Why we test what we test.

The Army Physical Fitness Test was implemented in 1980 and remained the Army's primary fitness assessment for forty-two years. Three events — two minutes of push-ups, two minutes of sit-ups, and a timed 2-mile run — scored against age- and sex-graded tables that yielded a 300-point maximum. For most of a soldier's career, the score on those three events was the single most legible number attached to their physical condition, and a 270 was the floor where the conversation about promotion really started. The APFT was not built from a sports-science consensus. It was built from a 1970s effort, drawing on Cooper's aerobic test research and the Army's own institutional history of calisthenic tests, to produce a fitness assessment that could be administered at scale by any NCO with a stopwatch and a flat half-mile track. That was its strength. It was also, in the end, its weakness — what scaled across a force of half a million people did not necessarily measure what mattered for combat.

The three events were chosen because each maps to a distinct physiological capacity. Push-ups load the chest, anterior deltoid, and triceps in a closed-chain endurance task — measuring muscular endurance of the pressing muscles. Sit-ups, performed with feet held and hands behind the head, load the hip flexors more than the abdominal wall, which the original program acknowledged and the critics never let go of. The 2-mile run is a submaximal aerobic event in the 12-to-20-minute range, sitting near the boundary of aerobic capacity and lactate threshold, well-suited to differentiate trained from untrained runners but a poor proxy for sprint or carrying capacity. Taken together, the APFT measured general physical fitness — the same things a high-school PE class measured — and did so consistently. Test-retest reliability was acceptable. Equipment cost was nearly zero. Administration time per soldier was under thirty minutes. From a force-management standpoint, that combination is hard to beat.

It did not measure anaerobic power. It did not measure carrying capacity or grip endurance. It did not measure asymmetric loading, change of direction, or the explosive single-effort movements that show up in real infantry tasks — getting a casualty onto a litter, throwing a grenade, breaking contact under load. By the late 1990s the gap between what the APFT scored and what soldiers actually had to do was widely understood inside the Army, but the institutional cost of changing it was higher than the cost of leaving it in place. The sit-up specifically became the locus of medical criticism. Multiple US Army Research Institute of Environmental Medicine (USARIEM) studies and broader sports-medicine reviews flagged repetitive spinal flexion under load as a contributor to low-back injuries, and Stuart McGill's spine-biomechanics work — widely cited by both military and civilian strength coaches by the 2000s — became the most frequent argument for replacing the sit-up. The Army knew. The replacement still took two decades to ship.

Three forces kept the APFT in place far beyond its evidence-based shelf life. First, the cost of change at scale — recalibrating a force of half a million active soldiers, plus Guard and Reserve, plus academy and ROTC pipelines, plus the BCT and AIT cadres that have to teach the new standard, is enormously expensive in time and training hours. Second, the political risk of a new test that suddenly fails large numbers of currently-passing soldiers, or suddenly passes large numbers of currently-failing ones, is a career risk for the senior officer who signs the policy. Third, the absence of a forcing function — no high-visibility incident or oversight pressure made keeping the APFT untenable for most of those forty-two years. That changed in the mid-2010s, when the combination of the post-Iraq / Afghanistan force review, the integration of women into combat arms (2015-2016), and a growing body of operational-physiology research from USARIEM and the Center for Military Health Policy Research at RAND made the case for an event-by-event replacement. The Army moved.

The Army Combat Fitness Test was piloted starting in 2018 and was, on paper, the most ambitious fitness-test overhaul in any branch's modern history. Six events selected to map to operationally relevant movements, age- and sex-neutral scoring intended to apply a single standard to every soldier in a given MOS, and a stated developmental philosophy — fitness as combat readiness, not as a calisthenic credential. The six events: three-repetition maximum deadlift (a strength event), standing power throw (a power event using a 10-pound medicine ball), hand-release push-up with arm extension (a refined endurance push-up), sprint-drag-carry across 50-meter shuttles loaded with a kettlebell and sled (a multi-modal anaerobic-aerobic event), leg tuck (a grip-pull-core combined event, later replaced by the plank), and the 2-mile run (kept from the APFT as the aerobic anchor). The test was longer, more equipment-dependent, and more controversial than its predecessor.

The Center for Initial Military Training (CIMT) at TRADOC, which led the ACFT development, drew on RAND's published operational fitness analyses, Army Research Institute work on physical demands of combat MOSs, and USARIEM physiology research to argue that each event mapped to a measurable warfighting task. The deadlift simulated lifting and moving casualties and ordnance. The standing power throw correlated, weakly but defensibly, with explosive power useful in throwing grenades and breaching obstacles. The hand-release push-up enforced full range of motion and removed the bounce-back that had inflated APFT push-up counts. The sprint-drag-carry was the closest the test came to a soldier-skill simulation — fire-and-movement, casualty drag, ammo can carry, all in a single timed shuttle. The leg tuck combined grip, pulling strength, and core function in one event. The 2-mile run carried over. The other ambition: gender-neutral, age-neutral scoring within MOS categories. The pilot version proposed that the same raw score earn the same points regardless of age or sex, with MOS-tiered minimums (heavy combat arms higher, support lower). A male nineteen-year-old infantryman and a female forty-year-old finance specialist would, under the gender-neutral framework, be evaluated against different cut scores tied to MOS, not to demographics.

The ACFT did not arrive alone. FM 7-22, the Army's Holistic Health and Fitness (H2F) doctrine, was published in October 2020 as the developmental program meant to prepare soldiers for the new test and for sustained combat fitness more broadly. H2F brought five domains under one roof — physical, mental, nutritional, spiritual, sleep — and was implemented through brigade-level H2F teams with embedded strength coaches, athletic trainers, dietitians, and behavioral health specialists. The doctrine itself was a leap forward, and aligned with two decades of civilian sports-science consensus on periodization and recovery. The implementation was uneven. As of 2024, only a portion of active-component brigades had received their H2F teams, and Guard and Reserve units largely did not. The doctrine the soldier was being tested against assumed support that, for most soldiers, did not yet exist.

Two streams of criticism emerged almost immediately. The first was operational-scientific: did each ACFT event actually predict the warfighting task it claimed to predict? RAND's 2020 evaluation, "An Analysis of the Army's Combat Fitness Test" (Wenger, Hodari, et al.), found that the relationship between specific events and operational performance was uneven — some events (deadlift, 2-mile run) had reasonable construct validity, others (standing power throw, sprint-drag-carry as constructed) had weaker correlations to the underlying capacities the Army wanted to measure. The second stream was political-equity. When the pilot scoring was implemented as gender-neutral, female soldier failure rates on the deadlift and standing power throw were substantially higher than male failure rates — a result that was statistically predictable given physiological differences in upper-body strength and explosive power, but operationally and politically difficult to defend. The standing power throw in particular became a flashpoint, with internal Army analyses showing failure rates as high as 60-70% for female soldiers in some age brackets at the proposed cut scores. Congress weighed in. The fiscal year 2021 National Defense Authorization Act directed an independent review of the ACFT, which RAND conducted. By 2022, gender-neutral scoring had been walked back to a hybrid model. By 2024, the Army announced that the ACFT was being replaced by the AFT — formally a rebrand and a partial retreat, in substance a recognition that the original ACFT vision was not politically deliverable at the speed the Army originally planned.

In 2024 the Army announced the Army Fitness Test (AFT), to phase in through 2025. The headline changes from the ACFT: the standing power throw was removed from general-population testing (retained only for combat arms MOSs in the initial proposal), the leg tuck was finalized as removed in favor of the plank, and scoring returned to age- and sex-scaled tables. Five events, more conventional pacing, more legible to the senior NCO who had administered tests for twenty years. This was framed officially as a refinement of the ACFT — taking what worked, removing what did not. In practice it was a partial reversion. The hand-release push-up stayed (the APFT push-up did not return). The deadlift stayed. The sprint-drag-carry stayed. The 2-mile run stayed. The standing power throw was the most visible casualty, and the gender-neutral scoring ambition was the most consequential one.

The leg tuck had been replaced with the plank earlier in the ACFT lifecycle — formally in the October 2020 directive that finalized scoring revisions, after pilot data showed female and older-soldier failure rates on the leg tuck that the Army assessed as untenable. So that change predates the AFT pivot; it is one of the partial-retreats that accumulated across the ACFT's six years. The standing power throw removal under the AFT was the cleaner break. It had drawn the harshest scientific criticism (low correlation with operational tasks), the highest female failure rates, and the most public political attention. By 2024 there was no internal constituency willing to defend it as a general-population event. Scoring reverted to age- and sex-scaled tables aligned with the broader US military tradition. The combat-arms MOSs were initially proposed to retain higher single-standard cut scores under the AFT, but the rollout has shown the gender-neutral combat-arms standard appearing in some MOS-specific guidance and not in others. The implementation has been less than perfectly clean.

The Army's public framing for the AFT changes leaned on three claims. First, that operational data from the ACFT showed certain events were not yielding the expected predictive value, justifying their removal or modification. Second, that retention concerns — particularly among female soldiers and older soldiers — were significant enough that the policy needed adjustment. Third, that the test should reward fitness without penalizing the demographic realities of the all-volunteer force. Each of those claims has truth in it. They are also, individually and collectively, the same set of reasons that Congress and the senior enlisted advisor community had been pressing publicly since 2019. The AFT was the policy outcome of a multi-year political process that produced a test the Army could defend in front of the House Armed Services Committee, not necessarily the test that the original CIMT-RAND-USARIEM analysis would have selected on pure operational grounds.

For the current soldier, the practical effect of the AFT pivot is that the test is shorter, easier to administer, and the scoring tables look more like the APFT tradition. The hand-release push-up is the unfamiliar element for soldiers who came up under the APFT. The sprint-drag-carry remains the event that exposes specific weaknesses (grip endurance, anaerobic capacity, change of direction) that the APFT never tested. The deadlift remains the event most directly tied to general strength. The 2-mile run is unchanged in form and not unchanged in significance — the AFT scoring tables and the broader narrative have made the run feel like a slightly smaller percentage of the total, but for a soldier on the bubble, the run still decides most outcomes. The fastest route to a comfortable AFT margin remains the same as it was under the APFT: build the aerobic base and the rest gets easier.

Once you strip away the doctrine documents and the press releases, the underlying question is empirical. Does an event on the AFT (or APFT, or any predecessor) predict something that matters — combat performance, injury risk, deployability, retention, career success? The research is mixed at best, and where it is clear, it is rarely the answer the institution most wants. The most thorough body of work is from USARIEM (the US Army Research Institute of Environmental Medicine), the Army Research Institute, and RAND. Civilian sports-science research on the predictive validity of standardized fitness assessments adds further context. The honest reading of all of it: some events have decent construct validity, some are essentially measuring themselves, and combat performance is multi-causal in ways no battery of five or six events can fully capture.

The 2-mile run, like the 1.5-mile run used by the other services, is a submaximal aerobic event in a duration range (9 to 22 minutes for typical service members) that has been studied extensively in civilian sports physiology. It correlates strongly with VO2 max, lactate threshold, and ground-based endurance performance more broadly. Where US military operational data has been examined — ruck-march performance, foot-mobile patrol endurance, casualty-evacuation aerobic demand — the timed mid-distance run is one of the more reliable proxies for operational endurance capacity. The caveats: the run is sensitive to body composition, to terrain familiarity, and to recent training specificity. A soldier who can run a fast 2-mile is not automatically a soldier who can ruck twenty miles, because rucking adds an axial loading and a postural-endurance component the run does not test. But across most general-population operational tasks that have an aerobic component, the timed run is a defensible measurement.

The three-repetition maximum deadlift, introduced with the ACFT and retained in the AFT, has reasonable construct validity as a measurement of full-body strength. It loads the posterior chain, requires bracing of the trunk, and produces a single number (the heaviest load successfully completed for three reps) that maps to the strength capacities studied in occupational-physiology research. USARIEM and civilian sports-science studies have shown moderate-to-strong correlations between deadlift maximums and tasks involving lifting heavy loads from the ground — casualty extraction, ordnance handling, vehicle recovery. The technical concerns: the three-rep max produces higher injury rates than the five-rep or ten-rep submaximal alternatives that the strength-and-conditioning literature generally prefers for testing. The Army accepted that tradeoff in exchange for the test producing a strength number that meant something. Whether the tradeoff was correct is a debate that continues inside the H2F community.

The standing power throw — backward overhead throw of a 10-pound medicine ball for distance — was the most criticized ACFT event on construct-validity grounds. The intent was to measure explosive power. The reality is that the throw rewards a specific technique pattern (hip-driven extension, timing, and release angle) that correlates only modestly with the underlying explosive-power capacities the Army wanted to assess. Specialized practice on the throw improves the score substantially without proportionally improving the operational capacities the score is supposed to proxy. RAND's 2020 ACFT evaluation flagged this directly. The throw distance was inflated by event-specific practice in ways that contaminated its use as a fitness measure. Its removal from general-population AFT testing reflected, in part, the empirical case against it. Where it remains in combat-arms-specific testing, the same construct-validity concerns apply.

The sprint-drag-carry combines five separate movement patterns into a single timed event. As a measurement, it has the virtue of feeling operational — the movements visibly resemble fire-and-movement, casualty drag, and ammo can carry. As an analytic instrument, it is harder to use. A soldier's total time reflects an unknown blend of sprint speed, change-of-direction efficiency, grip endurance, anaerobic capacity, and pacing strategy. Two soldiers with identical total times can have wildly different underlying capacities and wildly different training needs. USARIEM and CIMT both pursued analyses to validate the SDC against operational tasks. The results were mixed — moderate correlations with composite occupational performance, weak correlations with any single underlying capacity. The event survives in the AFT because it is the closest thing to a soldier-skill simulation the test contains, not because the analytic case is strong.

One claim that comes up repeatedly in defense of the ACFT and AFT design — that the events would predict injury risk and reduce non-deployable rates — has held up less well than the operational-performance claim. The most consistent injury predictors in military populations are prior injury history, sudden mileage ramps, and inadequate sleep — none of which a fitness test directly measures. Higher AFT/ACFT scores do correlate with lower injury rates in aggregate, but the relationship is largely mediated by training history, not by the test itself. The honest statement is that the test measures the output of training, and good training reduces injury risk. The test is not a screening tool for injury susceptibility. Using it as one — sending the slowest 2-mile runners for orthopedic screening, for example — has produced inconsistent results in the units that have tried it.

The equity question is the one where the disagreement has been most public and where the institutional position has shifted the most over the past decade. There are three roughly distinct positions, each with internal logic and external advocates, and the AFT reflects a compromise among them rather than a clear win for any one camp. The first position holds that combat-arms standards should be a single physiological cut score regardless of demographic — what an enemy combatant can do does not depend on your age or sex, so neither should your test. The second holds that the test should reward effort relative to physiological norms — a 50-year-old soldier should not be measured against a 20-year-old's biology — and that age- and sex-scaling preserves that fairness. The third holds that MOS-specific testing is the only honest answer — the desk-bound finance specialist and the combat-arms infantryman have different operational physical demands, and one standard for both is either too easy for the second or too hard for the first.

On average, adult males and adult females differ in measurable ways relevant to fitness testing. Mean upper-body strength differs by a larger margin than mean lower-body strength. Mean fat-free mass differs by a substantial margin. Mean explosive-power output differs more than mean aerobic capacity. These are statistical statements about populations, not about individuals — the strongest 5% of women out-strength the weakest 50% of men on most measures — but the population means matter when a single test cut score will be applied to an entire force. The ACFT's gender-neutral pilot scoring failed disproportionately female soldiers on the deadlift, hand-release push-up, and standing power throw specifically because those events most closely track the strength and power capacities where mean male-female differences are largest. The 2-mile run, where the mean differences are smaller, produced more comparable pass rates. None of this was a surprise to the physiologists involved in the design. It was a known tradeoff that the policy did not survive.

Age-bracketed scoring has been more politically durable than sex-neutral scoring across all branches. The Army, Navy, Marines, Air Force, Space Force, and Coast Guard all use age brackets in their primary fitness tests. The brackets are typically 5- or 10-year wide and apply scaled cut scores that reflect average declines in aerobic capacity, muscular strength, and recovery rate with age. The science behind age scaling is uncontroversial. Average VO2 max declines by roughly 0.4 to 0.5 ml/kg/min per year after age 30 in moderately active populations. Average maximum strength peaks in the late 20s to early 30s and declines slowly afterward. The age scaling on every service test acknowledges this — it is an evidence-based compensation, not a courtesy. What is more contested: whether the brackets are calibrated correctly. Some critics, including USARIEM physiologists, have argued that the age compensation in the APFT was insufficient for runners over 45 and excessive for push-ups in soldiers over 35 — the relative biological costs of aging affect different events differently and the linear bracket scaling does not reflect that. The AFT's age tables are a refinement on the APFT's but the underlying methodological question remains open.

The original ACFT design proposed three or four MOS-tier categories — heavy combat arms, combat support, combat service support, with different cut scores for each. The argument was operational: the physical demands of an 11B infantryman differ from those of a 92Y supply specialist, and applying identical cut scores either undertests the first or overtests the second. The practical implementation problem was that MOS-tiering creates equity friction inside units and across MOSs. Soldiers in adjacent MOSs would be tested against different standards. Cross-functional teams in combat would include soldiers tested at different levels. The administrative complexity of maintaining MOS-tier categories at scale — and updating them as MOSs are reorganized, as has happened repeatedly since 2010 — was significant. The AFT retains a limited version of this: combat-arms MOSs face higher cut scores on some events under the initial proposal, but the implementation has been uneven across MOSs and across components. The Marine Corps, by contrast, has tested differential standards within the PFT structure for years through the pull-up versus flexed-arm-hang distinction and through CFT events that effectively reward MOSs where the underlying capacities are routine.

Tracking who advocated for what across the ACFT-to-AFT transition is partly a matter of public record and partly internal Army politics that does not appear in publicly cited documents. What is verifiable: Congressional committees pressed for the independent RAND review, female senior officers raised concerns through both formal and informal channels, the senior enlisted advisor community signaled retention concerns repeatedly, and the broader Army strength-and-conditioning community was divided across all three of the equity positions. The result was an institutional compromise. The ACFT moved further toward gender-neutral, single-standard testing than any prior US Army fitness test in modern history. The AFT pulled back partially without returning all the way to the APFT scaling. Whether the AFT settles as the durable equilibrium for the next decade or itself gets revised within five years is open. The pattern of US military fitness test policy is that it changes, and the changes accelerate when retention pressures or political pressures get high enough.

The Army's APFT-to-ACFT-to-AFT arc is the most visible US military fitness-test reform of the past decade, but it is not the only one. Each of the other services has run a parallel evolution at a slower pace, with different events, different scoring philosophies, and different relationships to operational demand.

The Marine Corps maintains two separate physical tests: the Physical Fitness Test (PFT), which is the closer analog to the Army's general fitness assessment, and the Combat Fitness Test (CFT), which was added in 2009 to measure operational capacity more directly. Marines take both annually. The PFT historically used pull-ups for males, flexed-arm hang for females, crunches, and a 3-mile run. The crunches were replaced with a plank option in 2018 (allowed as an alternative) and the plank became the only abdominal event in 2023. The flexed-arm hang was retired in 2014 — female Marines now perform pull-ups on the PFT, with scaled scoring. The 3-mile run distance has been stable since the modern PFT format was implemented in the 1970s. MCO 6100.13A_W_CH3 is the current order. The CFT includes three events: an 880-yard movement-to-contact, an ammunition can lift (30-pound can lifted overhead repeatedly for two minutes), and a maneuver-under-fire course that combines crawling, body drag, fireman's carry, and grenade throw. The CFT scoring is age- and sex-scaled. It is the closest thing any US service has to a combat-task-simulation fitness test, and it has been broadly accepted internally as more operationally meaningful than the PFT — at the cost of being more complex and equipment-dependent to administer.

The Navy Physical Readiness Test has evolved in parallel with the Army, generally one or two years behind. OPNAVINST 6110.1 has gone through multiple revisions, with the current J-series instruction reflecting changes through the mid-2020s. The 1.5-mile run has been stable as the primary cardio event since the 1980s, with alternatives — the 2000-meter row, the 12-minute stationary bike, the elliptical, and a 500-yard swim — adopted progressively to accommodate sailors at sea, sailors with profiles, and sailors with knee or shin issues. The push-up and curl-up format mirrored the Army APFT for most of its history. The Navy added a forearm plank option as a curl-up alternative ahead of the Army, and the plank became the standard abdominal event with the J-series revision. Body composition policy — the BCA tape test and its successor protocols — has gone through its own parallel evolution, with the 2023 changes that effectively retired the tape-only failure pathway being one of the more consequential Navy fitness-policy moves of the past decade.

The Air Force's history is the most volatile. Through the 1990s, the Air Force used a cycle ergometer test that estimated VO2 max — no run, no calisthenic events, just a graded bike protocol. The cycle test was widely criticized as gameable and as a poor cohesion-builder, and was replaced in 2004 with a more conventional test: 1.5-mile run, push-up, sit-up, and a waist circumference measurement. The current DAFMAN 36-2905 governs both the Air Force and the Space Force assessment programs. Notable additions over the past five years: the 20-meter high-aerobic multi-shuttle run (HAMR) as an authorized cardio alternative, the forearm plank as a sit-up alternative, and component-specific implementations that vary between active-duty AF, Air National Guard, Air Force Reserve, and the Space Force's continuous-fitness pilot. The Space Force has been the first US service to publicly explore a non-event-based, continuous-fitness assessment model — instead of an annual test, a year-round wearables-and-coaching framework. Whether that pilot scales remains open.

The Coast Guard PFA, governed by COMDTINST M1020.8H, has tracked the Navy structurally — 1.5-mile run, push-up, abdominal event — with the addition of a 12-minute swim alternative that recognizes the operational reality of Coast Guard cutter and small-boat duty. The Coast Guard's smaller force size has allowed more flexibility in event alternatives and individual accommodations than the larger services typically extend. The Coast Guard PFA is pass/fail rather than scored on a points scale at the level of granularity the Navy and Army apply. The retention and operational consequences of a Coast Guard PFA failure track the broader DoD pattern of suspension of favorable actions and progressive remediation, but the administrative weight is somewhat lighter.

Allied and partner militaries vary substantially in what they count as fitness for combat. There is no single international standard, and the events that one country considers central often appear nowhere on another country's test. Reviewing what other militaries do is useful both as a sanity check on US choices and as a reminder that the specific events the US uses are not the only defensible ones. Honest MOS covers US allies and partner nations only. The countries below are public partners with publicly available fitness standards — what follows draws on the publicly published fitness instructions of each military. Where details are uncertain, the description is general rather than specific.

The Israeli Defense Forces (IDF) operate a multi-tiered physical standards system that differs sharply from the US model. General IDF fitness assessment includes a timed run (distance varies by role), pull-ups, push-ups, and abdominal events. Combat-eligible units — particularly Tzanchanim (paratroopers), Golani, Givati, and the special forces tiers — apply substantially higher cut scores and add unit-specific selection events: long ruck marches under load, combat swim qualifications, and unit-specific obstacle courses. The IDF's institutional acceptance of role-specific physical standards is, in practice, more thoroughly implemented than the US Army's MOS-specific tiering. Whether that reflects IDF cultural willingness to accept differential standards, smaller force size making implementation easier, or the operational tempo demanding it more visibly, is a matter of interpretation.

The British Army uses a two-test structure that parallels the US Marine PFT/CFT split. The Personal Fitness Assessment (PFA) is the more conventional event-based test, with timed run, push-ups, sit-ups, and pull-ups depending on role. The Role Fitness Test (RFT) is the operational simulation test: it includes a 4-kilometer loaded march at a prescribed pace, lifting and carrying events, and a series of soldier-skill tasks that map directly to combat demands. The RFT replaced the older Combat Fitness Test and has been progressively refined since its introduction. Pass standards are tiered by role rather than by sex, which the UK has implemented more cleanly than the US ACFT attempted. The British equity discussion around the RFT has been quieter than the US ACFT discussion, partly because of different demographic composition and partly because the UK transitioned more gradually.

The Australian Defence Force Physical Fitness Assessment includes a beep test (the 20-meter shuttle, also used for AF HAMR), push-ups, and sit-ups, with role-specific augmentations for combat and special operations roles. The beep test as a primary cardio measurement has the analytic property of producing a hard endpoint — you fail when you cannot keep up with the timed beep, which removes the pacing-error variable that affects the US run tests. Australia's PFA has been more stable than the US tests over the past decade, with adjustments mostly at the margins. The integration of women into Australian combat arms followed a different timeline than the US and produced different policy conversations about standards.

The German Bundeswehr's Basis-Fitness-Test (BFT) is a three-event battery: a sprint shuttle (sprints between markers in a timed shuttle), a flexed-arm hang for time, and a 1000-meter run. It is shorter than the US tests, easier to administer, and intentionally minimal — the Bundeswehr's institutional position has historically been that the fitness test exists to identify the unfit, not to differentiate finely among the fit. Additional role-specific assessments apply to combat and special operations units. The BFT's flexed-arm hang as a measure of upper-body endurance, rather than the push-up, is one of the more interesting cross-military comparisons. The hang removes the technique-and-rep-counting issues that plague the push-up event in every military that uses it, at the cost of being less explicitly tied to a movement pattern that appears in combat.

The French Armed Forces use a Contrôle de la Condition Physique du Militaire (CCPM) battery that includes the 12-minute Cooper test (run as far as possible in 12 minutes), pull-ups, and abdominal events. The Foreign Legion adds substantial unit-specific physical selection tests, including long ruck marches and combat swims, and the Legion's standards are publicly higher than the general French Army standards. The Cooper 12-minute distance run as a primary aerobic event is one of the older test designs in continuous use — Kenneth Cooper's original aerobic fitness research from the 1960s produced both the 12-minute distance test and the 1.5-mile time test, and most US services adopted the 1.5-mile time format while the French (and several other militaries) kept the 12-minute distance format. Either is a defensible aerobic measurement; the choice between them is a cultural-administrative preference.

There is no single NATO-wide fitness standard. NATO STANAG agreements exist for many areas of military interoperability — communications, ammunition, medical standards — but physical fitness standards have remained national prerogative. NATO partner militaries occasionally participate in joint events (like the Best Warrior or Cambrian Patrol competitions, which serve as informal comparators), but a soldier's fitness rating in one military does not translate to a defined equivalent in another. The implication for the US service member: the specific events on the AFT are a US choice, not an international consensus. Different allied militaries have made different defensible choices. There is no objective ranking of which set of events is "more accurate" — they are all compromises between operational demand, administrative feasibility, and cultural acceptance, and the US AFT sits within a wide international range of reasonable compromises.

A scientifically optimal fitness test — if such a thing could be defined — would probably look different from any test any military actually uses. It would be MOS-specific, individually calibrated, longitudinally tracked rather than once-a-year sampled, and would draw on wearables and physiological monitoring rather than on stopwatch-and-clipboard events. It would also be politically and administratively impossible to deploy at the scale of an active-duty force. What we have instead is a series of compromises between what the underlying physiology and operational demand suggest, and what an institution of this size, with this demographic composition, on this budget, with this political environment, can actually administer. Each test event is one of those compromises. Each test transition is a renegotiation of where the compromise lives.

Every event on every US service fitness test has to be administrable by a non-specialist NCO with a stopwatch, a measuring tape, and basic equipment. That constraint is not optional — it is the cost of testing every active soldier annually or semi-annually at unit level. The constraint rules out a large fraction of the events that exercise physiologists would otherwise prefer: graded VO2 max protocols, lactate threshold testing, isokinetic strength measurements, kinematic motion analysis. Within that constraint, the choice of events is a tradeoff between what can be administered fairly and what can be measured precisely. The 2-mile run scores well on administrability and reasonably well on measurement precision. The deadlift scores adequately on both. The standing power throw scored well on administrability but poorly on measurement precision against operational tasks, which is why it lost the political fight in the AFT pivot.

Fitness test events affect career outcomes — promotion points, retention, school selection. That structural reality creates incentives for soldiers to optimize against the test, not against the underlying capacity the test is supposed to proxy. When the APFT measured push-ups and sit-ups, soldiers trained push-ups and sit-ups specifically, and gym-floor sit-up technique evolved into something that maximized the score and minimized the actual abdominal contribution. The hand-release push-up was introduced specifically to defeat the bounce-back optimization that had emerged under the standard APFT push-up. This is not soldiers gaming the system in a deceptive way — it is rational response to incentive. Any test that ties career outcomes to specific events will, over time, produce training that targets the events rather than the underlying capacities. The only defenses are to choose events that are hard to game (timed distance runs and three-rep maximum strength events are relatively hard to game; technique-dependent events are easier), and to rotate events periodically to prevent specific-event optimization from fully colonizing training culture. The ACFT-to-AFT transition reset the optimization equilibrium for almost everyone. By the time soldiers had optimized for the standing power throw, the standing power throw was removed. That is partly the system working as intended.

Every hour spent administering or training for the test is an hour not spent training for actual operational tasks. The Army's testing burden alone, multiplied across the active and reserve components, is significant — preparation cycles, practice testing, the test itself, retesting for failures, body composition assessment cycles. The total fraction of unit training time absorbed by the fitness assessment program varies but is non-trivial. A test that measures operationally relevant capacities partially recovers that cost — training for the test is partially training for the job. A test that measures event-specific capacities does not. One of the defenses of the ACFT's design was that training for the deadlift and the sprint-drag-carry was closer to training for combat tasks than training for push-ups and sit-ups. The empirical case for that was modest. The principle is correct.

The pattern of US military fitness test policy is that the test changes every 10-25 years, with smaller revisions more frequently. The pressures that force change accumulate: new operational demands, new physiological research, new demographic realities, new political pressure, new senior leaders. At any given time, the test in place is the institutional equilibrium of a few years ago, slowly drifting out of alignment with the current pressures. The AFT will not be the final word. Whether the next change comes in three years or fifteen, and whether it brings the standing power throw back or removes the sprint-drag-carry or introduces a wearables-based continuous assessment along the Space Force pilot model, is impossible to predict. What is predictable is that change will come. Planning a career around the specific event set that exists right now is planning for a snapshot, not for a trajectory.

For the soldier reading this in 2026, the practical takeaway from forty-five years of US military fitness test history is straightforward: the specific events on the test will keep changing, the underlying fitness those events try to measure will not. The most durable career strategy is to train the underlying fitness and treat the current test as the format that happens to score it right now. That is not a recommendation to be cynical about the test. The test is what determines whether you are promoted, retained, and assigned. It deserves to be taken seriously. It does not deserve to dominate your training in a way that crowds out broader physical development.

The phrase that the best NCOs use about the test is "work through it." That means: pass the test by a comfortable margin, do not let it become the goal, and continue to develop the broader fitness that makes the test feel routine. The soldier who trains only to pass is one bad sleep, one minor injury, or one test-format change away from a failure. The soldier who trains for general operational fitness has margin in every direction. Concretely, that looks like: a year-round program with an aerobic base, a year-round program with general-strength development, periodic event-specific peaking in the 8-12 weeks before a scheduled test, and an attitude that treats the test as a status check rather than as the project. The Honest MOS training pages — the run-training guide, the deadlift programming, the plank progression, the sprint-drag-carry breakdown, the hand-release push-up technique — are written from this framing.

FM 7-22 (Holistic Health and Fitness) is publicly available, and most service-specific fitness instructions are. The instructions that determine how you are tested, how you are scored, what alternatives are authorized, and what your rights are in the failure cascade are not classified. Read them. The most common reason soldiers lose unnecessary points or accept unnecessary remediation is not knowing what the regulation actually says. This applies particularly to the alternative events (the row, bike, elliptical, and swim where authorized), the profile system, the age and pregnancy/postpartum accommodations, and the appeal rights in a separation packet. The institution has built more accommodations into the test than is widely known, and the burden is on the soldier to invoke them.

The history above predicts the future at a coarse level. The AFT is the equilibrium of the early-to-mid 2020s. By the early-to-mid 2030s there is a reasonable chance of another revision — driven by some combination of wearables technology maturing, continuous-assessment pilots scaling out of the Space Force, new operational demand profiles from whatever conflicts the US is responding to, and the ordinary turnover of senior policymakers. The soldier who treats the test as a snapshot and the underlying fitness as the trajectory has a long career. The soldier who treats the current event set as the permanent definition of military fitness has a brittle one. The cross-walks between APFT, ACFT, and AFT — for soldiers who have served across the transitions — make this concrete. The capacities that crossed all three tests cleanly were aerobic endurance, general strength, and consistent training. The capacities that mattered for one test and not the next were the technique-dependent event-specific optimizations.

If this page made you want to actually train, the practical follow-ups on this site are the run-training guide for the 2-mile or 1.5-mile, the AFT deadlift programming page, the plank progression, the sprint-drag-carry training breakdown, and the hand-release push-up technique guide. If it made you think about the consequences of failure, the AFT failure cascade page documents the flag, the bar, and the separation process in detail. If you are a journalist or researcher and you came here for sources, the references below cite the doctrine and the peer-reviewed and RAND-published research that this page draws on. Nothing on this page is invented. The history is real, the regulations exist, and the research is publicly available.