you have to “win” by far more workouts than you “lose.”  He didn’t mean, beat the times prescribed.  He meant, come out of the workout having knocked the workout out and your still standing, absorbing the training, and leave the track feeling good about it and not headed home to fall into a coma. There’s a fine line.  You certainly need both, but how and often isn’t as much as you would probably think.

Gorostiaga EM, Navarro-Amézqueta I, Calbet JAL, Hellsten Y, Cusso R, et al. (2012) Energy Metabolism during Repeated Sets of Leg Press Exercise Leading to Failure or Not. PLoS ONE 7(7): e40621. doi:10.1371/journal.pone.0040621

Get My New Guide on: The Science of Creating Workouts


    1. runninglosangeles on August 3, 2013 at 3:50 pm

      Always curious to see where you go with this idea. I was surprised. From your Fast Lane columns and basically since reading The Race Against Time piece, I thought you were headed the other direction. Thanks so much for this thoughtful work.

    2. Anonymous on August 4, 2013 at 7:49 am

      Great piece. Sadly there are too many coaches that aren't up to the intellectual muster to consider these ideas.

      Bill Bowerman was known for the philosophy of always leaving something on the track, for the next work out. To risk under training and not risk over training.

      There's an ancient runners world article or some such about how we are each an experiment of one when it comes to training, and learning to listen to the body is important.

    3. Anonymous on August 5, 2013 at 1:41 pm

      Good article, Any thoughts on bringing an athlete back once they've been "fried". Obviously rest is a factor but is it complete rest? Are you familiar with any specific nutritional/restorative methods that may expedite recovery (contrast baths, high fat diets, etc…)?
      Thank you for the great work.

    4. Anonymous on August 10, 2013 at 12:14 am

      Interesting about strength training comments. Ive done similar 'experiments'on myself where have done certain exercises with heavy resistance, few reps (2or 3) and finishing well short of failure, full rest between sets. This has resulted in significant gains without the stress and exhausted feeling from a 'to failure' type workout. Hardly a scientific study but supports what you're saying.
      My theory on this is that its the challenge presented to the muscle fibres by each individual rep that provides the stimulus, not the accumulated fatigue. So 10 sets x 1rep, with say a minute or two between sets, provides the same stimulus in terms of muscle damage (and therefore repair leading to increased strength) as 1 set of 10 reps. Without the negative of excessive fatigue. Runners drain themselves enough without doing it in the gym as well! Of course this applies to pure strength training but some of the principals can be applied to running, as youve pointed out. Hold some in reserve, 'draining the tanks' in training does just that, save it for racing and maybe one or two key workouts.
      One final thing…as a kiwi, i dont know why Lydiard was not a national hero…way ahead of his time!

    Leave a Reply