This article was originally published in September of 2014 as the first of four blog posts addressing The Principle of Specificity. Since then, the coaches at Barbell Medicine have learned a great deal more about the topic and felt that these posts deserved an update and expansion. We’ll be releasing the updated articles and corresponding YouTube videos throughout the month of October.
In an effort to show how our thinking has evolved, I am highlighting the parts of the article that are being updated using italics and quotation marks accompanied by the updated commentary in the normal font with additional clarifying content and citations where necessary. Hopefully this exercise in transparency regarding the evolution of our thoughts regarding programming is useful. As always, thanks for reading!
“In the strength and conditioning realm, specificity can be defined as how similar an exercise or conditioning modality is to another skill or set of physical traits “.
The statement above is incomplete, and has been described more thoroughly in academic journals, such as in Reilly’s 2009 where the authors define specificity as:
Specificity is a principle of training that is fundamental for securing optimal adaptation and improving performance. From a physiological perspective, a training programme must stress the systems that are engaged in performing a particular activity to achieve specific training adaptations. To stress the physiological systems associated with performance in an optimal manner, three components of specificity should be considered: (i) the energy system, (ii) the muscle group, and (iii) the skill of the sport .
Therefore, the Principle of Specificity would suggest that the more closely related the energy systems, muscle groups, and skills developed by a training program are to a specific task, the more it will improve the performance in that task. Conversely, the more dissimilar the components of a training program are to a given task, the less likely they are to improve performance of that task.
First, a brief exercise physiology review to make sure everyone is on the same page about these three components of specificity:
There are three main energy systems used to create adenosine triphosphate (ATP) and drive human activity:
- ATP and Phosphocreatine System (Alactic anaerobic)- Includes ATP already stored in the muscle and a rapid chemical reaction involving creatine kinase, creatine phosphate, and adenosine diphosphate (ADP) to make ATP without the presence of oxygen. Typically lasts 3-10 seconds and provides maximal force production.
- Anaerobic Glycolysis (Lactic anaerobic) – Uses glucose and it’s stored form (glycogen) without oxygen to make ATP rapidly, with lactate and hydrogen ions as by products of this process. Typically lasts 10-120 seconds and provides high, but not maximal, force production.
- Aerobic Oxidation – Uses glucose, muscle glycogen, fatty acids, and sometimes amino acids in the presence of oxygen to create ATP and other energy intermediates. Anything longer than 2 minutes relies increasingly on aerobic oxidation and limits force production as duration increases.
In truth, a blend of all three systems are used in all human activity and the time domains mentioned above refer to the relative contribution of one system to creating energy. For example, during a 1 rep-maximum (1RM) squat that lasts 5 seconds the main energy system being used is alactic, e.g. the phosphocreatine system and stored ATP. Conversely, during a 5k time-trial run that takes 20 minutes relies primarily on aerobic oxidation, though both anaerobic pathways contribute periodically during brief, intermittent periods of higher force production.
The muscle group component of specificity extends to not only the individual muscles, e.g. the quadriceps, erectors, abs, hamstrings, etc. used during a squat, but also specific aspects of the muscle’s contractions including the type (eccentric, isometric, concentric), velocity (fast or slow), and joint angles/muscle lengths as well [2, 3].
Stated another way, The Principle of Specificity suggests that muscular adaptations will be greatest in the specific manner in which that muscle is being trained, with progressively less transference to less similar conditions. Let’s look at a few concrete to clarify this point.
Example 1: Eccentric-focused training, e.g. the Nordic hamstring curl protocol, will have produce the greatest adaptations in eccentric strength and the structural changes occurring in the muscles that facilitate this, e.g. increased size in titin and costamere proteins and produce relatively less adaptations in concentric and isometric strength, especially when compared to training that includes or focuses on concentric or isometric-based force production [8-11]
Let’s say we have a lifter who is willing to do an eccentric only squat program. We know that most folks have about 20-30% more eccentric strength than they do concentric strength, so we let’s say that this lifter can lower 20% more weight to the safeties than he can lift concentrically during a squat . We have him do repeated singles of eccentric-only squats at his 1RM full squat, which would represent about 80% of his eccentric 1RM. We expect that his eccentric strength will improve to the greatest extent compared to his isometric or concentric strength. With that said there is likely to be at least some transference to those aspects of the movement as well. This is especially true if the lifter practices the full lift in other training sessions, and this makes a case for using tools like weight releasers in the right context.
The alternative situation is also true, whereby an isometric-focused program would principally improve isometric strength compared to concentric or eccentric force production.
Example 2: Using a decreased range of motion in a movement such as a half-squat will produce the greatest adaptations at the specific joint angles and muscle lengths used, but relatively less at joint angles that are not seen in the movement.
For example, if we have a lifter who is trying to improve their full squat, but only performs half squats in training, we would expect their results to be compromised compared to if they had trained full squats instead. Conversely, if we were trying to improve a lifter’s quarter squat or strength in the range of motion trained by the half squat because we had determined that the joint angles of the half squat were more closely related to the joint angle or muscular demands of their sport that we were trying to target, (e.g. the vertical jump), then we may actually choose to program the quarter squat and assess the outcomes rather than train for a less-specific, (potentially) resource-heavy goal like improving the full squat .
Rhea et al took 28 guys and assigned them to one of three training groups, quarter squat, half squat, or full squats that were performed in a 16-week training intervention. Strength measures were conducted in the back squat pre-, mid-, and post-training at all three depths. Vertical jump and 40-yard sprint time were also measured and it was found that squat strength improved at the specific depth trained in addition to quarter squatting having the greatest transfer to vertical jump height and sprinting speed.
I realize it is heretical to suggest that full squats may not always be optimal all the time, in all contexts, for all trainees…., but I implore everyone to consider this (and supporting evidence) with an open mind without relying on unsubstantiated anecdote.
Example 3: High velocity training, which requires rapid cycling of the contractile filaments of the muscle (actin and myosin), will produce greater adaptations for high velocity force production tasks than in lower velocity strength tasks which require less rapid cycling of those filaments. Additionally, changes take place in the primary muscles’ tendons, as well as within the antagonistic muscles, relating to muscle stiffness, relaxation properties, etc. These all limit the transference of a high-velocity movement to a low-velocity movement.
In this example, let’s say we are trying to assess the potential carry over between the power clean and the deadlift. Both lifts rely on the same energy systems and, when considering the first two pulls of the power clean only, use the same joint angles, muscles, and range of motion. That said, the movement velocity required for a successful power clean is much different than a deadlift and we’d expect a markedly lower transfer of adaptation from the power clean to the deadlift than if we used another deadlift variation in its place. Additionally, the carry over from the deadlift to the clean or snatch are also likely overstated when considering the velocity differences discussed above.
In general, the skill that is being used as the basis of comparison is either a particular movement that is tested in competition, (e.g. the squat for a powerlifter or the snatch for a weightlifter), or the skill can be the sum of athletic skills and physical traits seen in this highest level of a particular sport, (e.g. the conditioning, strength, technique, and other skills observed among gymnasts or golfers). In this regard, specificity can be used to categorize a particular exercise or conditioning protocol as being more or less sports specific based on how closely it [the exercise or conditioning] replicates that of the skill or sport being discussed.
In powerlifting, the sum of a lifter’s greatest squat, bench, and deadlift performed to the standard deemed “legal” by the judges at the meet, produces a lifter’s total. Therefore, the squat, bench, and deadlift are all very specific to the sport of powerlifting. Conversely, exercises like the press, reverse hyperextension, or lunges are all less specific.. In my view, there exists a continuum of specificity upon which we can plot exercises in order to rank a certain exercise’s relative specificity within a sport like powerlifting. I made this nifty picture below to visualize this.
So we have, in effect, laid the groundwork in comparing one exercise to another with regards to exercise application towards any particular task or sport; however, there are two more concepts that we need to discuss pertaining to specificity, the SAID principle and the Repeated Bout Effect (RBE).
The SAID principle, an acronym for Specificity of Adaptation to Imposed Demands, posits that however you train tends to yield a specific adaptation that is intrinsic to the stress applied by the training. For example, training the squat, bench press, and deadlift (i.e.-all low-velocity barbell lifts), using rep ranges of 4-6 reps is most specific in driving the adaptations, such as hypertrophy, improved force production, and neurological skills, etc., to improve performance in these specific lifts and rep ranges.
This is not to say that performing most training in the 4-6 rep range does not improve high velocity strength, power, or strength endurance, but rather that these less-specific adaptations are not as robust. “When applied to the specificity of exercise, the SAID principle suggests that if you want to squat more [weight], squat more. More what, exactly?”
In reality, the SAID is an extension of the Principle of Specificity discussed earlier. It’s logical conclusion, if you will. That said, I should’ve been more clear when discussing improving the squat in the previous paragraph. If one wants to squat more for 1 repetition, e.g. a powerlifter, then he or she will need to regularly practice squatting heavy singles and balance that stress properly with the rest of his or her program. A powerlifter may potentially waste a great amount of time trying to improve their 5RM that may or may not produce an improvement in their 1RM and if even it did, the magnitude of performance improvement is likely less than what would’ve been seen had the person trained more specifically. Put differently, if one’s 5RM goes from 500 to 520 in 6 weeks I am unsure if their 1RM has improved- especially if there is no data on how hard each effort was such as RPE, bar speed, or similar.
What’s more, is that specifically focusing on driving up a 5RM for a powerlifter, who’s test and therefore stated goal, is to increase his or her 1RM is as silly as specifically focusing on driving up a non competitive senior lifter’s 1RM. Both are wildly inefficient uses of training time and resources considering the goals of each population. It’s probably around this point where someone is yelling at their computer screen, “Yeah, well if a lifter’s 5RM goes up by 100lbs then I’m pretty sure their 1RM has gone up too!” Yes, Virginia, I agree. Large increases in performance in one strength test likely have some transfer to other strength tests, but you’re still missing the point. Who cares what a powerlifter’s 5RM is when a 1RM is all that is tested on the platform? Similarly, who cares what a non-competitive senior’s 1RM is in an arbitrary range of motion?
In the former, we should specifically focus on the objective- maximizing performance (to the extent this is the actual goal of the lifter). In the latter, I would argue that aggressively focusing on specializing in a single rep range or handful of exercises does a disservice to their overall fitness. Being hyperspecific can, at times, be deleterious to performance, fitness, and long term outcomes.
The argument against inappropriate specialization includes overzealous parents trying to get their child to specialize early in life, only to make them perform worse when it actually matters, e.g. college or the professional ranks and coaches who don’t apply the SAID principle correctly . Let’s take a sport like Golf, for instance, whose specific demands tend to gravitate towards the high skill and highly technical end of the physical requirement spectrum. If we strictly adhered to the SAID principle, we’d come to the logical conclusion that strength training, which causes a systemic general adaptive increase in force production, isn’t very “specific” to performance in golf. That would, however, be short sighted in my opinion.
I wouldn’t argue that increasing a golfer’s back squat will help his ability to pitch out of the sand at Pebble Beach by directly improving his technical mastery of the motor patterns used to execute the shot. I would, however, argue that by increasing the golfer’s ability to produce force in any and all muscles used during this shot increases the amount of force he can apply to the club, which is likely to come in handy when caught in the deep rough, facing a headwind, or when he needs to go long off the tee to setup for his second shot By increasing strength, our Tiger Woodsian golfer has more options when it comes to practically apply his highly refined motor patterns and technique, which he’s developed over years and years of golf practice. He can, if desired, choke down and blast the ball out by hitting further under the ball (into the sand) to get more loft- which he couldn’t do before since he didn’t have the greater force production (strength) to do so.
Make no mistake, I am not suggesting that the strength requirements of a golfer are the same- in absolute terms- as a powerlifter. I didn’t say that and hopefully, you didn’t read that. I am saying however, that increasing the strength athlete using methods that don’t interfere with the necessary time spent practicing the sport will make a better athlete. More importantly, this stronger athlete will be better than if he had committed the same extra time he spent training to more specific practice, as the return on investment for improving physical adaptations is quite high initially. In sum, there’s an inflection point that exists for each sport where additional technique work, which would be highly reflective of the SAID principle, is no longer indicated. Rather, the biggest return on investment for that athlete would be something that conflicts with the SAID principle -in our golfer’s case, strength training.
Repeated Bout Effect
The Repeated Bout Effect (RBE) suggests that when a person is repeatedly exposed to a previously novel movement or activity, the less muscle damage occurs when exposed to that same stimulus in the future . In other words, the more often a person squats, the less muscle damage they’ll incur from that same task unless other variables change markedly, i.e. if there’s more volume (reps x sets), average intensity of the volume (% of 1RM being used on average), total frequency (squat sessions/wk, in this example), total work (reps x weight/time) etc. This gradual accommodation, along with multiple other factors go into how much stress, fatigue, and ultimately, fitness adaptations, that a particular exercise generates for a given rep, set, and intensity scheme.
The RBE also has some carry over between similar movements, i.e. different types of curls were found to show this effect in a 2015 study by Zourdos et al. . As I understand it currently, the carry over of RBE between two exercises is more significant the more similar those exercises are. For example, a lifter who previously had exclusively low bar back squatted likely gets some RBE carryover when he high bar back squats, as the joint angles and demands are relatively similar, though a paused low bar back squat would likely have a higher RBE carryover as it is even more similar.
The main takeaway from this discussion on RBE is that in general, a person that regularly squats for the same rep range, the same average intensity, and the same variation, will accrue less muscular damage over time and through adaptive processes, incur less stress and fatigue provided no other variables change that would increase stress. Additionally, the adaptive processes secondary to the training will also increase the lifter’s recovery rate and he will get better at recovering from squats. So, a thought to ponder for next time:
What would we expect to happen to the rate of improvement in a trainee whose training stress is decreasing and recovery capacity is improving?
Thanks for reading and let us know what you think in the comments! Share the article with your friends on social media and help Barbell Medicine grow!
- Thomas Reilly , Tessa Morris & Greg Whyte (2009) The specificity of training prescription and physiological assessment: A review, Journal of Sports Sciences, 27:6, 575-589, DOI: 10.1080/02640410902729741
- Behm, D. G., & Sale, D. G. (1993). Velocity specificity of resistance training. Sports Medicine, 15, 374–388
- Aagard, P., Simonsen, E. B., Trolle, M., Bangsbo, J., & Klausen, K. (1996). Specificity of training velocity and training load on gains in isokinetic knee joint strength. Acta Physiologica Scandinavica, 156, 123–129.
- Rhea, Matthew & G. Kenn, Joseph & Peterson, Mark & Massey, Drew & Simão, Roberto & Marín, Pedro & Favero, Mike & Cardozo, Diogo & Krein, Darren. (2016). Joint-Angle Specific Strength Adaptations Influence Improvements in Power in Highly Trained Athletes. Human Movement. 17. 10.1515/humo-2016-0006.
- Jayanthi N, Pinkham C, Dugas L, Patrick B, LaBella C. Sports Specialization in Young Athletes: Evidence-Based Recommendations. Sports Health. 2013;5(3):251-257. doi:10.1177/1941738112464626.
- Hyldahl, R. D., Chen, T. C., & Nosaka, K. (2017). Mechanisms and Mediators of the Skeletal Muscle Repeated Bout Effect. Exercise and Sport Sciences Reviews, 45(1), 24-33. doi:10.1249/jes.0000000000000095
- Zourdos, M. (n.d.). Repeated Bout Effect in Muscle-Specific Exercise Variations : The Journal of Strength & Conditioning Research. Retrieved from https://journals.lww.com/nsca-jscr/fulltext/2015/08000/Repeated_Bout_Effect_in_Muscle_Specific_Exercise.24.aspx
-  Hedayatpour N, Falla D. Physiological and Neural Adaptations to Eccentric Exercise: Mechanisms and Considerations for Training. BioMed Research International. 2015;2015:193741. doi:10.1155/2015/193741.
-  Ervasti JM. Costameres: the Achilles’ heel of Herculean muscle. J Biol Chem. 2003;278:13 591–13 594.
- 10]Franchi MV, Reeves ND, Narici MV. Skeletal Muscle Remodeling in Response to Eccentric vs. Concentric Loading: Morphological, Molecular, and Metabolic Adaptations. Frontiers in Physiology. 2017;8:447. doi:10.3389/fphys.2017.00447.
-  Lepley LK, Lepley AS, Onate JA, Grooms DR. Eccentric Exercise to Enhance Neuromuscular Control. Sports Health. 2017;9(4):333-340. doi:10.1177/1941738117710913.
-  Munger, C. N., Archer, D. C., Leyva, W. D., Wong, M. A., Coburn, J. W., Costa, P. B., & Brown, L. E. (2017). Acute Effects of Eccentric Overload on Concentric Front Squat Performance. Journal of Strength and Conditioning Research, 31(5), 1192–1197. doi:10.1519/jsc.0000000000001825