Is training to failure necessary for muscular & strength gains?

There is a lot of splitting of opinions (more like hairs) when it comes to this topic, I’d like to start by saying that I feel neither way (failure & non-failure training) is right, & neither way is wrong; they both have application, & they both have a place, & this is the most important part of this to understand, application is dependent on the circumstances.


What this article is going to explore, is training to failure NECESSARY to maximise muscular gains? & I want to start with some simple definitions so that everybody is on the same page & there are no confusions;


1.      Training to failure – “The incapacity to complete a repetition in a full range of motion due to muscular fatigue” (1) (This is the definition used in the 2006 paper by Izqueirdo et al, this & derivatives of this definition is used most commonly in the research, so we shall keep to it.)

2.    Non-failure (submaximal) training – Training of which, each set is not taken to momentary muscular failure (as per the description above).

3.    Strength – The ability to produce & express force against external resistance (barbells, dumbbells etc.) with no time conditions attached to expression.


So, the research we have currently is a bit of a mixed bag in places, with some studies in favour of training to failure, & some others in favour of not doing so. Likely the differences between results are coming from differences in methodology, subjects used etc., this tells us that there is need for further, more controlled research in this field, along with the fact that neither training to or not to failure is the best for every case, it is going to be applicable in some situations, & inappropriate in others.


We shall start the discussion with how training to or not failure affects strength gains;


To answer the question of which is most beneficial for strength improvement, we will refer to the most recent meta-analysis & systematic review on the subject, performed by Davies & colleagues (2).


& the results? … no significant difference (skewed slightly in favour of non-failure training) in gains in muscular strength were seen between the two different training approaches.


So is that it then? Not quite, there are some confounding factors that probably should be considered before making a full conclusion on the subject.


Firstly, of the 8 studies selected, only half were controlled for volume (total work done); this is of note as we now know that volume can have a significant effect on gains in muscular strength (3), with a graded dose response relationship between volume performed & gains in strength, to a point. This means that in the studies were volume was not controlled, we cannot truly know if proximity to failure was the variable that caused the results of the study, or if the results would have been different if volume was controlled between groups as it’s effects on strength are known.


 Secondly, training status varied between studies used in this review, with half of the studies subjects being trained individuals, & half being untrained; this means that conclusions made on average, may differ slightly, when we look specifically at trained & untrained individuals, although the differences are likely to be fairly small. The reason for the inability to draw conclusions, is that we know that untrained individuals have a much lower need for overall stress, to induce a positive change (4), when compared to trained individuals. Another factor to note on training status is that 3 of the 4 studies performed in untrained subjects, utilised isolation only movements (leg extensions & bicep curls), so there is a skill & intermuscular coordination component that could be potentially stripped of its contribution to the results, although this is hard to conclude, but seems plausible.


Thirdly, the difference in exercise selection between groups was notable, with some studies such as this by Kramer et al (5), using the biceps curl only, & others, such as Izquierdo et al (1), using the bench press & back squat within their training protocols.


This is important as exercises like the bench press are going to be much more fatiguing, in terms of output needed from the nervous system & physical damage to our tissues, than a biceps curl, since heavier loads are generally used, as well as the need for stability from other surrounding muscles (6). So it would make plausible sense that individuals could train biceps curls or comparable single joint movements to failure with little to no detriment to fatigue vs. training a back squat or comparable multi joint movement to failure; with the same applying to larger & smaller muscle groups, the biceps recovery curve is going to be much shorter/smaller than that of the gluteal or quadriceps muscles.


Whilst this difference is important, when we look at the differences between failure & non-failure groups, whilst no significant results were seen when all the studies were pooled, the groups that performed isolation only exercise, favoured failure training more than those with compound movements, potentially elucidating to us that training to failure, when it comes to strength, may have some place or be more applicable, when utilising single joint movements, vs. more technical movements with larger ranges of motion & more muscle groups involved in movement & stability.


In spite of some of the confounding factors, I think we can come to some conclusions on how proximity to failure will likely affect our ability to gain strength:


·      Training to failure is no more or less necessary for gaining strength, than training sub-maximally, they will both have their place under different circumstances, with some studies suggesting that more advanced trainees potentially may need to use it more often to push through plateaus, although it still should be used sparingly.

·      No matter the training status, training to failure should probably used less often than training sub-maximally, in light of the trade-offs (1) (7) in terms of risk of injury, more strain on recovery (10), more potential for overtraining & the psychological stress (13) that can come with training to failure; as there are no significant differences in results, the risks are not really worth the rewards in most cases.

·      Failure training could be used effectively in weeks prior to a deload/taper etc., so that the excess fatigue generated has time to come down before high levels of performance are needed again, & atop of this, we could possibly take advantage of the concept of functional overreaching (8) (9), with the higher overloads generated, hopefully resulting in a higher positive adaptation, in terms of strength gained.

·      For those that really enjoy pushing themselves to failure, it must be understood that in doing so, you must make modifications to the protocol if it is to be done more often, likely reductions in volume, as per volume load, failure training is more fatiguing (7) (1), & takes longer to recover from (10), so to allow adequate recover from this to continue to produce overloading training, there must be a change made to make room for this extra fatigue generated; secondly, I would recommend training to failure on smaller muscle groups, or isolation exercises for bigger muscle groups, as these muscles/movements generally necessitate less recovery, as they aren’t as overall fatiguing, & the risk for breakdown in technique causing injury is potentially lower.


These views were shared by Williamson in the following study (11), that looked at the application of training to failure in a multi-set periodized approach, so if you want to read a bit further then check this paper out..


So, you ask, what about building muscle? Do we NEED to train to failure to maximise our muscular growth? Or is it the same case as is with strength? We are going to look into some literature to see further;


Firstly, in a 2016 paper by Sampson & Groeller (12), they took 28 resistance trained males, who were put through a 12 week exercise intervention, after a 4 week familiarization block, that saw individuals train 3 x a week, using 85% 1rm, under 3 different conditions, with groups being divided into 2 non-failure groups & 1 failure group (the difference between non failure groups were the speed at which the concentric & eccentric were performed); all groups performed elbow flexion exercise (bicep curls basically) throughout.


The first thing to say is, that this wasn’t a study controlled for volume, as groups that trained to failure typically did around 2 more repetitions per set than the non-failure groups, I know that I said before that we couldn’t conclude results, with variance in volume, but I have a reason for including this study & it will become apparent.


We would probably think that, because the group that trained to failure did more volume, they would see more gains in muscle? Nope, there were no significant differences between any groups across measures of force output or cross sectional area (muscle size). What is of note as well, is that this was a study in resistance trained individuals, & whilst we can’t conclude this for sure, there may have even been more of a benefit on the non-failure side had volume been equated, as volume plays a key role in gains in muscle size (13), with a dose response relationship been seen in research.

We will also look at a manuscript by Nobrega & Libardi from 2016, titled ‘Is Resistance Training to Muscular Failure Necessary?’ (14), this paper not only looked at whether muscular failure was necessary for gains in strength & size, they also compared how training to failure & not to failure was affected by the intensity used, between high & low loads; loads of 60% 1RM & above were categorised as high load, & low loads being anything below this, dipping down to as low as 15% of 1RM.


And the result? When high loads were used, there was no difference on average in strength or size, meaning that as long as load was 60% 1RM or above, gains were predicated on how much volume was performed.

When low loads were used, there was an advantage to the individuals training to failure vs. non failure; as the authors discussed in the paper, unfortunately there were no measures of fatigue & muscle activation, so we are unsure on the mechanism behind this difference. The plausible conclusion that the author came to was that to get the highest motor unit recruitment levels, with low loads we must go closer to, or to failure, with the higher muscular fatigue levels being associated with higher levels of motor unit activation. Heavy loads likely didn’t need to get to failure due to the fact that high force output likely recruits all motor units earlier in the set/further away from failure, so in these cases they are not necessary for maximum growth.

Despite this, there were some cases in which trained individuals saw higher levels of muscle activation with training to failure vs not doing, which could potentially contribute to growth, although there was not enough backing for this to be concluded in this case, so more research is needed in this field to solidify mechanisms behind the difference.


A recent study by Moran-Navarro et al, 2017 (10), looked to determine how training to failure affected performance in the subsequent days & the time course to this recover; 10 resistance trained individuals performed under 1 of 3 conditions, either a protocol of 3 sets of squats & bench press to failure with 75% 1RM, 3 sets of squats & bench press halfway to failure (measured by bar velocity) & 1 group performed 6 sets of squats & benches halfway to failure again; the 6 set group were used to equate volume with the failure group, all groups used the same intensity.


The results showed that there was a considerably longer time to recovery for the group that took all sets to failure when compared to the non-failure sets, even when equated for volume; with groups training to failure showing larger decreases in performance, with a longer time before return to previous performance.

Athough the non-failure sets were quite a bit away from failure, we could look at this research & infer that there was probably an unnecessarily longer recovery time for failure group, as we know that when volume is equated (15) & high loads are used (14), muscular gains are no different, so the longer recovery times may only prevent the amount of total volume performed.


When it comes to the research in muscular hypertrophy, only 1 man really fits the bill to take a wholesome approach to finding out what we know, & what we should do to know more, this man is Dr Brad Schoenfeld, so I’d like to reference his 2010 review titled ‘The Mechanisms of Muscle Hypertrophy & Their Application to Resistance Training’ (13) to help us delve further into this topic.


In this paper, Schoenfeld states that it is commonly believed that training to muscular failure is necessary to maximise hypertrophy, with the theory being this thought being linked to increased motor unit recruitment & the increased metabolic stress, with sets taken to failure being exposed to higher levels of metabolite accumulation within the muscles being worked.  


This metabolic stress is caused by training under conditions of anaerobic glycolysis (of which glucose is the primary source being used to produce ATP in the absence of oxygen), this absence of oxygen causes a vast build-up of metabolic by-products within the muscle, that have been seen to elevate certain hormone levels post exercise, potentially being linked by itself as a contributor to growth in muscle size.



In spite of all this, Schoenfeld weighs in on the downsides of training to failure, with links between training to failure & overreaching & psychological burn out, with evidence showing reductions in resting IGF-1 & testosterone levels over a 16 week protocol, that are highly relevant when it comes to building new tissues; Schoenfeld finally concludes that training to failure is something that can be implemented into most training protocols, & may infer its own benefits, but it should be done in a periodized fashion to reduce the risks of entering an over trained state.


So, I think when we look across all we have seen so far, we can potentially conclude a few things;


·      Training to failure can be done safely & effectively, as we have seen lots & lots of individuals training in this fashion & achieving great results; where the problem lies, is in the over application of training to failure, or doing so at inappropriate times; we should always look to periodize training to failure (11), & avoid doing so when already in a highly fatigued state, or prior to a state in which we will need to be highly ready to perform (10), as this will put us at an unnecessary risk for accumulating too much fatigue, risking overtraining &/or injury (1) (7).

·      Training to failure is not necessary to maximise muscular gains (12) (14), especially so in beginner & intermediate level trainees, & whilst there is still some contention as to whether there can be benefits to training to failure for advanced individuals, when we look back at what we have discussed so far, the potentially risks really don’t outweigh the unknown benefits in most cases, so training to failure should be used carefully & methodically to reap any potential benefit there is from it.

·      When we use loads of 60% 1RM & above, gains in hypertrophy are typically predicated on the amount of volume someone can perform (14) (15), recover & adapt from, so training to failure can be used effectively within a periodized programme, that moves between different intensities, allowing individuals time away from heavy loading to recover from this particular stress, especially for those with joint pain or history of injuries; we can do this & still see levels of muscular gain comparable to those with high loads.

·      As we discussed in the strength section, some individuals enjoy training to failure, especially those in the bodybuilding, physique & strength training realm, reporting that they can only train when they are pushing themselves to their limits; training to failure can definitely be utilised effectively & safely, & even with the trade off in amount of work you can perform & recovery times (10) considered, levels of muscular gain are no different (12) (14), so it is going to be a decision of personal preference, smart programming & effective recovery.


Thank you for reading this far, this one was a little bit longer than our others, I only hope it was insightful; I wanted to start to cover subjects a little more fully, & try to bring those who are scared of research, a bit closer to how our textbooks are written, & how we devise & revise strategies.


As always, feedback is LOVED, so either email us, message us or comment on the post related to this article.


Aaron @ Myonomics




(1)         Izquierdo, M., Ibañez, J., González-Badillo, J., Häkkinen, K., Ratamess, N., Kraemer, W., French, D., Eslava, J., Altadill, A., Asiain, X. and Gorostiaga, E. (2006). Differential effects of strength training leading to failure versus not to failure on hormonal responses, strength, and muscle power gains. Journal of Applied Physiology, 100(5), pp.1647-1656.


(2)         Davies, T., Orr, R., Halaki, M. and Hackett, D. (2015). Effect of Training Leading to Repetition Failure on Muscular Strength: A Systematic Review and Meta-Analysis. Sports Medicine, 46(4), pp.487-502.


(3)         Ralston, G., Kilgore, L., Wyatt, F. and Baker, J. (2017). The Effect of Weekly Set Volume on Strength Gain: A Meta-Analysis. Sports Medicine, 47(12), pp.2585-2601.


(4)         RHEA, M., ALVAR, B., BURKETT, L. and BALL, S. (2003). A Meta-analysis to Determine the Dose Response for Strength Development. Medicine & Science in Sports & Exercise, 35(3), pp.456-464.


(5)         Kramer JB, Stone MH, O’Bryant HS, et al. Effects of single vs. multiple sets of weight training: impact of volume, intensity, and variation. J Strength Cond Res. 1997;11(3):143–7.


(6)         Nisell, R and Ekholm, J. Joint load during the parallel squat in powerlifting and force analysis of in vivo bilateral quadriceps tendon rupture. Scan J Sport Sci 8: 63–70, 1986.


(7)         Fry, AC and Kraemer, WJ. Resistance exercise overtraining and overreaching: Neuroendocrine responses. Sport Med 23: 106–129, 1997.


(8)         Fry, AC. The role of resistance exercise intensity on muscle fibre adaptations. Sport Med 34: 663–679, 2004.


(9)         Volek, JS, Ratamess, NA, Rubin, MR, Go ́mez, AL, French, DN, McGuigan, MM, Scheett, TP, Sharman, MJ, Ha ̈kkinen, K, and Kraemer, WJ. The effects of creatine supplementation on muscular performance and body composition responses to short-term resistance training overreaching. Eur J Appl Physiol 91: 628–637, 2004.


(10)    Morán-Navarro, R., Pérez, C., Mora-Rodríguez, R., de la Cruz-Sánchez, E., González-Badillo, J., Sánchez-Medina, L. and Pallarés, J. (2017). Time course of recovery following resistance training leading or not to failure. European Journal of Applied Physiology, 117(12), pp.2387-2399.


(11)    Willardson, J. (2007). The Application of Training to Failure in Periodized Multiple-Set Resistance Exercise Programs. The Journal of Strength and Conditioning Research, 21(2), p.628.


(12)    Sampson, J. and Groeller, H. (2015). Is repetition failure critical for the development of muscle hypertrophy and strength?. Scandinavian Journal of Medicine & Science in Sports, 26(4), pp.375-383.


(13)    Schoenfeld, B. (2010). The Mechanisms of Muscle Hypertrophy and Their Application to Resistance Training. Journal of Strength and Conditioning Research, 24(10), pp.2857-2872.


(14)    Nóbrega, S. and Libardi, C. (2016). Is Resistance Training to Muscular Failure Necessary?. Frontiers in Physiology, 7.


(15)    Schoenfeld, B., Ratamess, N., Peterson, M., Contreras, B., Sonmez, G. and Alvar, B. (2014). Effects of Different Volume-Equated Resistance Training Loading Strategies on Muscular Adaptations in Well-Trained Men. Journal of Strength and Conditioning Research, 28(10), pp.2909-2918.


Aaron Brown