Getting Started With (I) – The US Army Combat Fitness Test

The GSW series presents focused applications of Analytic Fitness™ with a 2.500-word cap (excluding theory and analysis, kept in asides and footnotes, with no set limit) for an estimated reading time under 15 minutes.

Getting started with the US Army Combat Fitness Test

If you ask me, the most important piece of fitness news of 2018 was the inception of the US Army Combat Fitness Test.

Granted, shoving unsolicited opinions in other people’s face is an asshole thing to do (multiple-level pun intended). But my take on the ACFT is not so much an opinion as it is a logical consequence of arguments I’ve presented in this blog already. And so I figured that I could submit it to your appreciation.

ACFT events (from the official US Army website for the ACFT)

I’d have only good things to say about the ACFT as a military test, especially compared to its predecessor the Army Physical Fitness Test (APFT: a 2-min max sit-up test, a 2-min maxi push-up test, and a 2-mile run). But this post is not concerned about the history of the US Army physical fitness culture. If you are curious about how the ACFT fits in this history, I’d recommend this post by Cpt. Will Fuller and Cpt. Sally White [1].

Rather, I’d like to defend the proposition that anyone training for general strength and conditioning could use the ACFT as a benchmark for progress. Given what the ACFT includes and leaves out, I’m not expecting instant agreement. But I’m going to dodge that bullet early since training for the ACFT is not the same as training the ACFT.

In fact, you can train for the ACFT and still bench, squat and overhead press, or even power clean if that’s your thing. But that’s really a marginal topic, so I’ll keep it for the margins. With a 2500-word cap, I’ll stick to the essential, according to the following plan:

  1. Military vs “General” Fitness. How the military-specific tasks tested by the ACFT relate to “general” strength and conditioning.
  2. Training the ACFT? Be it for military or civilian purposes, the best way to train for the ACFT is not the ACFT itself.
  3. How to be ACFT-functional. The ACFT implies a definition of “functional fitness” worth training for and there are many ways to do that.

I’ll conclude on how to extend the current ACFT standards to the civilian population in practice because they sure don’t apply to those populations in principle.

Military vs. “General” Fitness

The ACFT tests a wide array of physical qualities contributing to performance in non-specific combat tasks, which makes it an equally valid civilian fitness test.

By “non-specific combat tasks” I’m not referring to any official US Army nomenclature but to the fact that the ACFT does not test physical qualities specific to combat roles (medic, communication, EOD, etc.). The US Army official YouTube video below showcases how the ACFT events relate to combat applications and illustrates their non-specificity (in the above sense).

The Army approved a new physical fitness test for all Active, National Guard and Reserve Soldiers that is highly correlated with combat readiness. The Army will execute the test through U.S. Army Training and Doctrine Command and the U.S. Army Center for Initial Military Training. Learn more at: #ACFT #ArmyFit (From the YouTube description box.)

Let’s go one step further, and note that there is nothing exclusive to military fitness in the ACFT. The notion that some physical tasks would be “exclusively military” is questionable but of marginal concern here (but see below Exclusively Military Fitness). The physical qualities tested by the ACFT are indeed the following:

  • Full-body maximal strength (3-repetition maximum deadlift [MDL]).
  • Full-body power and coordination (standing power throw [SPT]).
  • Muscular endurance (hand-release push-up and leg tuck [HRP, LTK]).
  • Mobility, speed, stability & anaerobic fitness (sprint-drag-carry [SDC]).
  • Cardiovascular endurance (2-mile run [2MR]).

The above correspondence between test events and physical qualities somewhat differs from those of the US Army’s website but the differences are marginal and easily explained (see Nitpicking). The take-home message is that for each ‘military’ application of physical attributes tested by the ACFT there is an equally relevant ‘civilian’ application of the same attribute. An exhaustive list would be an unnecessary use of words, but you can think of lifting a car to free a crash victim (MDL), playing ball and monkeying around with a bunch of kids (SPT, HRP and LTK), carrying heavy grocery bags in a parking lot without worrying about asshole drivers because you can jump out of their way (SDC) and climbing stairs without spitting your lungs out after reaching the second floor (2MR).

If you ask me, being able to do all of the above is as close as it gets to being a “functionally fit” individual. Accordingly, I’d take the ACFT to be an almost perfect test of “functional fitness” (be it military or civilian), where “almost” registers no serious issue, but merely reflects some practical compromises in the test design (see next section).

Exclusively Military Fitness.

There is nothing “exclusively military” in military fitness tests such as the French Foreign Legion’s 2-day, 50-km Marche du Képi Blanc or the British Royal Marine Commando 1-day, 30-mile (about 50-km) Dartmoor yomp (see OSS (III): Ad Optimam Valetudinem Fingendam, “Galea, Kepi Blanc and Coveted Green Beret“). Nothing other, that is, than the level of fitness considered to be an acceptable baseline. Civilian ruckers can indeed try and compare their performances to the soldiers’ performance as long as the weight of their backpacks is the same as the soldiers’. Thus, there could very well be a civilian version of the Marche du Képi Blanc indistinguishable of the military one in practice, because the only relevant parameters would be the weight of the backpack and the distance covered. And there is actually a civilian version of the Dartmoor 30-mile Yomp which is indistinguishable from the military one when considering the same relevant parameters. By contrast, the US Army 61-day Ranger School provides extreme endurance training that may appeal to amateurs of extreme sports together with military-specific tactical training under fatigue and caloric deficit. Accordingly, a “civilian Ranger School” would be Ranger School in name only. In that sense, the US Army Ranger School could be viewed as “exclusively military fitness” training but the exclusive military-ness stems more from the interplay of physical and tactical skills than from the specificity of physical skills.


Here are the grounds for my identification of physical attributes tested by the ACFT:

  • MDL. It is generally accepted that 3RM strength transfers to 1RM strength. It is also more stable than 1RM strength (i.e. less sensitive to fatigue, arousal, etc.).
  • SPT. The name “power throw” is a dead giveaway and the execution of the movement requires full-body coordination. Arguably, the implement used in the ACFT is too light (5kg) to train whole-body power. But it’s enough to display it.
  • HRP/LTK. Both events are cyclical and incorporate built-in rest (the release phase in HRP and the dead hang in LTK) allowing in principle to tap into the ATP/CP and aerobic (citric acid) systems rather than the glycolytic one (see AFD: Energy Systems). This qualifies as muscle endurance if anything does.
  • SDC. The lateral shuttle count as mobility (albeit very basic) as does the sled drag (for the backward component); the sprint tests speed (kind of obvious); the sled drag and the kettlebell carry add a mechanical stability challenge; and completing the whole 5*50m do not provide with enough rest not to pay back the oxygen debt, so anaerobic (glycolytic) it is.
  • 2MR. The 2-mile run is a proxy for VO2max test and VO2max is the best indicator we have of systemic cardivascular fitness (and endurance).

In the nitpicking department, I could register here minor qualms about how the correspondences between the ACFT events and their combat applications as stated in the official US Army YouTube feature. For instance: that the HRP is “similar to […] repetitive loading of ammunition and supplies” [1:57]; or that 2MR “replicates movements to and from contact” [3:16] (for the record, the first application is better tested by MDL+SDC and the second by SDC). Then again: (1) the official US Army website for the ACFT does a better job at stating correspondences and in particular does not claim similarities or replications (but I’d still nitpick about relating 2MR to rucking, since a US Army soldier’s combat load exceeds 1/3 of their bodyweight); and: (2) overall, the ACFT tests the physical aptitudes a soldier needs to perform any of the tasks listed as “applications” irrespective of the event listed as the closest to the application.

Training the ACFT?

As good as the ACFT is as a test of functional fitness it would actually make a rather poor workout.

[The APFT] was less a test of comprehensive soldier fitness than it was a measure of how often one performed pushups, sit-ups, and a two-mile run. 

Cpt. W. Fuller and Cpt. S. White.[1]

Some of the discussions surrounding the ACFT gravitate around the notion that “the test is the workout”, a remnant of the philosophy behind the ACFT’s predecessor, the APFT. The quote from Fuller & White sums up quite well the consequences of this earlier philosophy. Then again, the APFT and the ACFT still have in common to be poor workouts overall even if the latter is a terrific test of functional fitness and the former potentially dangerous crap not so much.

Now, I need to clarify and nuance the latter claim in at least two ways. First, the ACFT would be a much better workout than the APFT ever was. Without getting into too many details, the APFT was ill-conceived and potentially dangerous, while the ACFT is well-designed and safer to perform (for details, see Meet Push-Up and Sit-Up Dumb and Dumber).

Second, training ACFT events is not the best way to train for the ACFT. Actually, I’m about to go out on a limb here and propose that, with a one-and-fifth exception, if the goal is to max out the test, training the ACFT events is less efficient than not training them, the exception being MDL and a fifth of the SDC (Kettlebell carry). The reasons differ on an event-by-event basis and in the case of SDC, on a component-by-component basis. Assuming that the events are not trained as part of the same ‘”mock test” workout and that they are trained with more volume but at the same intensity as in the test[2], they sum up as this:

  • Some events are trade-offs. HRP, LTK, and 1/5 of SDC (side shuffle) are not optimal for the aptitudes they are testing but require less skill, specific training, and/or equipment than better options.
  • Some events are too easy to accommodate to. The SPT 10-pound med ball (4.5 kg) and the SDC 90-pound sled (about 45 kg) are too light to yield a training effect by ramping up the volume only.
  • Some events are best not trained ‘as is’. Throwing med balls overhead and practicing sprints, side shuffles, drags, and carries in anaerobic conditions are not the to train SPT and SDC (respectively).

The first bullet point is already covered (for details, see Meet Push-Up and Sit-Up Dumb and Dumber again). The second bullet point should be rather obvious for anyone familiar with the Law of Adaptation and, in the case of SDC, warrants some cool parallels with old school strength (see Learning from the Legion). Finally, the third bullet point is a direct consequence of the greater efficiency of aglycolytic training for anaerobic fitness (see the conclusion of AFD – Energy Systems). The latter deserves more than an aside, however, and is developed in the next section.

Meet Push-Up and Sit-Up Dumb and Dumber.

Push-Ups are a terrific exercise when properly done. Just in case you wonder, a properly done push-up halfway position should look like the picture below that I pulled from the excellent GMB Push-Up Tutorial.

Now, take one minute to try and remember the last time you saw someone looking like that halfway into the 10th push-up and before they’d bang 10 others. Or 20. Or whatever. See, what did I tell you. On the other hand, gals and guys who can do proper push-ups with the right amount of tension rarely do them by the dozens unless they’ve specialized in calisthenics. HRP solve the problem of poor mechanics by requiring more tension than regular Push-Ups at the bottom to actually end up looking like an actual push-up rather than a Cobra Pose (Bhunjangasana) and the presence of a referee during the ACFT does the rest. Barring similar countermeasures, max push-ups are pretty dumb. Max sit-ups are even dumber. Unlike push-ups, which (when properly done) actually train the muscles involved according to their function (triceps as movers, chest, back, and abdominals as stabilizers) while providing isometric tension in the legs, sit-ups train trunk flexion, a secondary function of the abdominals, when they train it at all. Indeed, someone with a weak ass inhibited gluteal and abdominal wall complex (for instance, from sitting too much) would default to hip flexors, including the psoas, a muscle whose function is poorly understood but so closely tied to back pain that biomechanists such as Stuart McGill recommend avoiding movements that stress it. There are better ways to train the abs and to test their endurance than sit-ups. Now, trunk flexion is important to soldiers and the LTK tests it much more safely in part because the contraction of the lats is protective, and in part due to the restricted range of motion. Furthermore, proper training for MDL and SDC actually strengthens the primary function of the abdominal wall and guarantees that soldiers won’t have weak asses can get away with 2-min of trunk flexion without excessive risk of injury.

Learning from the Legions.

In an aside of OSS (VI): Athleta Ex Oblivione, I made the case that the training effect of some of the public work performed by Roman legionnaires could not depend on increasing the intensity of effort (in the sense of footnote [2]) because the load-per-repetition would have been too constrained by regulations. With a fixed load, the only way to apply the Principle of Overload is through density, which for all practical purposes can be defined by the following formula: D=I/T where D reads “Density”, I reads “Intensity” (in the sense of footnote [2]) and T reads “Time”. However, there is an upper limit to the speed improvements that an individual can hope for in a 50-m run with a 90-pound sled (this may vary between individuals depending on the length of their limbs, muscle fibers make up, etc.). Therefore, increases in density can only delay accommodation for so long. This is not necessarily a bad thing if one ends up accommodated to a level of effort that would be optimal for one’s occupation. Back to the ACFT, 2*40-pound kettlebells are ok to accommodate a soldier to carry ammunition boxes but may be too light for, say, a heavy machine gun and a box of ammunition belts (please don’t nitpick here, I’m making the example up); and a 90-pound sled is enough to accommodate a soldier to carry a wounded comrade back to safety with the help of another soldier, but not on their own. Hence, the fixed loads for the ACFT are best viewed as the lowest-level performance an army soldier should be accommodated to rather than an optimal level of performance. Then again, there is some transfer from the ability to perform at a higher speed at a given intensity to the ability to perform at a higher intensity (at a possibly lower density), as demonstrated for instance by Compensatory Acceleration Training (see this post on T-Nation for a digest). Therefore, training with the test-weight while increasing density can in principle carry over to performance with heavier-than-test-weight.

How to be ACFT-functional

Designing a program to train for the ACFT would be an interesting intellectual exercise, but there are better things to do.

The first section suggested that adequate performance in the ACFT is tantamount to being a “functionally fit” individual. How “adequate performance” should be defined is an analytical sideshow but it basically amounts to assuming that the US Army minimum is a good starting point for minimal general civilian fitness within certain age parameters (I’ll come back to age groups in conclusion).

ACFT scoring system as of July 2018
(source:, right-click and open in a new tab for easier reading)

Even assuming this guesstimate, the variety of opinions about “functionality” keeps this section open to too much nitpicking. So let’s relativize “functionality” to the ACFT and refer to it as ACFT-functionality. With that done, we can ask: “what minimal constraints should a training program satisfy in order to guarantee ACFT-functionality?” This question is more general than: “what would be a good training program based on the ACFT?”.

As of today, there are already adequate answers to the second question but (to my knowledge) no satisfying explicit answer to the first [3]. However, there are at least two partial answers related to one another. Both answers are conditional on the following assumption:

Carry-Over Assumption (COA) Any program optimized for rucking with a load greater than 1/3 of the rucker’s bodyweight will improve performance in the ACFT.

The grounds for (COA) are that rucking performance under that kind of load depends less on aerobic capacity than on strength, mechanical stability, and lactate threshold. With an appropriate selection of exercises, improving lactate threshold (which is specific, see the AFD – Energy Systems) also improve muscular endurance in the body parts trained by the exercises selected. If you are already a reader of The Older Avocado, the above should suffice to jog your memory. If you are not (or need a refresher) there’s a digest on rucking.

On the surface, (COA) merely increases the range of available answers to the question about programs that can increase ACFT-functionality: to ACFT-specific programs (cf. footnote [3]), we can add rucking-specific programs. Then again, some rucking programs are based on an analysis of the constraints a training program must satisfy to improve ruck performance under a load over 1/3 bodyweight. Therefore, by (CAO), “analytic” rucking programs implicitly answer the question about the analysis of ACFT-functionality. Two examples of such programs would be:

  1. Mike Prevost’s Rucking Programs based on an analysis of: (1) basic movement patterns: upper body vertical & push upper body pull, hip hinge, squat, and loaded carries) and: (2) types of work: Strength, Metabolic Conditioning (cross between power and lactate threshold training) and Endurance.
  2. The Roman Legions physical training based on load-bearing marches, castrametation and/or public work, and weapon & tactical training.

Of course, (1) is based on systematic and explicit analysis, while (2) was based on reasoned practice and tradition. Then again, late-Empire author Vegetius (4th century CE, see OSS (II): Nemo Metuit Facere...) compiled recommendations for military training based on practices of the Republican era that are evidence of older systematic approaches to combat readiness. I conclude this section with one of Mike Prevost’ sample rucking programs based on a 2-week alternating template, for comparison with the US Army ACFT prep programs (cf. footnote [3]).

An Analytical Sideshow.

Defining the notion of “adequate performance” is tricky, as there are many options, with their pros and cons. Assuming that performance in the ACFT is given by a compound score (like a weighted average of scores for individual event), the options I’d favor would be the following three:

  1. Normative minimum. There should be a reference value, or norm, in the form of a minimum performance for ACFT-functionality for every individual in a given reference class (age group, gender, etc.).
  2. Context-dependent. The norm should be further adjusted according to contextual parameters: the same individual need not be ACFT-functional in all contexts.
  3. Empirically motivated. The norm for a given reference class in a given set of contexts should be grounded in statistical data rather than in informed opinions or educated guesses.

Constraint (1) allows answering yes-no questions about ACFT-functionality: one is ACFT-functional above a given compound score, and not so below. Constraint (2) reflects that ACFT-functionality is ultimately to be thought of in term of transfer or carry-over in different contexts. Finally, constraint (3) is meant to avoid the pitfalls of top-down definitions that could result from trying to make (1) too general.

An example of (1+3) would be defining the level of performance for the elderly that best predicts a lower risk of all-cause mortality. An example of (1+2) would be defining “US Army Grunt” ACFT-functional as the ability to carry 2 ammo boxes or drag a wounded comrade with the help of a fellow soldier at a given speed, and “US Army Sniper” ACFT-functional as the ability to carry a weight equivalent at least a fully loaded Sniper Weapon System (7.3kg) together with a ruck or weighted vest of a weight equivalent to body armor, spare ammo, secondary weapons, and ghillie suit (which requires adjustments to the test). Finally, an example of (1+2+3) would be to re-define “US Army Grunt” ACFT-functional based on data pertaining to the relationship between risk of injury during the performance of the ACFT and the and various training protocols for training for the test; or define “US Army Sniper” ACFT-functional based on empirically-established transfer between performance in the “vanilla” ACFT and performance in higher-load tasks.

The current US Army scoring system (as of July 2018, and still used as far as I can tell) readily provides with context-dependent, empirically-motivated norms for the ACFT. Interestingly, those norms do not differentiate by age, gender, height or weight (bullets coming at you on the battlefield won’t). But they do differentiate by military occupational specialty (MOS) which amounts to something similar to my US Army Grunt vs. Sniper example. As for the general population, I’d start with the US Army minimum for the general population within the 18-55 deemographics, and go from there with trial-and-errors with younger or older people, until some civilian study becomes available differentiating along parameters of age (more on this in conclusion).

A digest on rucking.

Rucking (US)/yomping (UK) denotes load-bearing marches, both civilian and military. Military rucks/yomps are sometimes also called “tabs” (UK) from the acronym of Tactical Approach to Battle. The existence of a cut-off value at 1/3 of one’s bodyweight has been established by NATO-funded research (see Mike Prevost: Ruck Training Programs – Part 1 for a summary). Given the relation between heavy rucking/yomping performance and strength and conditioning, accommodation to rucking/yomping with heavy loads, at a fast pace, and over long distances (say, greater than 30 km/18 miles), can be taken as a hallmark of elite military units, such as modern-day French Foreign Legion or British Royal Marine Commandos, as well as Roman Legions from the late Republic and the Marian Reforms (107 BCE) to at least the end of Hadrian’s reign (138 CE) before the generalization of auxiliary troops (see OOS (III) – Ad Optimam Valetudinem Fingendam). Roman Legions of this period also offer an example of a comprehensive strength and conditioning system organized around the regular tasks of army life, including skill training (see OSS (V) – Poscam Bibere, Caseum Edire, Sub Sole Otiari and OSS (VI) – Athleta Ex Oblivione).

Wrapping up: Setting a civilian standard

The ACFT is a fantastic test of general fitness but its civilian applications require some careful thinking.

Thanks to the breadth of physical qualities the ACFT tests for, and the extensive research the US Army has carried prior to selecting the ACFT events, the ACFT has the strongest claim to being a science- and evidence-based test of general strength and conditioning any fitness test has ever had.

However, the test is responsive to contextual parameters set so as to reflect the reality of tactical deployment, hence it’s being age-, gender- and anthropometry-neutral (again, because incoming bullets are). But by the same token, the US Army research does not support civilian applications of the ACFT, in particular, because civilian populations vary more in age and anthropometry than military ones, and the US Army research does not consider extensions to those populations.

Then again, the ACFT standards are normative, context-sensitive and empirically motivated, and so the methodology underlying the design ACFT can be extended to obtain “civilian” standards sensitive to parameters military standards must be blind to (cf the analytic sideshow). Sure, this methodology has still to be so extended, leaving us to our educated guesses. But given what we know about the US Army methodology, they are higher-education guesses.

At the beginning of How to be ACFT-functional, I promised that I’d come back to age parameter in the conclusion, and here we are. Well, it’s time for a higher-education guess. Here go the premises:

  1. the 18-55 civilian demographic matches the class of tactically-deployable Army personnel;
  2. the Army minimum is a minimum for able-bodied (aka tactically deployable) military personnel in Military Occupational Specialties (MOS) with low physical demand.
  3. army personnel who pass the test with the Army minimum can help lift a car to free a crash victim, monkey around with their kids, carrying heavy grocery bags in crowded parking lots while dodging asshole drivers, and climb stairs without spitting out their lungs.

By (2) and (3), the Army minimum for personnel in MOS with low physical demand is actually desirable for a civilian. By (1) and (2) able-bodied army personnel in MOS with low physical are within the 18-55 demographic. Therefore, by (1), (2) and (3), the Army minimum for personnel in MOS with low physical demand is desirable for civilians within the 18-55 demographic.

That was easy, wasn’t it? So much for higher education, then. Now, what of young punks and old farts younger people and older ones?

Well, it’s not much harder: reaching the Army minimum would correspond to physical maturity (young) and maintaining it would correspond to physical autonomy (old). In the latter case, one could also devise longitudinal studies to refine the Fitness Hypothesis using the ACFT as a standard fitness test within the 18-55 demographic to predict performance in toned-down variations of the ACFT appropriate for the 55+ demographic.

That would be quite an improvement upon testing leg extension and grip strength.


[1] Full reference: Captain Will Fuller and Captain Sally White, “So you’re telling me there’s a chance: observations from the Army Combat Fitness Test pilot”, War on the Rocks, accessed February 21, 2019.

[2] The common definition of intensity in exercise science studies is I=Wt/R, where I reads “Intensity”, Wt reads “Workload (total)” and R read “Repetitions”. In the case of loaded events of the ACFT with fixed weight (SPT and SDC drag and carry), Wt is obtained by multiplying the weight of the implement (10-, 90- or 2*40-pound) by the number of throws (SPT) or 50-m runs for (SDC dag/carry) performed during a workout, the latter giving R immediately. Given this definition, one would perform at the same intensity as the test as long as one sticks to 10-pound med ball throws and 50-meter 90- and 2*40-pound drags/carry.

[3] For an answer to the second question (about programing) you can check the US Army-issued Training Aid for the ACFT in July 2018 (the link is to a re-upload on and may or may not be an up-to-date version) which provides with a one-month, 5-day/week, 60-min/session prep for the test (pp. 38-39) and a more condensed 14-day, 6-day/week, 30-min/session prep (p. 40). For an attempt to answer the first question (about analysis) you can check the non-US Army related website which offers assorted resources if you sign-up with Facebook, Google or email, including a training ebook analyzing and quantifying visually the demands of each ACFT event. However, the principles of the analysis are left implicit, and the method for comparing and quantifying the demands for various physical qualities is never stated, and they probably boil down to the author’s gut feelings. Since said author remains unidentified, there’s no way to gauge their competence, and subsequently, the value of their analysis.

Leave a Reply