Articles  >  General Training
CrossFit Criteria
Greg Everett
February 29 2008

[NOTE: I wrote this article in February 2008, at a time when CrossFit as an actual competitive sport was very young and still in its earliest phase of development. Since that time, more standards have been implemented in the CrossFit Games and those of you who were not around prior to the current era may not understand what all the fuss herein is about. You'll just have to believe me that this was an enormous point of contention within the community. - Greg Everett, 5-16-2014]

This whole CrossFit thing is getting big. Thanks to the interminable wisdom and foresight of one Robb Wolf—a kind and generous man as long as you don’t ask him if he’s a runner—I was personally introduced to CrossFit in its public infancy. At this time, the finer details of the theory were still developing rapidly around the very solid yet rudimentary foundation. Gone are the simple days when the affiliates list on the website was only an inch long and prospective clients had to be beaten into submission and forced to train at gunpoint. In the last few years, CrossFit has been evolving so rapidly that the community is in a continuous race to simply keep up with themselves.

Aside from the scale of the operation, CrossFit has been evolving in terms of theory and performance in ways that have provided no shortage of material for passionate debate. One argument that has remained in play and wholly with merit from the earliest days is that of technical performance and the possible compromises that arise from racing a clock.

Technique & Time

Very early in a CrossFitter’s training development, there is invariably a clear negative association between time and technique. That is, the more precise the athlete’s technique, the worse their times on a given workout. As the athlete progresses, this disparity attenuates and will generally become a positive association—the improving technique brings a new efficiency to the performance, which reduces time. However, as the athlete advances even further, precise technique will often begin to actually slow performance again in a sense by forcing the athlete to adhere to movements he or she is now capable of altering or bypassing. This final stage is the setting for many of the technique-versus-time arguments.

The importance of speed in the performance of CrossFit workouts is so well-established and entrenched that it often assumes priority on a nearly unconscious level. To draw on another relic of first generation CrossFit, Greg Glassman often repeated to the twelve seminar attendees that “Men will die for points.” The establishment of the now common CrossFit gym practice of recording workout times on publicly visible whiteboard arose from the recognition of the power this natural competitiveness had to encourage clients to push themselves harder. However, the parameter of comparison was time and so the details of actual movement performance often suffered due to this compulsion to best others’ times. With the rapid spread of CrossFit to individuals outside of affiliate gyms and the ostensibly watchful and authoritative eyes of expert CrossFit trainers, debate began to arise among athletes regarding the legitimacy of certain performances being claimed. While in the earlier days of CrossFit, there seemed to be a much greater uniformity among trainers and athletes in terms of acceptable technique for the movements employed, there is now less consistency with the influx of people from different training and education backgrounds as well as those in possession of little if any of either.

Contributing to the problem is the ambiguity of the word technique and its application to a broad category of criteria. For the sake of clarity herein, let’s establish a more precise definition: Technique refers to the manner in which an exercise, in isolation, is expected to be performed in terms of motor patterns and range of motion; in other words, starting position, ending position, and the movement that connects the two. The key to the usefulness of technique as an evaluative tool is its standardization. If expectations aren’t made official, technique remains open to interpretation by all parties involved in the argument and distracts from the core issue. Until those official definitions are published, arguments will remain ongoing indefinitely, and anything I present here is nothing more than a suggestion, although not arbitrary by any means, and presumably not too objectionable.

What are the objectives of CrossFit and how are these supported by a given approach to exercise or workout performance? The answer to this question has the power to resolve the argument by providing clear guidance regarding what constitutes appropriate and acceptable technique. The problem is that there are an essentially endless number of possible answers. The objective of CrossFit is entirely specific to its application—a wrestler, rock climber, fitness generalist, obese grandmother and a CrossFit athlete each have unique demands, expectations and goals, and these are what drive the implementation and consequently the standards of performance. The critical point here is that CrossFit, like any other training methodology or individual exercise, must be evaluated within its given context.

The Limitations of Work & Power

One of the recent changes in CrossFit theory has been the increasing emphasis on work capacity as the measure of fitness, often to the point of obscuring other important details. The detail most commonly neglected in the pursuit of power and work capacity is, of course, movement technique.

Formerly the presented definition of fitness was far more complex and inclusive. This original definition consisted of four parts that together described the state of fitness. The first part dictated that fitness required balanced development of the ten physical elements as defined by Jim Cawley of Dynamax: Strength, power, speed, endurance, stamina, flexibility, agility, balance, accuracy and coordination. The second part was the expectation that the fit individual would be capable of outperforming on average the unfit person in any conceivable physical task. The third part was the notion that fitness resided at the opposite end of sickness on a continuum of health and was essentially super-wellness. Finally, fitness was the balanced capacity of the phosphagen, glycolytic and oxidative metabolic pathways. These four parts created a partially redundant but extremely clear definition of fitness; nowhere within was the idea of work capacity explicitly mentioned. However, improved work capacity and power output were unavoidable if these standards of fitness were met. In other words, the qualities themselves have not changed—only their place in the discussion—but this has been sufficiently confusing.

The term work in physics is very specific and differs considerably from the word’s common usage, essentially synonymous with either or both the intensity or amount of an effort. Instead of the restrictive idea of work, what we should be concerned with is this relationship of effort and productive movement. If we subscribe to the precise physics definition of work and consider improving work capacity and power output as the goals of fitness, the technical performance of exercises and workouts will differ considerably from the manner in which they will be performed if instead of work capacity we consider productive effort capacity. It will differ further if in addition to productive effort capacity we evaluate the training within the framework of the various elements defined by the earlier definition of fitness, as well as with regard to the needs and goals of each individual.

The Practical Implications

With the shift of emphasis from the original definition of fitness to the improvement of work capacity has come reduced attention to the multitude of aspects that conspire to produce the fitness we’re ostensibly pursuing. For perspective, a simple question can be asked: If we could develop better work capacity and power output with exercises such as leg extensions and preacher curls, would we choose this approach over the present one? If we agree the answer is no (If you don’t, you can stop reading now), we must value the movements themselves for reasons other than their ability to help us improve work capacity alone. These reasons are myriad, and vary among athletes and training sessions, but can generally be contained within the category of functional value in the mechanical sense rather than the metabolic sense. If we were able to somehow produce improvements in work capacity and power output with curls and leg extensions, we would still lack the motor skills to express those characteristics in the most meaningful, productive ways.

We select movements based on their functional value; that is, we select the squat not because of its potential for work and power output, but because of the motor qualities that train the body to move in a manner conducive to healthy life and performance in sport. That these motor patterns are also capable of producing more work than their isolation exercise counterparts is interesting and convenient but coincidental. Work capacity and power output improvements are to a large degree specific to the movements with which the training effects are attained.

Work and power are useless terms in the absence of a medium—the medium through which these physical abilities are expressed is movement. That being the case, consideration of either without a parallel evaluation of the movements producing them tells us little.

Loosely analogous to this problem might be weight loss through nutrition in the absence of training. If we consider bodyweight and body composition our markers of fitness, diet-induced body composition changes may appear to produce fitness. Yet that improvement in body composition does nothing in terms of the motor functions of the body (aside from the reduction in stress on the structures and possible improved mobility from reduced weight and girth), and we’re left with a nice-looking body that is still incapable of anything but the most basic functions. In other words, by narrowing the focus too greatly, we may neglect integral components of fitness and may mistakenly shape our methods of achieving it.

If we agree we’re training for life and sport, we want the greatest degree of transferability possible from our training to these applications. If we consistently train a limited range of motion of an exercise, it’s precisely that range of motion within which we’ll excel. If we instead train the fullest range of motion dictated by the interplay of anatomy and functionality, all of the subset ranges of motion will be trained to the same degree along with the range excluded by a limited approach. It’s difficult to argue that ability through all potential positions isn’t preferable to ability through fewer positions.

Just as we need to prepare for the full range of possible motion, we need to prepare for the spectrum of other parameters such as time, speed, and loading. At present, CrossFit is essentially split into two basic training categories—strength and metabolic conditioning. In the strength category, technique has been prescribed clearly and is generally adhered to without resistance. However, on the metabolic side, technique suddenly becomes open to interpretation, even in consideration of the same movements seen in the strength realm.

Using the thruster as an example, I’ve witnessed less argument in regard to the finishing position overhead when considered as a strength movement rather than as a conditioning exercise. It’s recognized more easily that in order to manage heavy loads, the overhead position must be fixed properly, i.e. scapular retraction and upward rotation with a slight forward torso lean.

What occurs, then, is a gap in ability between the heaviest loading of the strength workouts and the much lighter loads of the metabolic workouts. This of course is less accurate for beginning CrossFitters who typically have lower strength bases and consequently for whom the prescribed weights of the metabolic workouts are much heavier than for the more advanced athlete.

An athlete who can do 225 lbs in a max-effort thruster will still only be using 95 lbs for the thrusters in Fran, only 42% of his maximal effort. If that athlete wants to train, as he or she should, across the entire spectrum of loading and repetition, he or she would be best served by preparing accordingly. As the thruster weight increases, the ability to cycle quickly will decrease considerably and the need to maintain structure overhead for the sake of rest, balance and the inability of the weaker muscles to support the load out front will increase. This seemingly inconsequential detail of overhead position may suddenly become the weak link in the chain and be responsible for significant reductions in performance. This unnecessary drain on performance can be avoided simply by training movements with consistent technique across the range of parameters.

As an aside, it only makes sense to claim that it’s inefficient to bring the barbell all the way back into the correct overhead position if we’re considering work alone and viewing the movement as nothing more than a method of displacing a load vertically. Efficiency is how well the athlete minimizes effort while moving from the starting position to the ending position of a movement; it has nothing to do with determining where those positions are. We can’t say it’s more efficient to stop short of the full overhead position because we’re now comparing two different movements; if we compared the counterpart segments of those two movements, the manner in which they’re performed would be identical and therefore there would be no difference in efficiency.

If we consider a series of repetitions, we may indeed perform more work in less time by stopping short of the full overhead position, but again, that’s not increased efficiency—it’s an alteration of the movement, in this case a reduction in range of motion. That a movement through a reduced range of motion and more direct path will be able to be executed in less time is no surprise, and it certainly doesn’t necessarily improve the training effect of that movement itself. If we reduce the time of performance of the exercise by the same amount of time saved through reducing the range of motion, we’ve made no progress in power output if we consider that power output to be predicated on productive effort instead of the restrictive term work. In other words, we’ve reduced the effort to a degree equal to the reduction of time and have consequently effected no net change. (This is akin to saying that the car that drove 1 mile is more efficient because it burned less gas then the car that drove 2 miles.)

A final consideration is that in most cases, including that of the thruster, resistance to such precise performance based on the fear of extending the time of execution is the result of misunderstanding the situation. The final lockout of the thruster is neither a complicated nor time-consuming movement—certainly no more so than the preceding front squat—yet appears to be for many athletes simply because of their lack of experience performing it, and the inadequate speed on the bar when nearing the top to help carry it through to the final position. With consistent practice, the movement will become remarkably quick and demand an inconsequential addition of time.

In the case of the thruster or any other overhead pressing movement, while the final effort to bring the barbell and body into a structurally sound overhead position may not contribute to the amount of work performed, this movement offers clear and genuine benefit that should in most cases override the desire to define exercise technique according to work alone in order to improve power output. Those most immediate are shoulder girdle and thoracic spine mobility, shoulder stabilization, and the balanced development of the local musculature. The question each athlete must ask in each situation is whether or not such a compromise produces a net benefit; the answer will determine how that athlete should perform the exercise.

Fitness, Athletic Training and Sport

The three basic contexts within which CrossFit performance will need to be interpreted are fitness, athletic training and sport. Fitness refers to the pursuit of the most general notion of physical preparedness; that is, no specific athletic goals exist and no particular element of fitness is being given priority other than to establish balance. Athletic training refers to the use of CrossFit-style training for the purpose of improving the performance of a given athlete in his or her sport. Sport refers to CrossFit itself acting as the sport of choice for the athlete; that is, the athlete is interested in improving his or her performance in CrossFit workouts for their own sake and has no interest in training effects in any capacity other than their influence on future CrossFit workout performance.

Performance standards for fitness are the easiest to establish. These standards will be the most demanding of thoroughness because the intention is essentially to train the body for all functional contingencies. The goal of these standards is largely to balance the training effects of the movements employed to satisfy as many of the criteria of fitness as possible. For example, we can’t alter a movement to produce greater power output if that alteration considerably reduces another training effect with which we’re concerned. In the case of thrusters, for example, this would mean that the rep must be taken to the final proper overhead position. The impact this small movement will have on the time of completion of a workout and therefore the power generated is negligible, while its exclusion may have considerable consequences in terms of strength, flexibility and coordination—arguably far more important elements than power output.

Standards for athletic training will vary somewhat based on the sport and athlete. Workout performance in these cases should be tailored according to the benefits for the sport and athlete in question with no regard to the ability to compare results among the community; athletes need to compare performances in their sports, not in their training. In other words, if the athlete and coach determine a certain standard of performance for an exercise provides the greatest benefit for the athlete, there is no reason to perform it any other way. This will generally mean occasionally relaxed standards of technique—such as stopping short of proper lockout at the completion of repetition thrusters—for greater power output for glycolytic athletes such as fighters. In most cases, however, the standards should be similar to those of the individual interested in fitness: movement performance should reflect the desire to improve as many elements of fitness as possible rather than sacrificing some for the sake of others unnecessarily.

Developing standards for CrossFit as a sport itself is a tricky endeavor. The argument can be made that because CrossFit is in essence a method of producing fitness, the sport should reflect this, and accordingly the standards for CrossFit as a sport should be identical to those for fitness. However, the pursuit of impressive times and feats encourages a relaxation of standards as a means to gain an edge on competitors. Of course, this is ultimately futile—comparison of performances must be made with regard to standards of execution. If an athlete turns in a lower time than a rival because of, for example, a wider push-up hand placement and a consequently shorter ROM, comparisons of the two performances cannot be accurate. This sort of disparity typically encourages the field to conform to the methods of the leaders, producing a gradual deterioration of technique standards that produce meaningless improvement.

Movement should be valued for reasons other than work capacity or power output regardless of the application, but individuals will need to decide what approach to training best serves their needs and interests, and accept any attendant drawbacks or limitations.
Please log in to post a comment

Pierre Auge
March 2 2008
I agree...
Rafael Flores
March 3 2008
Excellent Info!! I train with Jon Gilson and agree with his quality of movement mind set! I have increased my strength and overall fitness because of it.
Pierre Augé
March 7 2008
I also agree with that other Greg.

"Yes, both!"

Excellent technique will carryover more effectively to work capacity in the long term than sheer brute effort. But brute effort can take you quite a long way further when combined with that technique, and a willingness to allow it to deteriorate to within acceptable limits.

What is an acceptable limit? That point in which the movement is still safely performed with minimal risk to serious long term injury.

Not like this:

Brace yourself, everyone I have shown this to has injured themselves simply as the result of viewing it. I know this is one of OPT's boys and I know this is not indicative of OPT's abilities and instruction, in fact I'm quite certain it's simply a result of self instruction. It just happens to be that way. And it is the same in every one of his lifts.
Pierre Augé
March 7 2008
The most interesting part of it are the comments by the athlete attached to the video. All that maters apparently is getting the weight up...

March 7 2008
" This sort of disparity typically encourages the field to conform to the methods of the leaders, producing a gradual deterioration of technique standards that produce meaningless improvement. "

Greg..interesting that you'd choose hand placement...and producing meaningless improvement is arguable....

As a coach and athlete I have interests in all 3 fields - fitness, performance and sport...

fitness - form matters
performance - form matters most times to carry over
sport - scores matter

...and in the end, if it comes down to me winning with a 411 DL, i'll take it...and Geoff (video) is a great friend and superb coach that would not teach others the DL like this - and it is not taught this way at OPT...but when it comes to points, unless you have actually'd never understand what it means by "getting the weight up"

great article...good reflection
Greg Everett
March 7 2008
As far as sport goes, yes, the scores matter - that's all that matters. But those scores mus be the result of some kind of standardized game. I can score a lot more in soccer when I hold the ball with my hands - but that's outside the parameters of the game, so any scores I achieve with that method don't mean much if none of my competitors are using their hands.

The real concern with hand placement, for example, is less an issue of the actual placement itself than its relation to that of other competitors. Again, without a standard against which to compare competitors, the meaning of scores begins to wane. Honestly, it doesn't matter one bit to me how people perform CrossFit. My only argument is that if there is to continue to be comparison and competition, there must be standards.

With form, there's no such thing as a perfect maximal deadlift - it doesn't exist - rounding backs etc. are the nature of such movements with maximal efforts, and that's to be recognized and accepted.

You're at the top of the game - just about everything you do is going to be scrutinized to an obnoxious degree. I remember the first time there was a video posted of me doing Diane - I was crucified for my "form" even though not one person making the comments could come anywhere near my time. That said, the article was in no way directed at you, nor was it any kind of value judgment of your or anyone else's training or performance.

I haven't looked at the video posted above, and I'm not going to, because I with OPT - get the weight up. If it starts at the bottom and ends at the top, it's a legit deadlift (assuming there's no hitching). I couldn't care less about safety - this isn't a bunch of middle aged moms in Pilates - it's competitive athletes.
Pierre Auge
March 7 2008
I will refrain from sharing my opinions any further. I will only end up insulting people.

Who am I to talk?
Pierre Auge
March 7 2008
OPT you can tell Geoff I apologize for saying anything.
Robert Wolf
March 8 2008
Sometimes it is less to do with the message than the delivery. Criticism and debate are good, how we enter into those fray's determines a lot of what comes out.
March 8 2008
Pierre, never refrain from sharing, it's how we learn...we all have points to share..
Greg, agree on the standards, i'll wait for part 2..until then don't watch the knee bend in the dip of my push jerk, ha
March 11 2008
It's called leading from the front. You have crossfit big shots demonstrating WODs where they are doing "ring dips" that are nothing more than head nods and bouncing 225lbs off the ground like it's a basketball. Even the most casual observer would not be surprised that Xfit has become synonymous with incomplete reps.

Stop calling "Grace" 30 C&J for time and call it what it is. Lose the wildly kipping pullups. Do a pause at the top and bottom of each rep. Take some of the weight off the bar and do it right.
March 12 2008
Steve, are you saying that momentum is ever used in sport..i.e. ballistic movements, jerking, kipping..?
and if the weight travels the same distance (as in ring dips and dead lifts) then why is it not OK when done quickly as opposed to slowly?
last time i dropped 225 lbs on the ground from my hip height, it did not bounce that much..
Jeff Tincher
March 26 2008
Interesting article. I did not realize there was this debate going on.(between Jon and Andy) Some very solid points are brought up. I have also been blasted for form on the CF site (deservedly so) I know that it forced me to clean my technique up. The trick is cleaning up your technique and maintaining a high level of intensity.

I can't wait for part 2 of the article. Good stuff as always Greg.

James Kitcher Jr.
April 14 2008

I really appreciated this article. Generally I find that happens when someone with more experience than me in a given area cogently expresses what were vague notions swimming around in my head and builds on them. Looking forward to the next part of the article. Looking forward to the standards you're putting forth in part 2.

On a side note....
Greg, I was at the cert where Brian Contosh did Grace in 2:05. While there were a lot of "holy cow how about that!" comments on the board, there were a slew bashing him for his form when no one could come close to that time -- just what happened with you and the Diane video. Both performances are definitely in the "getting the weight up" school (but hardly unsafe) and very impressive achievements. I'd take the 2:05 and some "he rounded his back" comments any day.
Dale Saran
May 18 2008
Greg, very thoughtful and high quality writing. I have some of my own thoughts on the issue and I suspect that within this group of people (we're talking about higher-end competitors) the disagreement about whether or not a thruster "counts" is very much a matter of degree and not of kind. Let me offer this thought, as well, even in sports with multiple judges and referees and well-established parameters, we still have disagreement over what "counts". e.g. I am certain you hagve been at oly-lift meets and had someone have a lift not count and you've gone "Hunh? WTF?" and seen someone's lift count and you've said "he pressed that out!" In BJJ tournaments, "advantage" becomes critical and yet is very much a matter of judgment. Ditto for wrestling, boxing, baseball (read the rules on the strike zone vs. what they call) and on and on. I am not sure that you're prescriptions will actually alter that, although they may "tighten it up" a bit.

Nice article, nonetheless.
Greg Everett
May 18 2008
Dale -

Yes - there will always be a degree of disagreement with such activity because it's relatively difficult to evaluate, particularly live - with video everyone becomes an expert after the fact. However, I don't think the kind of standards we're talking about here really address the kind of minutiae that would be problematic on any considerable scale. But of course, people will always argue any judge's decision. The real point is that there needs to at least be a starting point from which to judge - right now, it's unclear and too much of the time people are spending arguing/discussing is unnecessary.
November 6 2008
Really enjoyed the article on a topic all the more relevant with the posting of videos and times on the Crossfit website, in particular the most recent sub 2min Fran. Thanks!
Andrew Cattermole
December 14 2013
It seems even more simple to me
Training movements have no result hence they are training movements
No matter how you standardize 30 C&J the real difficulty of any external Load event is decided via Weight Divisions and heaviest Load.. but not address these parameter you don't really get a result just a training effect usually conditioning via anaerobic threshold.
BW movements are generally judge but form and precession so doing 100 HSPU without any judgement on form or precession again you just don't get any real result

The games are 4?? days of attrition based conditioning with a bias towards weight lifting and I truly don't know what they prove...