“Revolutions are not made by men in spectacles”  Oliver Wendell Holmes Sr.

You’ll see the term ARX or ARXFit kicked around here a bit, and listed quite often in my posted workouts.  As such, I’ve had quite a few readers ask me to give a more detailed explanation of how this equipment works, and why I like it so much.  Machines just don’t figure prominently in my grand fitness scheme, so what’s so special about the ARXFit?

Explanations of how this equipment works is rather a tall order, no doubt.  Actually, the intellectual and phisiological basis is not all that tough to explain (as I’ll attempt below).  What is tough, however, is relaying just how effective this mode of of exercise is.  The fact of the matter is that one session on this equipment with a quick follow-up white board talk could take the place of volumes of writing or blathering.  Imagine trying to explain to someone who has never sprinted, just how effective this mode of exercise is.   Watching a clip of some one sprinting — or doing battle with ARX equipment — just can’t convey the difficulty or effectiveness of each.

Heh…yeah, so that was back before I found a good flat top barber (they are hard to come by);  back, too, when these devices were marketed as CZT (Critical Zone Training) equipment.  We think the newer ARX (Adaptive Resistance eXercise) moniker much better conceptualizes this training philosophy.  And hey, a nice cameo by Anthony Johnson, of the 21 Convention, is included 🙂

Let me begin this discussion by saying that attempts to more perfectly match the body’s inherent strength curve  via various mechanisms/manipulations (again — adaptive resistance) is as old as Milo and the calf.  Eugene Sandow devised and sold a home exercise device in the early 1900’s that used rubber tubing as a resistance element.  Why rubber tubing?  Because it was relatively cheap and, if employed correctly, was actually pretty damn effective.  Chains have also been around since the early 1900’s, and were draped over the bar so as to add accommodating resistance to the deadlift and pressing movements.  These old-time physical culturalists might not have been able to plot a strength or force-velocity curve (or maybe they could have?), but they did know that these lifts were a hell of a lot more challenging at the bottom of the exercise as opposed to approaching the lockout position.  Chains were a pretty crafty answer to this inefficiency.  As the bar is elevated (and the body shifts into a more biomechanically advantageous position, (i.e., able to handle additional load) more links of chain come off the floor which increases the load, forcing the lifter the push (or pull) harder to keep the bar moving.

The strength curve graph above, and the following explanation, is an excerpt from this fine NSCA article.

The musculoskeletal system contains three types of strength curves (ascending, descending, and bell-shaped) with different force generating capabilities depending on the changing joint angles and the number of joints involved during the movement. Single-joint movements (e.g., elbow flexion and extension, knee flexion and extension, etc.), generally have bell-shaped strength curves where maximum strength occurs around the mid-phase of the lift. Multi-joint pulling movements (e.g., bent-over rows, pull-ups, etc.) have descending strength curves where maximum strength is produced at the beginning of the movement. Lastly, movements with ascending strength curves include variations of the squat, deadlift, and bench press as maximum strength and force capabilities occur near the top of the lift. Based on the mechanical properties of bands, chains, and the musculoskeletal system, variable resistance modes would be best suited for training ascending strength curve movements.

In the 1970’s elastic bands were promoted and sold as a cheap and viable alternative to the new-kid-on-the-block isokinetic machines that were beginning to gain in popularity.  I’d first tinkered with bands and chains in the mid 90s, after devouring everything that Louie Simmons had to say about the subject.  And then there was Arthur Jones, and the Nautilus cam.  A pretty damn awesome idea, actually — if your limb lengths happened to match the cam specifics; the Nautilus pullover is still one of my all-time favorite machines.  And many people don’t realize that one of Jones’ first attempts at revolutionizing exercise equipment was essentially a welded hook to which chains could be hung from a barbell.

But again, all of these innovations were attempts at more perfectly matching the body’s natural strength curve with some manner of accommodating resistance.  Louie Simmons took this idea a step further (with bands, at least), attempting to drain more out of the force-velocity relationship of particular movement patterns.  And now the hunt was on to find ways to maximize the power output of traditional closed-chain exercises (squat, deadlift, press) as well.

On the power front, it’s fairly safe to say that what limited science there is on the subject agrees with the dictates of ages-old empirical evidence: that it is the intent to move fast over the entire range of motion, and not the actual movement speed that leads to gains in speed strength. This is what training with chains and bands is all about — crafty manipulation of both the strength and the force-velocity curves.  Here is a great article from the NSCA if you want to dive into this concept a bit more.

In layman’s terms, chains and bands serve to force the lifter to push as hard as possible throughout a said movement’s entire range of motion, thus creating maximum tension in the target musculature.

According to Zatsiorsky in Science and Practice of Strength Training, 2nd ed. , “[t]he heaviest weight that is lifted through a full range of joint motion cannot be greater than the strength at the weakest point”  This obvious truth refers to sticking points and it is based on the fact that strength is joint-angle specific. Strength values will change as joint orientations change, and the weakest orientation is the most limiting one. Therefore, improving the ability to exert force at the weakest joint angle can lead to an improvement in maximal strength performance. Additionally, improving the ability to exert force at every joint angle will have a similar positive effect. When the submaximal effort method (or repetition method) is used, the second phases of lifts are somewhat neglected, and these portions of the ranges of motion don’t receive sufficient training stimulus. If this neglect occurs over a period of several months, muscular strength will not only fail to improve but will begin to drop. In the pursuit of maximum strength, this training outcome, combined with the faulty intermuscular coordination patterns that are created, isn’t favorable.

Now, let’s apply what we now now about effectively matching the strength curve with accommodating resistance to the force-velocity curve.  What we want to concentrate on here is effectively and most efficiently targeting the maximal strength area of the force-velocity curve.  How best to accomplish that?  Barbells and dumbbells can do the job for sure — but, as already discussed, there’s much left on the table.  Chains and bands are an awesome step in the right direction, but this method still isn’t perfectly calibrated to each individual’s strength curve; we’re still leaving some efficiency on the table.

So we’ve got two different ideas to think about now: the force velocity curve (power production), and the strength curve.  We’ve talked about how chains and bands can be used in an attempt to maximize efficiency of the barbell lifts in these moves, now let’s talk about where the ARXFit equipment maximizes efficiency in these movements.

Remember that the strength curve is the result of force production at various points along the range of motion, and that force production changes because of changes in the joint angle, muscle length, and the involvement of other muscles. The long and the short of it is this: every lifter knows that much more weight can be lifted at the top of a movement than at the beginning. Take the bench press for example; the load used in the exercise is limited by what one can control at the chest (let’s assume no stretch reflex is involved here, for ease of explanation).  The upper part of the range of motion, is easier — but the bar is slowing down.  Why?   Because the arm is only so long, and the bar must return to zero velocity at the arm’s terminus.  This means, again, that less force is being applied to the bar at the top of the range of motion.  Ugh, more inefficiency.  What we need is some device that would perfectly match the strength curve, allowing for maximum force development throughout the entire range of motion, with no need to slow down at the movement’s terminus.  This is exactly what the ARXFit accomplishes.

Not only the the ARXFit accomplish this task in the concentric (raising) portion of the lift, but in the eccentric (lowering) portion as well.  Eccentric movements have been shown to involve a lower number of muscle fibers than the concentric movement with the same weight.  Another way of looking at this is that one is able to handle about 40% more load on the eccentric as compared to the concentric portion of a lift.  What if I had a device that allowed for an instantaneous and perfect strength curve match not only concentric, but on the eccentric as well?  You guessed it — the ARXFit does it.  And now we’re talking some serious efficiency.

ARX Fit Omni – Unveiling at 21 Convention from Mark Alexander on Vimeo.

Check out the full version if you like, here.

I’ll just drop this anecdotal nugget in and leave it as food for fodder.  Since I’ve been associated with Efficient Exercise, and had ready access to the ARXFit equipment, I’ve gained between 7 and 10 pounds of body weight.  I wish I’d had a DEXA completed prior to working with this equipment, but I know that I’ve lost a good couple of percent bodyfat in those two years as well.  So do the math.  My eating habits haven’t changed (i.e., have remained solidly Paleo) in the last 7 years, nor has my workout style and/or frequency.  The only things that have changed are that I’m on my feet much more during the day, and I have steady exposure to the ARXFit equipment.  In other words, what max effort training I do now is much, much more pinpointed and effective than what I had been able to accomplish via the use of chains and bands.

Make plans to head out to the ATX next spring, where I’ll be demoing the ARXFit equipment and speaking about Efficient Exercise’s HIIRT style of training at the PFX13 symposium.  It’ll be one hella good time!

In health, fitness and ancestral wellness –




  1. Keith,

    Great stuff as always. I had a question for you that’s somewhat off topic. Have you ever had hamstring issues? My workouts consist of varied lifting (stick to the basics of squat, bench, deadlifts, and power cleans) and sprints. I pulled my left hamstring three weeks ago playing softball and it has been HELL recovering from it. After doing much research on the topic, I believe I had an imbalance between quads and hams. I’ve always put up big squat numbers and a lot of my leg muscle is front loaded in the quads.

    I’m still able to continue my lifting routine (gradually increasing the weight on lowerbody lifts and adding in hamstring exercises and plenty of stretching), but my favorite workout, hands down, is sprints and I’m sidelined from that action for now.

    From a person that has a similar workout style, do you have any suggestions for me?



    • Alan,
      You’re right in assessing these type of hamstring issues as being (for the most part) a result of strength imbalances between quads and hams. The hamstring acts eccentrically during the reach/stride to slow/stop the powerful forward motion of the leg which is generated primarily from concentric quadriceps action. I’m speaking in generalities here, but that’s the gist of things. Anyway, this is where the pull usually occurs — at the point of maximum eccentric loading of the ham. The fix is, as you probably have already guessed, lots of dedicated posterior chain work. If you have access to any kind of a glute/ham raise or Russian leg curl, work the hell out of it as this move really hones in on the hams like no other hip hinge motion can. And really, the problem is usually a knee extension (or extension prevention) problem anyway. Also note that I rarely do directed quad work. Any quad work that I do perform is indirect (squat cleans, for example), or indirect stimulus via deadlifts. If I do backsquat (rare), it’s performed as a box squat with the emphasis being on hamstring involvement.

      Hope that helps.

      • Keith,

        You’re the man! I understand this was a “consult your physician type of question”, but I wanted to seek advice from someone who has similar goals as myself in health and fitness.

        Thanks again,


  2. Keith, I don’t mean to try to push you into a pissing match with Joshua Trentine from Renaissance Exercise, but presumably you’re aware of his criticisms (outright dismissal would be more accurate) of the ARX machines. Is it possible to provide more of a point by point response to the types of criticisms he makes in the comments section of the Body by Science website, and also the criticisms at the Ren Ex website? Or, do you think this would simply be a waste of time?

    • Yeah, I’m not big for getting in a pissing match either, though I don’t see the harm in responding to criticism. Everyone has a right to an opinion, and I suppose I have a right to counter. Feel free to post a link to those criticisms here.

      • I sent a message via your ‘Contact’ section – I assumed the pasted links would get caught up in a filter on this comment section.

        I’ve enjoyed reading your posts over the past two years; always interesting food for thought. Thanks for sharing your insights and experience.

          • Ok, so what follows are two links containing criticisms of the ARXFit design, operation and efficacy. This is via an email sent to me by Will, and I’m paraphrasing some of his words below:

            I’ve pasted in a couple of links. The first is from the Renaissance Exercise site, where, this past June, they published a series called “Dumpers,” which consisted of a series of impressionistic criticisms of various types of new pieces of equipment, including ARX.

            1.) http://www.renaissanceexercise.com/dumpers-iv-segment-d/

            The second is from the “Body by Science” website – there are more than 200 comments but (and I think I have the right one) somewhere in the middle of that comment string, Joshua Trentine of Renaissance Exercise launches into a criticism of ARX – he does so by way of trashing/dismissing a fellow named Anthony Johnson. I have to say that Mr. Johnson is a pretty poor advocate for ARX (in my opinion, ignorance and arrogance are not an attractive combination), but, for what it’s worth, Trentine strikes me as an equally unappealing advocate for his ‘side’.

            2.) http://www.bodybyscience.net/home.html/?p=1191&cpage=3#comments

            Since it’s the election season, I’ll use an old Bill Clinton line and say that I don’t really have a dog in this fight. I don’t have access to ARX equipment (or Ren Ex equipment either), and I probably wouldn’t make use of it if I did – I’m perfectly happy with free weights and body weight exercises. I’ll soon be 55, and I’ve trained since I was 15; I still find the basic movements (Squats, Chins, Dips, etc.) perfectly satisfactory for my present needs. That said, I’ve appreciated reading various blogs and related materials available on the internet – your site being one of them – and I find new approaches and modalities a source of interest (it provides me with a break from my regular work). I’ve felt a certain kinship with your approach and I appreciate your thoughtfulness and creativity. So, reading your positive comments about the ARX equipment, combined with Trentine’s (and others at Ren Ex) complete dismissal of the very same equipment, and, to say the least, I’m curious.

            So about Anthony — I consider Anthony Johnson a good friend, and I’ve worked with him both in the Paleo f(x) symposium and the 21 Convention. Anthony is an enthusiastic customer and supporter of the ARXFit equipment and is not in any way paid to support our systems. And true, he’s not an exercise physiology geek, so his assessments in that realm my not be 100% accurate.

            I guess the bulk of these critiques revolve around safety. Let me just say that nothing is 100% safe. And if it is, it’s sure as hell not very effective. That said, I’ve trained countless people on these machines whose ability ranged from just out of physical therapy to accomplished athlete, and I’ve had an exactly ZERO accident rate. But then again, I’ve never had anyone get hurt performing power cleans (or any other exercise, for that matter) under my tutelage. I guess I should tout myself as a guru 🙂 Any way, let me just say that I’m a stickler for good form. GOOD form. The body is not robotic, and can react quite well to outside stimulus and inconsistencies. The only thing that incessant, nanny-worry over every slightest form break produces is squelched intensity. Squelched intensity results in lackluster results. The name of the game is max intensity with good (not robotic) form.

            I have no problem at all with RenEx machines; it’s quality stuff! But the smooth turn around that Josh speaks of is important only under the *very* narrowly focused SuperSlow protocol. If you ever stray away from that — which, of course, I think people should — then RenEx machines become far less effective. Or at least, their advantage has been nullified. Hell, we even employ some SuperSlow Systems machines at Efficient Exercise. To use a metaphor, it’s not that the mallet is any “better” than the chisel — what matters is selecting the right tool for the job at hand.

            I’ll be happy to answer any other criticisms/questions. I just don’t want this to turn into a pissing match because, really,there’s no need.

  3. With free weights or machines, the load on your body is determined by geometry, mechanics, and the size of the weight or weight stack, and the variation in load through the ROM doesn’t necessarily match your capabilities. But you do get immediate feedback on what your output/effort is relative to the load If your output exceeds the load, the weight moves up. If your output is below the load, the weight falls. To the extent you can control the movement, you can primarily vary the speed by modulating the difference between output and load.

    With the ARX, the speed of movement is fixed, and the output is strictly under your control, since the force generated is entirely determined by how hard you make your muscles contract. So in theory, the resistance curve is perfect. But isn’t it only perfect in practice if you can exert the same degree of effort or engagement throughout the full ROM? So how difficult is it to acheive that in practice….

    • Yes, you are correct in that the ability to generate “intensity” (for lack of a better term) is key. In fact, this is key with any movement/modality, especially those targeting instantaneous or repeated power output. Top-of-the-foodchain athletes are, of course, very adept at this. Many of my clients do have to learn to express this ability, and there is a vast spectrum of ultimate ability here. No surprise. For instance, I can perform 3 repititions of an ARX movement and be totally shot, whereas a new trainee may be able to perform 2 sets of 5 and still have some left in the tank. A real advantagento the ARX in this instance is that I can teach instantainious and sustained intensity without fear of dropping weights, bad form, or worse. On the other sidemof the spectrum, and advanced trainee can go balls-out without worrying about those same negative consequences. Again, it’s not the only tool — but it is one hellova tool to have in the workshop.

  4. In the above response you mention that you rarely do back squats. Can you explain why? I was surprised by this as I assumed it would be a staple of your programming.

    • It’s quite simply this: I get plenty of quad-specific stimulus *for my specific goals* from other exercises (deadlifts, squat cleans, sprinting, fixed-speed cycling, some RFESSing, etc.). This is in no way to dismiss the effectiveness of the squat, but rather it’s ROI relative to my goals. If I were a bodybuiler or powerlifter, things would for sure be different.


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.