Generality, Specificity, and Emergence in Sport Preparation

I just got done listening to Joe Rogan's interview of John Danaher (1), and came away seriously impressed. Danaher's training as a philosopher shines through in his critical, rigorous, and systematic way of thinking about grappling.

Towards the end of the interview, Danaher and Rogan have a brief discussion about strength and conditioning for mixed martial arts. I want to address some of the things that were said, and more broadly aim to answer the question: what are we doing when we use strength and conditioning in preparation for sport?

Generality and Specificity

In discussions of strength and conditioning for sport preparation, the term specificity is often used, but rarely defined. It is generally understood to mean something like strength and conditioning practices which have significant and direct carryover to the athlete's primary sport. I will refer to this as the direct carryover (DC) account of strength and conditioning.

Specificity is an important concept for strength and conditioning, but I think this definition is at least lacking, if not totally misguided. This is because strength and conditioning is necessarily general and necessarily indirect. No matter how specific you strive to make your S&C, it will always be a (very) distant second to your sport practice in this regard. Some degree of specificity is required (football players shouldn't train to run marathons), but there is a sharp rate of diminishing returns, and attempts at increased specificity can quickly become asinine.

Instead, strength and conditioning should aim to improve general fitness characteristics for indirect carryover. I will refer to this as the general carryover (GC) account of strength and conditioning.

As an example, consider the potential benefits of using the squat to develop strength for MMA.

The squat is wholly unlike anything that happens in MMA. For the most part, fighters are in a staggered stance, they will typically produce the majority of their force from one leg at a time, and if they find themselves standing with their feet in a line, pushing into the floor as hard as possible, something has likely gone very wrong. On these metrics, squatting is extremely non-specific to MMA. So why not substitute leg strengthening exercises with a greater degree of specificity? Lunges are more specific on this account. Performing double leg shots while wearing a weighted vest or a heavy resistance band would be more specific still.

Certainly there are coaches who take this approach. It is possible that those exercises have some merit, but I think they fall short of using simple squats to develop strength, despite the apparent lack of specificity. This is because squatting does a far better job of developing general strength qualities than the other options listed above, and the improvement of general qualities is the point of strength and conditioning, at least on the GC account, which I will argue for below.

Emergence

This leads us to the question: how do we build a bridge from the general to the specific? Once we have developed general fitness characteristics, how do we maximize transfer to the athlete's primary sport?

The DC account is built on the idea that by utilizing movement patterns in S&C which are as close as possible to the movement patterns exhibited in the primary sport. This assumption causes the DC account to miss a vitally important concept in sport preparation, one which, unlike specificity itself, is rarely discussed. This concept is called emergence.

When we spout that tried and true cliché "the whole is greater than the sum of it's parts", we are referring to emergence. Emergence is what happens when a combination of phenomena creates new phenomena which are not explicitly part of the constituent elements of the original phenomena. We call these new phenomena emergent properties.

Danaher alludes to emergent properties when he describes MMA as a "transcendent sport" as opposed to an "eclectic sport". When you combine Muay Thai, Boxing, Wrestling, Jiu-Jitsu, and whatever else, new properties emerge which are not present in any single one of the original sports. The possibility of wrestling takedowns changes the boxing in MMA in ways that neither boxing or wrestling alone can account for, or even anticipate.

The DC account runs into trouble when we begin to consider emergent properties. This is because it begins by deconstructing the primary sport into a series of movement patterns, and then making the leap to assuming that those movement patterns, removed from the context of the sport, maintain their most important qualities.

As an example, consider the double leg. Above, I said that on the DC account, lunges would be more specific to the double leg than squats, and some form of resisted double leg shot would be more specific still. Looking at the movement pattern alone, this is certainly the case. But the similarity begins to fall away when we look at the double leg in the context of the sport as a whole.

When an athlete initiates a double leg takedown in sporting context, they do so from a state of constant movement. They are using feints to set-up the entry, they may change stances to throw their opponent off, and they will move from side to side to get the angle they want. All of these are crucial elements of an effective double leg, and none of them have anything to do with the movement pattern of the double leg itself. Where does this leave the DC account? Lunges may be more specific than squats, but they are still very non-specific in context. If there is a way to account for this variety of factors (without turning strength and conditioning into just more sport practice), I am unaware of it. I do not think it is possible.

If we take the DC account to it's logical conclusion, it becomes self-contradictory in application. Squats are less specific than lunges, which are less specific than resisted double leg shots – which are less specific than unresisted double leg shots on a tackling dummy, which are less specific than unresisted double leg shots on an unresisting opponent, which are less specific than unresisted double leg shots on a resisting opponent, which are less specific than just giving up on S&C and practicing the primary sport. Why bother with S&C at all, if the DC account is correct?

By way of contrast, the GC account draws a clear distinction between S&C and sport practice. We select for the general qualities most important to the sport and train those qualities. This allows sufficient specificity (when I said "football players shouldn't train to run marathons", what I meant was "the general fitness characteristics needed for football are very different from the general fitness characteristics needed to run marathons"), and that is as much specificity as is required from strength and conditioning practices, because carryover from strength and conditioning to the primary sport is an emergent property. It arises from the combination of sufficiently specific general S&C with appropriate sport practice. There is no way to create this effect with a top-down, reductionist approach. The whole is greater than the sum of it's parts.

An Ontology of Sport Specific Fitness

Alright, but what's all this got to do with John Danaher and Joe Rogan?

During their brief discussion, two points were made which I think indirectly describe the general way of thinking about S&C in mixed martial arts (and, I think, in sports in general.) I want to explore those points because I think they are suggestive of a conception of the role of S&C which could be more critical, rigorous, and systematic. In other words, I want to try to do for strength and conditioning what John Danaher has done for grappling: namely, do philosophy to it.

“One of the more interesting (schools of thought), this is Nick Curson, Marv Marinovich, that camp, they believe that you already know how to fight, and that what the camp really should be about is just radical strength and conditioning.”

— Joe Rogan

This line of thought is very common. It treats strength and conditioning as a simple skill amplifier. Take an athlete who is sufficiently skilled at their sport (whatever that means), improve their physical capacity, and the outcome is an athlete who is more able to apply their sport skill against a resisting opponent.

This account ignores the fact that there is a constant tension and interaction between the athlete's physical capacity and their sport skill. As the athlete's capacity changes, their sport skill must absorb and adapt to the changes in capacity.

We see this frequently with the snatch. When a weightlifter makes significant improvements in leg strength, their snatch technique often suffers temporarily. Their force curve changes, they are pulling the bar higher, and as a result their timing is off. The athlete must adjust their sport skill to accommodate the new capacity.

The same effect can be observed in endurance sports. If, in the off-season, a 5k runner has a great basebuilding period and comes back to sport specific training with a much higher aerobic capacity, they must practice their pacing (a hugely important sport skill in endurance events) in order to adapt to their higher level of general fitness. If they fail to do so, the capacity will not result in better competitive outcomes.

All of this is suggestive of emergent properties which result from improvements in general physical qualities combined with sport practice, and there is no reason to believe that the same phenomenon does not exist in combat sports. Pacing is a hugely important component of MMA, and if a fighter enters a camp with much improved physical capacity than they had in their last camp, there will be an adjustment period. If the camp is viewed simply as a period of improvement of physical capacity, there exists the very real and very risky possibility that the fighter enters the cage with tremendous fitness, but poor application to the primary sport.

Danaher seems to take the opposite approach.

“I’m not a big believer that you’re going to create big, significant, fight-changing physiological changes in six to eight weeks. It’s not really my experience. But I can show someone a single technique which can have a direct impact on a fight, I can show them that in five minutes. I’m not gonna claim to be a medical expert who has a deep understanding of these things, but my experience in coaching is that physiological changes take time, and you’re not gonna get it done in a fight camp.”

— John Danaher

Danaher is likely wrong about the ability to make significant physiological change in a six to eight week period (see the Tabata Study (2), for one famous example). It is especially possible to make such changes if the athlete trained intelligently leading up to the camp, in way which set them up to improve their physical capacity greatly in that short period. In fact, athletes do this all the time, and it's a fundamental part of sport preparation, usually referred to as "peaking".

But I think there is something more important going on here. When Danaher asks Rogan "how much could you raise your VO2 max in six weeks?", he reveals the underlying current of thought about strength and conditioning in sports. The thought is that improving some set of physiological metrics is equal to improvement in physical capacity. This is not dissimilar from the physiological model of health, which I've written about before (3).

The problem is that at high levels of competition, these metrics don't necessarily correlate with sport performance. Sometimes, they might even be inversely correlated with some aspects of sport skill (4).

I think all of this alludes to a problem with the way we look at the idea of fitness for sport performance. It seems that the common view is to take physiological metrics independent of the athlete's real application of their capacity, and then either prioritize building those metrics, or building skill, in the lead-up to a competition. Of course, I'm sure that members of both camps still consider the "other" part to be important – that is, I very much doubt that Danaher thinks an athlete can safely neglect their physical fitness during a fight camp, and I doubt just as much that Curson and Marinovich think fighters can spend the last six to eight weeks prior to competing doing nothing but strength and conditioning. But nonetheless, both camps seem to make an ontological distinction between the athlete's physical fitness and their sport skill.

I think that a more robust definition of fitness for sport performance encompasses the emergent properties that result from the combination of physical fitness and sport skill. To be as "fit" as possible going into a fight, an athlete needs to have found the happy medium of improving their physical capacity to the highest point they can without impairing sport specific skill. This means that the athlete cannot spend so much time and energy on their strength and conditioning that their sport practice suffers, but it also means that the athlete's sport practice must be executed in such a fashion as to allow them to make strategic and tactical adaptations to improvements in physical capacity.

By taking this view of sport specific fitness as the intersection of physical capacity and sport skill – rather than viewing physical capacity as a complementary but ontologically distinct amplifier of sport skill – we enable ourselves to write training plans which make the best use of the athlete's time and effort in preparing them for the day of competition. Improve general physical characteristics, practice your sport, and allow the emergent properties that result to define your fitness.

Bibliography

(1) Joe Rogan and John Danaher, https://www.youtube.com/watch?v=qkBXzkD6tis
(2) Tabata, Kouzaki, Ogita, Miyachi, "Effects of moderate-intensity endurance and high-intensity intermittent training on anaerobic capacity and VO2 max"
(3) Tsypkin, "Physical Autonomy"
(4) Lucia, Hoyos, Perez, Santalla, Chicharro, "Inverse relationship between VO2 max and economy/efficiency in world-class cyclists"