Today I want to explain a bottom-up framework for skill trees, starting from the most granular decision and working up to the most complex. To teach those concepts, I'll analyze the most popular format in the history of Yu-Gi-Oh! through the lens of that framework and argue why it's overestimated.

That format is the Spring 2005 Advanced Format, known colloquially as Goat Control.

The Goat Illusion

The most skillful format is the one that requires the most frequent and most difficult decision-making. There's a near-unanimous agreement on that definition, and if we accept that, then the most skillful format isn't a matter of opinion–and it isn't Goat Control.

The Spring 2005 Goat format is, and probably always will be, the most played format of all time. Partially due to its mechanics and partially due to nostalgia, that era of Yu-Gi-Oh's regarded by many as the greatest format of all time. But that's an illusion. As of 2017, Goats isn't even in the top 20 most skillful formats.


Goat Control is glorified because of its emphasis on technical play. Long games afford many opportunities to make mistakes, and in those instances, players slightly above average or higher can consciously observe how a skill gap creates marginal advantage over time, culminating in victory. While grinding marginal advantage can be an impressive and exciting feat, both to watch and to participate in, it creates the illusion in many players' minds that this is the only, or most important, form of skill.

The reality is that skill encompasses more than the ability to repeatedly make the correct selection in a binary decision. By breaking skill into smaller segments and quantifying each one, you can dissolve the illusion that the Goat Control format is outstanding in its difficulty, and thereby appreciate the format for the lessons it actually offers, without glorifying it disproportionately.

The Decision Tree

Factor 1: The frequency of single branches

To use a decision tree as an extended metaphor, a single game of Goats may present a player with more decisions in the form of two-pronged branches than many modern formats.

Can you make the correct selection in a binary decision more often than your opponent, in a game that offers many of those decisions? That's the distinguishing question that the Goat format poses. Will you play Exarion Universe or Magician of Faith? Will you set Exarion or summon it? The first skill factor I want to introduce is the frequency of the choice of what to do in one single moment of play. That often means binary decisions, but may at times include decisions of more than two options.

If you were to think of playing a game like taking a test, a largely technical format like Goats would be a lot like answering several hundred True/False questions. The sheer length of the test, as well as the fact that one right or wrong answer affects the score in a distinct and measurable way, is understandably convincing evidence for it being difficult, right? "Which of these cards do I play, and which of two, sometimes three, options with said card do I choose?" You'll have to answer that question repeatedly.

But bear in mind that a high-volume test isn't the only form of test you might encounter. Yu-Gi-Oh! offers the potential for decision trees of far more to the depth; trees that are complicated by more factors than just how many branches you encounter per game.

Factor 2: The complexity of single branches

A second consideration for skill is how many prongs there are per branch, which is analogous to the number of options you have for one given play. This factor is vastly more demanding in modern eras than in the Goat Control format.

If a Factor 2 format were a written test, it would be like a multiple choice test with fewer questions, but more answers for each of them.

Is that more difficult than the True/False test? That depends entirely on how many choices are offered per question, and how great the difference is in the number of questions. Now it's more a question of, "Which of these singles, pairs, or triplets do I lead with, and which of the three to five options do I choose?"

That sort of question isn't posed very often in Goats.

Factor 3: The sequence of multiple branches

It's also important to recognize the situations where branch clusters are interconnected, analogous to the memorization of play sequences. That factor was virtually nonexistent in past eras of competition. In fact, it's the single most useful milestone that divides the beatdown era of Yu-Gi-Oh! from the modern era.

If Factor 3 were a test, it would be a free response test. A game of this nature may be decided in fewer turns than a Factor 1-dominant game; or metaphorically, the test is completed by answering fewer questions. But that doesn't mean you're expending less effort per game. Whereas there might be only one free response question in Dragon Rulers for every ten True/False questions in Goats, that free response question might require an entire paragraph of thoughts ordered in a correct sequence.

Stream, Dragon Ruler of Droplets 

"How do I play this hand in a sequence that baits Maxx "C" and doesn't lose to Crimson Blader when I have only Stream, Dragon Ruler of Droplets and Lightning, Dragon Ruler of Drafts?" Situational shortcuts like those, along with bread-and-butter combos, can be thought of as sequences of multiple branches.

Factor 4: The selection of sequences of multiple branches

Yet another consideration is the decision you have to make about which branch clusters to even use, versus which to ignore, which is analogous to matchup knowledge. That factor was also largely nonexistent in the first era of the game.

If Factor 3 were a free response test, then Factor 4 is not only free response, but a free response test that requires the student to synthesize. For example, while a question of the Factor 3 type might ask, "Describe this medical procedure," a question of Factor 4 type will ask, "Given a patient that presents with the following symptoms, justify the procedure you would select and then describe how it's done."

Maxx "C

Extrapolating into Yu-Gi-Oh, Factor 4 may ask, "How do I alter the sequence I would do here, given that we're in a sided game; or given that I know my opponent plays Deck B instead of Deck A; or given that I know my opponent plays hand trap B but not hand trap A?" You might discard Maxx "C" for the effect of Reactan, Dragon Ruler of Pebbles to vary your opening sequence against Spellbooks, because Maxx "C" draws little value as a hand trap in that matchup. You might do a different variation of the Zoodiac Combo  that ends with Zoodiac Hammerkong instead of Zoodiac Ramram on board, if you know your opponent plays no Raigeki or Dark Hole, but you know they do play Forbidden Chalice.

Analysis Of Goats

While Goats may be one of the most Factor 1-demanding formats in history – it is, if anything, tied with a few others – a format that requires a great amount of that one type of skill can't surmount the demands of modern eras that require lots of all four. While the True/False skill factor might apply most to Goat Control, it still comes up in any modern format that can produce simplified game states: in any format, any time you reach a simplified game state with very few cards in the mix, you've effectively begun to play Goat Control. Think of a Dino Rabbit mirror without Rescue Rabbit as an example.

How does the Goat format stand up when scrutinized under the other factors of skill demand? Factor 2 comes up in Goats at various junctures, but not nearly with the same complexity and frequency as it does for other modern decks. Consider that Mermails have to make a Factor 2 decision for virtually every opening hand; with that deck, you're not making decisions with just single cards, but with pairs and triplets.

Factor 3 effectively didn't exist in Goats. Two-card sequences were sparse. Compare that to modern eras, where play sequences can have ten moves, twenty moves, or more; and on top of that you need to memorize variations on those sequences to truly play well.

Factor 4 also effectively didn't exist in Goats. Although Spring 2005 was not a one-deck format, the card pool was very limited, as were play sequences. In other words you had fewer tools AND fewer ways to use them. Consequently, the amount of adapting you needed to do based on the matchup or based on siding pales in comparison to what needs to be done in modern formats.

In fact, even one-deck formats of the modern era surpass Goats in Factor 4, because due to the immense size of the card pool, a single deck's variations could all effectively be considered different decks. I might side Level Limit - Area B against a Karakuri Geargia deck, but certainly not an Xyz Geargia deck. I might side Mystical Space Typhoon  against a Dragon Ruler build that played Vanity's Emptiness, but certainly not one without it.


Jedi Tricks

One might object that the employment of "Jedi tricks" – that is, using table presence – to open up more options in Goats also argues for its skillful dominance. But that perspective's often lax in the way it ignores certain truths. First, Jedi tricks aren't a separate skill factor as far as the quantitative measurements go. While a person who's great at True/False technical play may be weak at Jedi tricks or vice versa, both are choices made within the True/False domain of skill factors.

"Do I set monster A or monster B? Do I set monster A slowly, or do I slap it onto the table with confidence? Do I summon and attack, or do I summon, ask to see my opponent's graveyard, and then attack? Do I set two backrows simultaneously, or do I set one first, and then thoughtfully contemplate before setting the second? Should I ask my opponent a question to misdirect his attention away from what's in my grave?" All of those questions are still part of binary decision making.

Jedi tricks do employ a talent beyond recognizing the correct card to use, but don't let that fool you into thinking there are more skill categories being employed. Yes, selling your opponent into letting you come back with a clutch Morphing Jar a phenomenal feeling, but as impressive as it may look, at the end of the day, it was still the outcome of a series of selections, each with just two or three choices.

The Jedi argument also conveniently ignores that similar tricks are prevalent in all formats. There's no reason you can't ask to see your opponent's grave to get his attention away from the super cool play you set up in yours, or set face downs at a deliberate pace, or change your posture or attitude, in any modern format.


One might object that many decisions in one play during a game of Goats aren't binary, but actually a choice between three, or even four options (though I'd argue that your hand is terribly clogged with Normal Summons if you're sitting on four viable options). Still, that doesn't subvert the underlying logic of my argument; it just wrestles with semantics.

Whether I compare binary choices against choices of three to five options, or whether I compare choices of two to four options against choices of four to seven, it doesn't change that I'm comparing fewer choices to more choices.

Mastery time

Although I might be able to do some multiplication and averaging to demonstrate that many formats have more total branches per game than an average game of Goats, there's still the potential objection that Goats has "sturdier branches." Or in other words, that arriving at the correct selection in a binary choice is not the same in the modern era as it was in the past. A large body of more difficult choices, with the number of choices held equal, would result in a greater demand of time to master a given deck or format.

While I don't deny that it would take a considerable amount of time to master Goats, individual decisions are still more complex in the modern era. In a Goat Control mirror, I am evaluating your facedown backrow for about five or six different trap possibilities and two or three spell possibilities, excluding bluffs and cards I may already see in your graveyard. In modern formats, those same evaluations may be done per matchup. And recall once again that the opponent's deck after siding for Game 2 may be considered a different matchup than that same deck in Game 1, and that variations of an archetype can be considered different matchups.

The other problem with using mastery time to measure skill is that, as valid an indicator as it may be, it has no bearing on the match at present. Perhaps more time can even produce a larger results gap in Goats compared to other formats like in the chart above, but once that gap in time is closed, so is the gap in results. This leads me to my final point.

Goats Is Solved

There's one final, and tremendously significant, problem with replaying not just the Goat Control format, but past formats in general: they're solved or nearly solved. That's true both in terms of technical play, and deck building. Having less to discover means that there's yet an additional factor that the current format will always possess, that the retro format does not.

Factor 5: Navigating the forest for the correct trees

Thus far, I've introduced skill factors from the bottom up: starting with the forked branch, moving up to clusters of branches, and then further to clusters of clusters and even entire trees. Next, there's the matter of navigating the entire forest, uncovering the trees and branches to begin with.

This skill factor is all about deciding what to do in a given format. Decisions include what to do about your play, technical play; what to do about your opponent's play, mind games; what to put in your deck, deck building; and what to do based on everyone else's deck, metagaming.

This final factor is present in whatever format happens to be the current one, and becomes less present - inevitably disappearing - when you revisit past formats. The first person to ever sell a good Morphing Jar play deserves an Oscar. But eventually all good players realize the importance of knowing how to do that and suddenly it isn't the elegant finesse that it once was. The first person to attack over a wall of Sheep Tokens with an Asura Priest was probably quite the clever fellow. However at some point there are no longer any hidden "techs" to discover to counter Sheep Tokens; you now simply decide whether to play one of the known techs.

The more solved the format, the less opportunity there is for an individual to translate superior talent into a victory. Perhaps the most elegant proof of this is the game tic-tac-toe. Deep Blue may be a chess computer so sophisticated that it could defeat me in a million games out of a million at chess, but if I go head-to-head against it in tic-tac-toe we'll still draw every time. That's because the format of the game allows for no opportunity on the part of Deep Blue to translate its superior computational ability into game results.

Even that hypothetical aside, this is observed in the real world across all arenas of competition. Whether it's chess, Magic the Gathering, dance, fighting games, or even activities we typically think of as completely talent-based like sprinting and lifting, the best in the world in a former era cannot even break Top 100 in the modern era.

As time passes, new knowledge reaches diminishing returns. Players reach an effective ceiling on how much more they can improve at either technical play or deck building, and then it truly does become a game of luck, as both players take turns making 100% correct decisions.

The Value Of Goats

None of this is to say that Goats isn't skillful, or that Goats isn't very important, because I think it is. If I were to write an essay debunking the notion that hitting groundstrokes against a wall is less skillful than playing the full game of tennis, it would be to make just that point, and nothing more. It wouldn't be because I think groundstroke drills are unimportant. In fact, they're a key part of training to become good at the full game of tennis.

In that same vein, Goat Control teaches indispensable lessons that shape a player's overall arsenal of abilities. The very first YCS I topped was a sealed event and one of the easiest tournaments of my life. By virtue of being a sealed event, just about every match outcome was entirely dependent on technical play. I lost no matches to being outplayed, and only took one loss in Swiss to an opponent whose randomly assigned pool had much better cards.

Based on the numerous errors in technical play that I witnessed, I surmise it would be a safe bet to say that none of my opponents had any significant degree of experience in Goats, the ultimate teacher in technical play.

Goat Control's a wonderful format that stands over the unforgettable first era of the game. However, we mustn't laud it for being more than it is.

A bottom-up skill tree can serve many purposes. Today I've discussed how it can be used to evaluate formats, but you can also use it as a self-inventory, to evaluate what parts of your game you can work on at any given time. There are other uses as well, and if there's one general conclusion I wish to impart beyond a statement about one particular format, it's that we should recognize the utility of thinking about, debating, and playing the game in terms of frameworks.

Thank you for reading.