AMRAP With Friends

Sourced review

Fatigue in an AMRAP: velocity, holds and movement choice

You move more slowly as an AMRAP goes on. Some of that is fatigue and some of it is a decision — in a direct comparison of the same work run three ways, the AMRAP produced the lowest movement velocity, and the researchers read that as athletes regulating their tempo to survive the duration rather than simply running out of capacity.

By · Sources last checked

The format changes how you slow down

Twelve trained athletes performed the same work under three formats on separate days: as many rounds as possible, every minute on the minute, and rounds for time. Movement velocity was measured throughout.1

FormatMovement velocityVelocity loss within a set
AMRAPLowest of the threeHigher than EMOM
EMOMHigher than AMRAPLeast of the three
Rounds for timeHighest, especially at light loadsGreatest of the three

Read the two columns together, because separately they mislead. The AMRAP was the slowest format but not the one that decayed hardest within a set — rounds for time was. The authors attribute the AMRAP's lower velocity to its prolonged duration prompting a self-regulated tempo.

That is a meaningfully different story from "the clock drives you to failure". In an AMRAP you are choosing a speed you can repeat, which isthe pacing decision showing up in the barbell rather than in the round times. Twelve participants, so hold it lightly.

Rest protects movement quality

Thirty-four subjects completed three different workout types with jump testing before, during and after: a 20-minute AMRAP of pull-ups, push-ups and air squats; a metabolic piece of skipping intervals with rest built in; and a five-minute weightlifting piece at a light load.2

Jump performance was reduced after the AMRAP and after the weightlifting piece. After the interval piece it had recovered by the end — the rest periods were long enough to restore the immediate energy stores that jumping depends on.

The practical reading is not that AMRAPs are bad. It is that continuous formats leave you measurably more fatigued at the end than interval formats doing comparable work, which matters if you have planned something demanding immediately afterwards, or if tomorrow is abenchmark retest.

Holding is harder than acting

There is a real and under-appreciated distinction between two isometric tasks that look identical. Holding means reacting to an external load — supporting a weight against gravity.Acting means initiating force against something immovable — pushing or pulling on a fixed resistance.

At matched intensity they do not last the same. In a controlled task at 60% of maximal torque, time to failure was significantly shorter for the holding action than for the acting one.3

Now the caveats, because they are large. That was twelve subjects, one joint — the elbow flexors — at a single fixed intensity, in a laboratory. Nobody has tested whether it transfers to a plank hold in minute fourteen of a twenty-minute AMRAP. I am reporting a mechanism that looks relevant, not a finding about AMRAPs.

It is worth knowing anyway, because ourmovement library is full of holds — plank, hollow, side plank, bottom squat, superman, bear crawl hover. All of them are the holding kind. If the distinction transfers, they will fail sooner under fatigue than their duration suggests.

What this suggests about picking movements

Stated as suggestions, because that is the strength of the evidence.

  • Put holds early, not late. Aplank hold that is comfortable in round two may be unholdable in round eight, and a hold you fail out of mid-rep is worse than one you never programmed.
  • Expect the technical movement to break first.Velocity falls as the workout runs,1 and the movements that need precision have the least room to lose.
  • If you want to preserve quality, build in the rest.That is what the interval format did in the jump study.2 An AMRAP is the wrong tool for a session about movement quality — use it for the thing it is good at.
  • Do not stack a hard AMRAP before a test.Jump performance was still down at the end of one.2 That is the measurable case for aneasy day before a retest.

What the research does not show

  • A specific velocity-loss figure for AMRAPs.The comparison above reports directions and rankings, not a percentage you should expect. If you have seen a precise number quoted for AMRAP velocity decay, ask which study it came from.
  • That the holding-versus-acting difference applies to bodyweight holds in a workout.It has been shown in a lab, on one joint.3 Everything past that is extrapolation, including mine.
  • Anything about long-term adaptation. These are acute studies — what happens during and immediately after one workout. Whether repeated AMRAP exposure changes fatigue resistance is a different question and these do not answer it.
  • Much about women. Twelve participants with two women;1 thirty-four subjects described as men.2 Sex differences in fatigability are a real research area and this literature is not where you will find them.

Related questions

Why do I slow down so much in a long AMRAP?

Partly fatigue and partly choice. When the same work was run as an AMRAP, an EMOM and rounds-for-time, the AMRAP produced the lowest movement velocity — which the researchers read as athletes self-regulating their tempo to sustain a long effort, not simply failing.

Is an AMRAP or an EMOM harder on the nervous system?

They fatigue differently. AMRAP produced greater velocity loss within and between sets than EMOM in a direct comparison, but rounds-for-time showed the greatest intra-set loss of the three. Built-in rest appears to protect movement quality, which is what an EMOM is for.

Should I put holds late in a long AMRAP?

Probably not, though the evidence is thinner than confident advice suggests. Lab work shows holding a load against gravity fails sooner than pushing against something immovable at matched intensity. That was one joint at a fixed torque, so treat it as a reason to be careful rather than a rule.

Programme around it

Every workout in the library lists its movements up front, so you can see where the holds and the technical work fall before you start the clock rather than discovering it in round eight.

References

  1. Barba-Ruíz M, Hermosilla-Perona F, Heredia-Elvar JR, Gómez-González N, Da Silva-Grigoletto ME, Muriarte-Solana D (2024). Muscular performance analysis in "cross" modalities: comparison between "AMRAP," "EMOM" and "RFT" configurations. Frontiers in Physiology, 15, 1358191. doi.org/10.3389/fphys.2024.1358191Twelve athletes (10 men, 2 women) performing the same work under three formats. Small, and the only direct format comparison of its kind.
  2. Maté-Muñoz JL, et al. (2017). Muscular fatigue in response to different modalities of CrossFit sessions. PLOS ONE, 12(7), e0181855. doi.org/10.1371/journal.pone.018185534 subjects, three workout types including a 20-minute AMRAP. Jump testing before, during and after.
  3. Dech S, Bittmann FN, Schaefer LV (2022). Muscle oxygenation and time to task failure of submaximal holding and pulling isometric muscle actions and influence of intermittent voluntary muscle twitches. BMC Sports Science, Medicine and Rehabilitation, 14. doi.org/10.1186/s13102-022-00447-9Twelve subjects, elbow flexors only, 60% of maximal torque. A tightly controlled lab task, not a workout.
  4. Tibana RA, de Sousa NMF, Prestes J, Voltarelli FA (2018). Lactate, heart rate and rating of perceived exertion responses to shorter and longer duration CrossFit training sessions. Journal of Functional Morphology and Kinesiology, 3(4), 60. doi.org/10.3390/jfmk3040060Nine trained men, two sessions each. Small, but it is the direct duration comparison.
  5. Kliszczewicz B, Snarr RL, Esco M (2014). Metabolic and cardiovascular response to the CrossFit workout 'Cindy': a pilot study. Journal of Sport and Human Performance, 2(2), 1–9. jhp-ojs-tamucc.tdl.org/jhp/article/view/jshp.0038.2014Pilot study, nine participants. The authors call it a pilot; treat the numbers as indicative.
  6. Rios M, Becker KM, Cardoso F, Pyne DB, Reis VM, Moreira-Gonçalves D, Fernandes RJ (2024). Assessment of cardiorespiratory and metabolic contributions in an extreme intensity CrossFit benchmark workout. Sensors, 24(2), 513. doi.org/10.3390/s2402051314 highly trained male CrossFitters. Measures the three pathways directly, in a real workout.

This page is a practitioner's synthesis of published research, not original research and not peer reviewed. Every claim above is sourced; where a study is small, a mechanism is extrapolated, or the evidence does not reach as far as the advice, it says so. It is not medical advice — see our terms.