If strength is the objective – and it should be for everybody – understanding the difference between training and exercising is fundamental to being an effective athlete and an effective coach.

So is understanding the difference between the basic barbell movements – the primary exercises – and the assistance exercises, the ones people seem to worry about the most.

  • What's the difference between an effective strength coach and a physical therapist pretending to be a strength coach?
  • What's the difference between a lifter who gets big and strong and the gym rat who's been the same skinny kid for 3 years?
  • What's the difference between a successful strength athlete and a frustrated P90Xer or CrossFitter?
  • What's the difference between a productive trainer and a babysitter (highly paid, ineffectual personal trainer)?

In each of the above examples, the former understands the difference between the primary exercises and the assistance exercises; they understand what they're used for and how they're programmed.

And in each of the above, the latter – the physical therapist, the gym rat, the fad trainer, and the highly paid ineffective personal trainer – doesn't have a clue.

The former structure their training around primary exercises and program them for long-term progress, using assistance exercises only when progress has slowed on the primary exercises.

The ill-informed think that variety is the objective and that boredom is the enemy, that the pump, sweat, fatigue, and soreness are the hallmarks and the objective of an effective workout, not realizing that these things are just the side-effects of what happened today, and aren't the indicators of progress.

The uninformed don't understand that athletes getting stronger are not "bore-able," that measurable increases in the weight used on the basic exercises are required, and that the inclusion of new "moves" in every workout – exercises that inherently lack the ability to drive basic strength – don't accomplish a thing if strength is the objective.

They lose sight of the fact that "muscle confusion" is a rather odd concept, and that doing dozens of different exercises actually prevents productive training for strength acquisition. This is the nuts and bolts difference between effective strength training and wasting time and potential.

How many times have you seen the following?

  • Magazine articles about the variety of exercises available for "doing arms," "legs," "back," "abs," or, worst of all, "The Core." (Ah yes, the fucking core – the thing that's always stronger on a man with a 500-pound squat.)
  • Infomercials for weight loss or "ab" development programs that emphasize the number of exercises available to accomplish this miraculous transformation.
  • New York Times pieces about the new "moves" (exercises) for 2013 for firming and toning your abs/inner thighs/outer thighs/arms/neck/"sidemeat"/"Buttissimo."
  • Books written by physical therapists that detail the different exercises that are absolutely necessary to strengthen your core – all of them absolutely necessary, none of them programmable for improvement for 6 months, or even 6 weeks, because they can't be.

Successful lifters, bodybuilders, and strength athletes all have one thing in common: their training is based around six or seven basic barbell movements, and the variables that are manipulated are volume, intensity, and rest, not the number of exercises.

Because there aren't very many exercises that can actually be trained.

First, "Training" and "Exercise" are different things entirely. Training is the process of directed physical stress, which results in an adaptation that satisfies a performance goal.

Since different performances require different physical abilities, and different tasks produce different types of stress, and since stress causes an adaptation (if you can recover from it), different physical abilities are therefore acquired by doing different physical tasks – and the training stresses that facilitate these different adaptations must be specific to the performance goal being trained for.

Duh, right?

Exercise ignores this fact. Exercise is what happens when you go to the gym and do exactly the same thing you did last time you went to the gym, or when you do P90X, CrossFit, or any other randomized program.

These activities are performed for one reason: the effect they produce for you today, right now. Fucking around in the gym is merely punching the ticket – you showed up, moved some stuff around, got sweaty, tired, and maybe out of breath, but you did the same thing Friday as you did Wednesday, and now that you think of it, the same thing you did Monday.

The harder programs like CrossFit and P90X are about the "burn," the sweat, the heart rate, the feeling of being "gassed" or "thrashed" or "fried" or "crushed" – they're about the perceived physical effects of the workout or immediately after the exercises are performed.

The random nature of the exercises ensures that you'll be "crushed" every time, because it guarantees that you won't adapt to the work. Either way, they're really about what happens today.

Training is not about today. It's about the process of going from where you are now to where you want to be later for the purpose of meeting a specific performance goal – usually at a specific time for more advanced trainees on a competitive schedule, but at first for the simple purpose of completing the novice phase of training, the first few months when it's easier to add weight every workout and get stronger very quickly.

During a training cycle, each individual workout is important only because of its place in the whole process. Subjective judgments about how the workout felt are important only insofar as they provide information that would make the next workout better. Much more important is this basic question: Did you meet your numbers on the squat, press, and deadlift today?

Training can't be accomplished randomly, because randomly applied stresses do not create a specific adaptation. For example, squatting 5 sets of 5 once every 3 to 5 weeks, or whenever Pukie posts it, doesn't apply a stress that can be adapted to – and built upon – before detraining to the stress occurs.

This is especially true if during the intervening period many other stresses have been applied that conflict with the adaptation necessary to get better at squatting heavier sets of 5, like doing hundreds of reps of unweighted squats, for example. Or doing hundreds of reps of anything, if the performance goal is doing 5 reps or 1 rep with a very heavy weight. Or doing nothing at all, which in itself is a "stress" that will be adapted to.

Since Training is a process designed to produce an adaptation, this process necessarily entails more than a short period of time, because the adaptations necessary for high-level performance take time to accomplish.

Strength, for example, can be improved for many years if the processes that produce it can be continued uninterrupted by injury or distraction. The closer you get to your genetic potential for any given adaptation for performance, the slower progress will be and the more critical the method by which the stress is applied will become. This is merely the principle of diminishing returns, whereby a value approaches a limit asymptotically, and is in evidence throughout the universe.

It must be said that not everybody is interested in Training. For many, Exercise is good enough. They just want to burn some calories, get a little conditioning work, and have better abs. This is fine, for those people. But the second you want more – when you decide that there will now be a goal to accomplish with all this gym time – you've graduated to Training.


Now that you're all grown up, which exercises will you use to get there? Considering our previous analysis, you'll need to choose movements that have the capacity to produce the desired adaptation over the long timeframe that profound, transformative adaptation requires. If you're a runner, you'll run; a swimmer, you'll swim.

After your initial novice phase, where the gains come quickly, you'll vary your workloads by manipulating volume and intensity over longer periods of time. If you're a smart runner or swimmer, you'll add some strength training to the program to help with force production, but running and swimming will still be the primary tools you rely on to improve at running and swimming.

Lifters, bodybuilders, and strength athletes often lose sight of the fact that they're really in the same situation. For us, heavy work on squats, deadlifts, bench presses, presses, power cleans and snatches, leg presses, and maybe barbell rows for bodybuilders will comprise the vast majority of the productive effort we'll expend throughout our training careers.

Basic barbell training will be the foundation of our progress for the entirety of our gym lives, and all the other work we do – the assistance work – must be kept in the proper perspective.

The primary lifts and the assistance exercises differ in one very important respect: the primary exercises can be trained and improved for years at a time...if the aforementioned injuries and distractions don't occur. The assistance exercises cannot.

This is because the primary exercises are systemic in nature – they inherently affect the entire body because the entire body is involved in the exercise.

Squats, presses, and deadlifts produce enough stress to induce hormonal changes and structural adaptations system-wide. These fundamental movements all have a kinetic chain that starts at the floor and ends with the bar in the hands. The arms aren't in the kinetic chain of the squat because they don't move the bar, but they stabilize the weight.

The bench press is the only primary exercise with a short kinetic chain (the legs not being a part of the kinetic aspect of the exercise, even though they're critical to benching heavy weights), but it includes enough muscle mass to meet the criteria by being able to improve throughout the lifter's training career, if care is taken to balance it with the press so that the shoulders stay healthy.

Assistance exercises use less muscle mass, a short kinetic chain, or are some variant of the parent exercise that's less efficient at allowing as much weight to be lifted. In my book, Starting Strength: Basic Barbell Training 3rd Edition, I categorize these exercises as either assistance exercises, which are variations on the parent exercise, like stiff-legged deadlifts, Romanian deadlifts, close-grip benches, and low-box squats, or ancillary exercises, which work a group of muscles in a way that the primary exercises do not, like a chin-up or a back extension.

Partial movements, like rack pulls, partial benches, and presses that use heavier weights through a shorter portion of the ROM of the parent exercise, can be improved right alongside their parent exercises, and can be used to drive progress for as long as they're trained.

But they do not constitute a replacement for the parent exercise; rather, they're used to drive continued progress on the primary lifts for more advanced trainees. My comments in this article refer to exercises that are designed to train a muscle group, not a movement pattern.

Ancillary exercises like curls, wrist curls, seated curls, preacher curls, dumbbell curls, dumbbell preacher curls, dumbbell front raises, dumbbell lateral raises, dumbbell flyes, anything done on one leg or with one hand, and 1615 or so of the 1621 exercises described in Bill Pearl's Keys to the Inner Universe don't qualify as basic exercises, because they can't be Trained for long-term progress – they can only be Exercised.

All the basic barbell exercises can be assessed with a one-rep maximum effort. This doesn't mean that they should be tested for 1RM – it just means that they can be. Novices have a new 1RM every workout since they're getting stronger every workout by doing sets of 5, so 1RM testing is pointless for novices. It's probably pointless for intermediates too, since they're still getting stronger every week.

In contrast, assistance exercises can't effectively be done for 1RM. Imagine a 1RM dumbbell fly, a 1RM wrist curl, or even a 1RM Romanian deadlift, and you can easily see my point. These exercises are designed to be used for higher reps because they're designed to work a muscle group, and a 1RM test applied to an isolated muscle group is an excellent way to get very injured.

Most machines are designed to exercise a muscle group. So, all muscle group machines are automatically optional as hell. All the others are designed to mimic a barbell exercise, and to remove the balance component of the movement. Hammer Strength makes lots of these things.

I think it's important to be able to fall down when you do a barbell exercise so that you have to make sure you don't. The balance aspects of the movement are critical to the training effect, and when this is removed you're left with a Glorified Exercise. Leg presses are a good example of a Glorified Exercise, but I'll admit that some bodybuilders have used them successfully to build massive legs.

(I wonder how much they can squat? It's always been my observation that a guy with a big squat has a big leg press, but that the converse is never true.)

Likewise, bodyweight-only exercises like sit-ups, push-ups, burpees, air squats, one-legged squats, handstand push-ups, bodyweight dips, exercises done on rings, and kettlebell exercises – any exercise whose loading variable is the number of reps or the length of the set, and which doesn't have a 1RM – can't drive a strength improvement.

This is because after about 10 reps, and depending on your bodyweight, they're not limited by your force production ability – they simply become endurance exercises. Their repetitive nature means they're inherently sub-maximal in terms of force production. They can't make you stronger unless you're very weak, and they can't continue to make you stronger for more than a couple of weeks even if you are.

Look in your notebook and ask yourself these questions: How long have I been able to add weight on a weekly or even monthly basis to my back extensions? My triceps pressdowns? My sit-ups, curls, lunges, dumbbell rows, and behind-the-neck lat pulldowns?

Can you swing a set of 5 with a 300-pound kettlebell? And if you can, what made you that strong, kettlebells or squats? Even very useful ancillary movements such as chin-ups have a limited ability to continue to strengthen for long periods of time. And none of them get you strong for as long as your squat does.

This doesn't mean that these exercises aren't useful in the context of a correctly designed training program. But it does mean that programs based on them are not Training for the same reason that programs that do not drive regular, programmed improvements in the primary lifts are not Training.


There are many successful "strength coaches" who've made careers out of writing programs that are actually just lists of exercises done in some way or another that appear to work satisfactorily.

The inescapable fact that many pro-level athletes get away with Exercising instead of Training is a function of elite-level genetics, not exercises posing as Training. Every professional or D1 athlete spending his time in the weight room adding to his collection of unilateral balancing tricks is wasting his potential for strength improvement, and strength is the most basic of athletic adaptations.

All other aspects of performance depend on strength – this is why athletes take steroids. There are no "balance steroids" and no "agility steroids" and no "endurance steroids" and no "core steroids." And it's why they should be squatting, pressing, and deadlifting instead of playing around with "Bulgarian split squats" and other such silly distractions from the real task at hand.

Even elite-level athletes who haven't actively pursued a linear increase in barbell strength still have the potential to do so, and the failure to get stronger always represents wasted potential. This is especially tragic if that failure is the result of following the advice of a "strength coach" who doesn't really know how to increase an athlete's strength. Catching up will always involve squats, presses, benches, and deadlifts, and may also involve getting another strength coach.

Look around your gym and count the guys who are just exercising. They're physically quite easy to spot: they're neither big nor strong. What are they doing? Full squats and deadlifts, or standing in front of the dumbbell rack?

Are they standing with heavy barbells in their hands, or sitting with dumbbells in their laps? Where has most of their time been spent in today's workout, on one foot or two, in a balanced symmetrical stance or staggered in some version of a split?

How many have a training log?

History tells us what works in the gym, and everything else walks down the road with a carrot in its ass. Training results in long-term progress in a specific direction, while Exercise gets you tired and sweaty, satisfying the desire for a feeling of accomplishment for people who are easily satisfied.

Training uses loaded movements that have the ability to generate long-term progress – squats, presses, bench presses, deadlifts – and Exercise can use any damn thing it wants to, because Exercise doesn't care about the long term, it just cares about today.

Basic heavy barbell movements are what the strongest, biggest men of the past century have used to get that way. Assistance exercises are merely the things these men do in the gym after they've Trained, while they're resting.

Quit wasting time and start training the basic lifts, and save the assistance exercises for later. If there's time.