The Task Complexity Taxonomy
Version 1 of a classification system to help you reliably scale games from simple to complex--and how to use it.
A recurring problem among new CLA coaches is the ambiguity of what constitutes task simplification, and inversely, complexification.
I originally formulated the Scalable Live Training construct with two scaling levers, task simplification/complexification and intensity control. These are directionally correct, I think, but they do not provide enough guidance on the actual design of skill games for martial arts.
What does a more or less complex version of a game look like? What are its features?
Over the last two years minimum, I’ve been trying to cook up some sort of tool to help coaches scale games in a more systematic way—and it proved extremely difficult.
I took inspiration from Gentile’s taxonomy of motor skills, but it was hard to dial in on dimensions general to combat sports but specific enough to be useful.
After torturing over it on and off for years, I think I’ve finally devised something workable: the task complexity taxonomy (TCT).
I’m affectionately calling it the “taskonomy.”
Disclaimers
There are some inevitable misunderstandings about this taxonomy that I want to get ahead of, so I’m placing the disclaimers section before the actual tool.
Keep these things in mind before attempting to apply it.
The taskonomy presupposes aliveness.
Aliveness is the necessary foundation representative learning design in martial arts.
There are no categories in this taxonomy that are not predicated on unscripted and uncooperative interaction as a condition of participating in the task.
Dead patterns—rote drilling, flow drills, one steps, kata, overly cooperative exercises, etc.—necessarily fall outside the categorical hierarchy of the task complexity taxonomy. They are at best conditioning exercises, at worst junk volume, and are not part of skill work, as defined by scalable live training.
If you are trying to classify activities with scripted or cooperative elements in them with this taxonomy, you are doing it wrong.
The taskonomy is not a complete picture of complexity, intensity, or difficulty.
The taskonomy is a guide operating on the most observable ways to make tasks simpler or more complex chiefly via task design. It is not an exhaustive catalog of all factors influencing complexity or difficulty of a given task environment.
Constraints not accounted for that can influence complexity and difficulty of a task include, but are not limited to, affectivity, uncommon equipment constraints, additional non-sport scoring methods (e.g., adding takedowns to taekwondo clinch practice), more skilled or physically dominant training partners, individual injuries or weaknesses, tightened or loosened time constraints, and/or limitations on or expansions to defensive tactics (the taxonomy only indexes scoring methods).
Therefore, it is still incumbent upon a coach to monitor the activity on his mat and make adjustments accordingly.
The taskonomy is NOT a proper representativeness measure.
It has a relationship to representative design—and a degree of representativeness is necessarily woven into its structure—but a higher category is not always higher representativeness than a lower category.
I noted the subtle difference between RLD and task scaling in a previous article:
There’s certainly a relationship between task complexity and higher levels of representativeness, but the two are not one and the same. In fact, it’s possible to make the task much harder than the performance environment, which is technically less representative of the game itself but nevertheless could be hugely beneficial to an advanced player.
In the above cited article, I put together a different tool, the combat action fidelity index, to help more specifically measure representativeness within combat sports practices. It is linked below.
Other, more established tools also exist, like the RPAT and the representativeness dial.
The taskonomy is not a finalized tool yet.
It’s in version 1 as of the writing of this article, which I define as “sufficiently useful.” I expect feedback from real usage by other coaches, and additional thinking and research, to refine the tool over time.
The taskonomy is also not a perfect tool. Because it’s likely in its most unrefined form right now, there are aspects of it that require some instruction to know how to use correctly.
Those aspects also require some additional conceptualization provided by the individual seeking to apply the taxonomy (always a perilous thing when it comes to martial artists).
Introducing the Task Complexity Taxonomy
The TCT or “taskonomy” tracks complexity along two main dimensions:
Tactical Demands
Task-Environment Type
Tactical demands refer to the amount of (generally offensive) scoring methods available to successfully complete the task, which also necessarily stipulates a lesser or greater number of variables to manage for defense.
Task-environment type refers to the physical conditions, as dictated by the task parameters, that most commonly contribute to a more simple or a more complex task environment.
This dimension breaks down into four subdimensions following the theme of attachment and unattachment and enforcement of a meaningfully limiting practice boundary.
I unpack some theory and background on task-environment interaction in the article linked above.
The TCT Visualization Chart
The figure below is a crude visualization of the TCT, modeled after Gentile’s multidimensional classification system.
Each “C”—Class or Category—has an adjoined number to denote the specific intersection of each subdimension on the X and Y axes. C1 is these least variable and complex, and C16 is the most.
There are several elements here that need further definition and instruction. I break everything down below.
Explanation of Tactical Demand Subdimensions
Scoring methods are primarily defined by the rule set that you and your students compete in. There are other considerations here, however, which we will discuss in later sections.
The tactical demand subdimensions, concisely explained:
No scoring for one player. This is a highly asymmetrical game parameter where one attacks and the other defends. Only one had offensive scoring options. Not recommended.
High scoring limits. This is a game where both players have at least one scoring method available to them and as many as half of all available scoring options for your sport, as as defined by your own method breakdown.
Low scoring limits. These are games where most scoring options are available but at least one is constrained out. Limitations are minimal. These games are much more variable and representative and therefore of higher skill acquisition value generally.
All scoring methods available/no scoring limits. This is typified by free sparring to the full rules of the competition rule set, but it can hypothetically encapsulate enhanced rule sets even freer than the anchor modality.
Explanation of Task-Environment Types
An attachment is any entanglement or grip between one or both players. It is commonly understood in the eco community that starting in or from an attachment lowers the overall variability that has to be managed in a learning environment, at least for one of the players, but often both.
No/large boundary means there is no defined practice boundary as part of the task or the defined area is so large that it does not meaningfully add complexity to the play dynamics (i.e., a boxing ring).
Enforced boundary means the boundary is small enough that there is real and consistent danger of stepping out, and the stepping out is penalized by point deduction or match forfeit.
Attached, no or large boundary (ANLB). Where the players start attached to one another, but the total practice space has no boundaries enforced by point deduction or loss, or the boundary is so large that it does not meaningfully demand mindfulness of it.
Attached, enforced boundary (AEB). Where players start attached to one another, and there is an enforced boundary small enough to change the dynamic of play. Enforcement includes point reduction or match loss.
Unattached, no or large boundary (UNLB). Players start without connections on one another thus must mediate connections to facilitate task engagement. Thee boundary is not an important consideration for successful task completion.
Unattached, enforced boundary (UEB). Players start without connections on one another thus must mediate connections and also be mindful not to step out of bounds. In some combat sports, this might represent greater complexity than the performance environment.
Understanding the Color Coding
Below is a key to understand the color coding on the taskonomy chart. You’ll notice the color progression matches that of Renshaw’s representativeness dial.
However, since all taxonomically classified exercises are at a base level alive, the meaning of these colors for the taskonomy differs and thus does not cleanly map onto a representativeness scale.
Red: always over-constrained, should be rarely or never used in skill work
Orange: probably useful, but possibly over-constrained, especially beyond a few months of training
Green: generally useful for various purposes across skill levels and training goals
I have two comments to add:
First, I hope the colors do not serve as a point of confusion regarding the differences between Renshaw’s dial and my so-called taskonomy.
Second, these color codes are heavily based on my own speculation but informed by research, especially regarding variability and RLD. In other words, I believe they are theoretically sound, but they are not all empirically validated.
Your mileage may vary.
How to Apply Tactical Demand Subdimensions 2 and 3
The tactical demands dimension has four subdimensions, the middle two of which require some calibration before they can be applied. This lends itself to a high risk of misapplication, so I’ve written out extensive guidance on it here.
Please read this section carefully, because a sloppy attempt here will ruin your ability to accurately use the taskonomy.
Rule of Thumb: Anchor it to Your Most Dominant Competition Modality
These subdimensions are tied to available scoring methods, which are defined by your competition modality. It’s for this reason that you, as the applicator and coach, must first structure your conceptualization of scoring in a way that can be operationalized by the taskonomy.
The simplest and most easiest place to start is by anchoring the your understanding of tactical demands in the competition rule set you most often compete in or most value.
If you’re like most grapplers, and even many strikers, it’s likely you compete across different rule sets and even different sports (e.g., many Olympic TKD athletes also compete in point karate). Designing games with those elements is recommended, but your program should be structured around the most dominant rule set, which will calibrate your usage of the task complexity taxonomy.
For example, if you compete most often in IBJJF tournaments—perhaps you consider IBJJF Worlds the top tournament, because you’re a gi grappler—you should understand the scoring subdimensions of the tactical demands in terms of officially recognized scoring methods by the IBJJF.
The spread of scoring methods looks like this:
Successful takedowns
Achieving recognized mounts
Sweeps
Guard passes
You might also consider including advantages, but I’m afraid it would complicate things too much.
If you want to be an over-achiever, you can spin up multiple versions of the taskonomy calibrated by all the different sport modalities you and your students participate in.
I don’t think this is logistically prudent, but it might be a useful intellectual exercise for coaches. In fact, this might help you test different curricular focuses and decide which one is best for your mat’s needs and your goals as a coach.
A caveat
I understand pinning in terms of Greg’s nomenclature: chest-to-chest, chest-to-back, and perhaps we can through in knee-on-belly as an outlier. On the other hand, the IBJJF recognizes side mount, full mount, back mount, and knee-on-belly.
This is important:
Even if you understand the theory of the game one way, the tournament organization still understands and scores it their own way, and this can affect in-game strategy.
This is something to keep in mind for practice design.
Edge cases like boxing
Boxing’s only objective scoring methods are landed punches, knockdowns, and knockouts.
These are too few safe top-level scoring methods to productively use the taskonomy in many cases where method subtypes (e.g., “straights” vs “hooks”) might be useful to target in learning environments and when other scoring methods are too damaging for training.
Boxing is not the only case where this might be useful. Sumo has pressing out of the ring and takedown as the broadest scoring (in this case, winning) methods, but there are many different subtypes of pushout and throw that might be useful to target in some environments.
For these cases, existing conventions for classifying those methods are very useful, although you can amend those as-needed for your own purpose or according to your own philosophy.
I say this because we want to avoid working on singular techniques to prevent over-constraining, preserve representativeness, and encourage creativity and exploration.
Overall, some of these alterations won’t surface in the taskonomy, so I must reiterate that this tool can’t automate anything about your practice designs. It’s a guidepost.
Tell me what you come up with. I’m confident there are multiple productive ways to structure scoring methods compatible with the taskonomy.
A Note to Coaches on the Danger of Labels and Systematizing…
In labeling these elements, and coding them as increases in complexity, there’s a danger that coaches will artificially limit the rate of learning of their students out of fear of too much complexity.
This is an irrational fear, an unfounded anxiety. For example, a jump from AEB to UEB might be a meaningful change in complexity, but it’s realistically not a large jump, and is always doable within the same session—even for day one beginners.
The taskonomy should never be used as an excuse to hold students back. It’s a structure to help you scale complexity and simplification in a more systemic, and thus consistent, way.
To the karate folks, I have to be extra clear: if at any point you are stratifying these task categories across belt ranks, you are horribly off course and entirely out of step with the spirit of this project.
Think about it: you can compete in the whole game of taekwondo at yellow belt in a tournament, so it makes no sense to wait until a more advanced belt to conduct C15 or C16 tasks, i.e. sparring with the whole rule set.
I Covet Your Feedback
Ask questions liberally.
Take and run with it.
Tell me your findings.
I want to hear from everyone. You can meaningfully develop a tool that thousands of future coaches might benefit from. Please give me your input, because you never know how useful it might turn out.
Let’s make the constraints-led approach more accessible to all.