General [Design Discussion] Perfect Imbalance

[ahadabans]

One thing at a time though. Let us first talk about mechanisms that allow a cube format to self-regulate in a healthy manner.

In my mind, the best way to do this is to remove broken cards and flatten the power curve. And I think the vast majority of people posting here are already doing that. In addition, it probably also makes sense to include a wide range of threats and answers so that there are multiple paths to victory and enough solutions to any one path that people can reasonably deal with them without having to resort to hate drafting and/or tutoring silver bullets (etc). Lowering curve to increase decision density early in the game may help as well. But honestly it probably just makes it harder for people to solve your cube meta versus actually make it more balanced. I still think this is a good idea though because it cuts down on auto-pilot deck win situations which really cheapen the game in my mind.

I don't know how often you all cube, but I need to be somewhat realistic about what's actually possible. There's just no way I can get enough playtime with my cube to dial it in to a level that approaches where I feel like this conversation has gone.

 

[Quirk]

Lets step back and look at the forest for a moment here. Historically, cube formats get solved. This should not be an argument, given the number of references I see towards "dragon formats", "good stuff formats", "Grim Monolith cubes" or phrases like "MODO drafting." People min/max these formats until they come up with a drafting strategy that allows them to do well, and by and large seek to repeat said strategy, until everyone is miserable/bored. I know I am not the only person that has been there, because I read about it on the forum, or get PMs about it.

Now what most people do to combat this is make fairly constant card swaps, which I think is ok in certain circumstance, but quite poor in others. As someone who finds cube an enjoyable hobby, and an opportunity to actually engage in game design on some level, it feels like a cop-out. In addition, I have drafters that don't like the constant swaps because it sometimes can feel like you are playing a never-ending beta.

Now, if my format can naturally have the tools needed to address a dominate deck type, than I don't need to solve this problem by patching in cards, which I like.

Are you familiar with the Nash equilibrium?

Let us also acknowledge that card evaluations can (and have) changed in limited formats over time. The spider spawning deck is testament to this. This means yes, we should be able to (at least for a time) create self regulating environments in a limited format of MTG.

No, because "cycles" don't happen the way they're touted to happen. Nobody, to the best of my knowledge, has made this theory work in practice - partly because it doesn't appear to be proper "theory"; I've seen no-one attempt to put it into a game-theoretic mathematical basis, and it flies in the face of what we actually know about how such games work. Wizards and League of Legends and others who're claimed to use this model are in actuality patching constantly to keep their environments unstable, unsolved.

Now, in very simple environments, where you have binary options, optimal strategies can be probabilistic: fifty percent of the time I choose A, fifty percent B. In something with as many moving parts as a Magic deck, it turns out that some cards just aren't that playable in the format, and some are all-stars.

The troll example I provided is a binary example. The troll represents that initial dominant deck, and the edict represents the controlling tech that prevents the dominant deck from being the format solution. This is the big picture.

To scale this up to something Wizards tried on a broader scale, there have been quite a number of pretty brutal colour hate cards printed, like Boil, to provide anti-dominant-deck tech. It doesn't work that way, though. Formats tend to settle down over time to have metagames diverse enough to make the narrow hate cards sufficiently bad not to be worth running, but with a relatively small number of clearly dominant decks, most of which have some level of game against each other. The rock-paper-scissors switching out of the number one deck doesn't happen cleanly; instead the proportions of decks in the meta shift until the hate for that deck is too narrow to run maindeck.

Now, I have found that smaller meta chain are quite enjoyable as well, and this is because they provide a space for card evaluations to change and grow as players grow more knowledgeable of the format. One thing at a time though. Let us first talk about mechanisms that allow a cube format to self-regulate in a healthy manner.

Evaluations most definitely do change. Often this is because they become more accurate when balanced against the format. But, because they become more accurate, continuing cycles are rare. People going back to the first strategy don't just adopt it as it was, they improve it, find ways to shore up its weaknesses. There are usually a finite number of improvements that can realistically be made; sometimes it's apparent very early on no amount of tinkering will improve a hopeless strategy.

People learned to combat trolls and lumberknots with more aggressive options? Great. But the option tree from here isn't really cyclic: perhaps the strategy of the troll deck can be improved to make it more successful, perhaps the troll deck is good enough that it can afford to lose one game an evening, perhaps it turns out that every deck can prepare against it without giving up too much and it's a bad strategy from there on out. This is a step along the way to solving the format, not evidence of it being solution-proof.

(Though, I have to thank you; I didn't really make the connection that the rock-paper-scissors school of design had disconnected quite so hard from actual games theory until having this discussion and getting to think it through in terms of things like the Nash equilibrium. I think they may actually be genuinely mathematically wrong.)

 

[Laz]

The draft process acts as a strong self-regulation tool. The initial Hexproof -> Edict -> etc chain was simplistic, and I am not convinced that it reflects a meta-game cycle as opposed to a learning cycle. As drafters become more familiar with an environment, they learn to the positioning of certain effects and therefore how important those effects are to prioritise in the drafting phase.

The drafting phase also significantly alters how one can perceive a 'meta-game' for their cube. If there is a perception that a certain type of deck is the strongest, then chances are that the increased competition for those cards will dilute it. This is especially apparent if the perception is of a strongest colour. However even without such a strongly defined meta game, the order of cards in packs and the fact that cards work in multiple types of decks significantly change how decks shape week to week. I know someone stated this early, but it seems like there are far too many moving parts in a cube draft for it to really solidify into a meta-game cycle. I think the best way to avoid the emergence of a dominant deck or strategy is to avoid narrow cards, such that competitive demand handles any specific strategy from draft to draft.

 

[Grillo_Parlante]

Evaluations most definitely do change. Often this is because they become more accurate when balanced against the format. But, because they become more accurate, continuing cycles are rare. People going back to the first strategy don't just adopt it as it was, they improve it, find ways to shore up its weaknesses. There are usually a finite number of improvements that can realistically be made; sometimes it's apparent very early on no amount of tinkering will improve a hopeless strategy.

Great, so now we can get back on topic; I'm glad that we have been in agreement all this time. This is exactly what I want to be fostering.

So, the original proposal was to present a slightly pushed card, which causes the evaluation of another card to change, which than causes the evaluation of another card to change, which creates an evolving problem.

We've already added the idea of understanding it as format oscillations, which I quite like.

 

[FlowerSunRain]

That's not a product of perfect imbalance. That very same pattern happens in the exploration of basically any game of substance, including abstract strategy games that are symmetrical and have no hidden or random factors. The cause of this isn't some designed imbalance, it is opaqueness and depth. If a game is opaque, people have to spend more time exploring it to make progress. If a game is deep, people have to explore further to understand it. If a game is opaque and deep enough,it can last a lifetime, but the journey is always progressive, never cyclical. I agree with quirk, that concept goes against everything I know about game theory.

I don't think anyone has anything against what you want to be fostering, but I think you are attributing the wrong cause to the effect you are seeking.

 

[Grillo_Parlante]

That's not a product of perfect imbalance. That very same pattern happens in the exploration of basically any game of substance, including abstract strategy games that are symmetrical and have no hidden or random factors. The cause of this isn't some designed imbalance, it is opaqueness and depth. If a game is opaque, people have to spend more time exploring it to make progress. If a game is deep, people have to explore further to understand it. If a game is opaque and deep enough,it can last a lifetime, but the journey is always progressive, never cyclical. I agree with quirk, that concept goes against everything I know about game theory.

I'm really not that concerned about the label we want to attach to it: opaque game design is fine if thats what you want.

If you want to frame it as an ouroboros rather than a cycle thats fine too.

 

[Lucre]

Where draw selection and hand manipulation comes into this is a chief concern of mine.
I'm at the point where I find it very difficult to design white value cards for modern era magic.

 

[Quirk]

So, the original proposal was to present a slightly pushed card, which causes the evaluation of another card to change, which than causes the evaluation of another card to change, which creates an evolving problem.

We've already added the idea of understanding it as format oscillations, which I quite like.

Right, here's the problem: this doesn't happen.

Adding one card does not materially change the evaluation of other cards. If I'm playing an Upheaval deck I try not to put O-Ring in it. If I'm not playing an Upheaval deck I grab O-Ring happily. Whether Upheaval was or wasn't in the format wouldn't bother me. It's one card and I may never see it.

It's exactly as wrong to play round a single card you might never see in a several person draft as it is to hate-draft cards you don't want to meet - i.e. not totally, but your very lowest priority.

We evaluate single cards in the context of the whole format, and changing one card makes little difference to that format.

Okay though, but what if you really change the format? What if you add a bunch of regenerating creatures and drop all the sweepers that can handle them? Those guys, if pushed and good enough to bother more than control, might end up in multiple decks.

Then perhaps there's a small shift; the control player learns to prioritise things that can handle those creatures higher and is slightly less keen on sweepers. There is no corresponding shift where people stop taking the regenerating creatures. The control player still plays sweepers, because they are still good against the field, and while the regenerating guys are slightly less awesome against control than they were before control learned not just to rely on sweepers, they still provide needed threat diversity. The creature decks react to the format change by picking new creatures, the control players react by building their decks differently. There it stops, because the control players getting better at the format is not a format change.

Now, there is the possibility that the format changes without adding cards because someone discovers a new archetype: suppose someone puts the regenerating creatures and sweepers in the same deck, and this strategy works. That'll ripple for a little bit; they will be taking cards away from different archetypes while drafting. There are as many sweepers and regenerating creatures in the format as before though, so effects will be subtle; perhaps aggro gets slightly more interested in instant speed removal so they can overload regeneration on the critical turn, but it's not like they never had a use for it before. They were, as ever, drafting to beat the whole format.

 

[Quirk]

Or, to put it another way: if there is a small number of players playing a selection of decks they're used to, they are at an equilibrium. Change the format substantially enough and you can disrupt this equilibrium and there will be a period of moving to a new equilibrium. Any "oscillations" come from people figuring out where that new equilibrium lies.

 

[Quirk]

I'm really not that concerned about the label we want to attach to it: opaque game design is fine if thats what you want.

"Opaque" in this instance means "hard to evaluate". It's not hard to evaluate an obviously pushed creature; it's good. It's hard to evaluate cards with drawbacks, unique effects, heavily synergistic cards. It takes time to figure out what they can actually do at their best and how that interacts with a format of other cards with partially understood potentials.

 

[Jason Waddell]

the thing about this is that it describes mtg perfectly.
By extension, we've had the expectation of the same type of constant change forced onto our cubes.

In fact, I'd argue that it describes most popular multiplayer games, the games you might describe as exceptions to this are either extreme flukes and also probably had few players stick around once they are done with their personal discovery phase. (examples: marvel vs capcom 2, street fighter 2 super turbo, smash melee, counterstrike, quake, shit even games like chess). The success of all these games relied on so many factors that were far outside of the vision of the creators. I almost put starcraft broodwar in that list, but it had constant updates in the form of new 3rd party maps that constantly shifted the optimal strategies.

Note that all of these games (even chess) have been criticized for their longstanding imbalances!

And one might bring up that for example, in Super Turbo (a 20 year old game), Dhalsim beats Boxer, and Boxer beats Claw, and Claw beats Dhalsim. Seems like a simple enough RPS relationship, a metagame balancing thing. But the thing fighting games have is very simple: you can switch characters between games in a match, or between matches. You can go from being the control deck to the aggro deck. Imagine if in Starcraft, you rushed during game 1 in the first round of a tournament, and now you're stuck with your 5Pool Zerg Rush for the rest of the tournament, and that's it, no switching. That'd be ridiculous, it wouldn't make sense, unless such a strategy isn't just a simple RPS choice. But that's what trying to build RPS matchups into magic is equivalent to. and that's why they're undesirable.

Someone could bring up that fighting games have notably many possibilities to force through bad matchups with differences in player skill. But for every one of those, there's a thousand of these (i just found the last two by typing "dhalsim yun" and "3rd strike chun" into youtube without having any particular vid in mind!)
in a sense, that's sort of false hope. but in a lot of ways, mtg doesn't even give you that. and honestly that's not really a good thing, like i said, it's a side effect of actual good things. but at least in these other games, you can change your strategy mid-tournament. with regard to draft you're fighting over resources from start to end, and a sideboard doesnt really actually fix a bad matchup.

I'd also argue that MTG simply isn't an inherently interesting enough game to keep people entertained forever without constant updates and shifting. Or rather, the updates and shifting balance are part of what is inherently interesting about MTG. "Swing of the pendulum" to go for the maro-speak.

also "balance" means like 30 million different things to different people because I suspect there won't be any consensus on these questions:
if a matchup has a 50/50 win rate, but every time you draw card X you win that game for free, is card X balanced?
is a game less balanced if you have 16 equally good options and you add a 17th less good option?
if a game has 6 equal "character" choices, but 4 of them force you to make very similar strategic choices (and deviating from that is playing suboptimal), is this game more balanced than a game with 6 character choices where the characters are less equal but there are several optimal strategies?
in less confusing terms, if there's 5 exact copies of a character in a game, they are balanced, but there's a 6th character that is different that is worse, is that more balanced than a game with 6 very different characters?
is it more useful to talk about the full range of balance or ignore certain outliers?
Should every option even be viable? The answer is probably no, because then you could just pick cards at complete random during the draft and end up with a viable deck, right?
if there's a magic card that wins every game it's played in unless the opponent knows a secret password that counters that spell for no mana cost, is that balanced?
if a deck somehow existed where you could literally make no choices, but you would still win 40% of your games, is it balanced? What if there was no way to use skill to increase this rate about 40%?
is goblin charbelcher killing you on turn 1 balanced?
if everyone in legacy can play wasteland decks if they wanted to, yet the card "warps the format" around it, is it balanced?

Here's a question I've always had in mind. Do you think it would be possible to build a static set of, say, 8 "duel decks" that are approximately balanced with interesting and compelling gameplay? It would have to be a set of deep and skill-intensive decks, that ideally take years to master, and need to be piloted significantly differently based on the matchup (in the way that Fox v Falco is super different than Fox v Jigglypuff).

To my knowledge nobody has ever attempted to create such a setting.

 

[sigh]

So, the original proposal was to present a slightly pushed card, which causes the evaluation of another card to change, which than causes the evaluation of another card to change, which creates an evolving problem.

All of the game theory discussion aside about what shape our changes take, and why they take place, and what the changes are called, I think that we can figure out a way to actually have this occur. Maybe not in the exact way the video or OP was hinting at, but in some way. Cubes are highly complex with a lot of dimensions to them. Personally I think this is less about one individual card (which grillo has pointed out was mostly for discussion sake), and one dimension of cards. An example of this could be "red burn spells".

I was thinking about previous discussions on this forum, and one potentially interesting example of this is the instant v. sorcery burn debate. My cube right now has instant speed burn, and also has several double striking creatures/support. I would imagine switching all my burn out for Hammers and Fancy Flames would have quite an interesting effect on how decks play out over the next few drafts. Again, I won't assign any particular geometry to this, but I've altered one dimension of the design space and it should make some cool things happen to how decks go together.

Short: I think to make this happen a change has to occur on a "dimension" of play.

The initial Hexproof -> Edict -> etc chain was simplistic, and I am not convinced that it reflects a meta-game cycle as opposed to a learning cycle.

We are also reading way to much into the troll/edict illustrative example, which is (as stated) a simplified illustration.

I believe Grillo pointed this out himself about the simplistic nature, but at the same time I don't think you can isolate "learning" from "meta-game" especially in a complex drafting environment. And I'm not so sure it's super important to determine which it is. If there is some sort of evolving happening amongst the players, then I think the goal He's looking for is complete.

 

[Quirk]

All of the game theory discussion aside about what shape our changes take, and why they take place, and what the changes are called, I think that we can figure out a way to actually have this occur.

I think it depends what "this" is. If it's some kind of eternally cycling metagame, no, this is very probably mathematically impossible. If we make changes and see ripples out across the way other cards are valued in the changed format, and a period of players oscillating between over-valuing and under-valuing new and old cards: that is absolutely a thing and to be expected, but eventually the new environment hits equilibrium in the same way the last did.

The thing people are taking issue with is the notion that off-balancing things creates a perpetually dynamic metagame. It doesn't. The premise of the video is false and its ideas of "balance" are confused. Changing things will shift your metagame, but it will settle down again over time. It'll probably still evolve a little even in a "stagnant" state as people become better players, but that's a slow process, frequently invisibly so.

And, as an aside, I think making combat tricks better in your environment by dialing back the instant speed removal definitely has the potential to change how games play out and how decks are built, and I'd be interested to know how the experiment goes.

 

[FlowerSunRain]

All of the game theory discussion aside about what shape our changes take, and why they take place, and what the changes are called, I think that we can figure out a way to actually have this occur.

The best way to go about this is have the having the implications of your decisions being non-obvious and far reaching as possible. And the best way to do that is to create a game where you yourself as the designer have no fucking clue how to play it. Follow strong design principals and don't worry about what the end product looks like. If you can actually accurately predict what the end state of your game looks like, it has no [strategic] depth.

 

[Quirk]

Here's a question I've always had in mind. Do you think it would be possible to build a static set of, say, 8 "duel decks" that are approximately balanced with interesting and compelling gameplay? It would have to be a set of deep and skill-intensive decks, that ideally take years to master, and need to be piloted significantly differently based on the matchup (in the way that Fox v Falco is super different than Fox v Jigglypuff).

To my knowledge nobody has ever attempted to create such a setting.

Have seen it done in Magic with 2 decks numerous times but never 8. There are other games which have attempted this - e.g. Yomi, Summoner Wars - but the more decks you add the more matchups there are to balance, and when juggling 28 possible combinations it's hard not to end up with some of them being one-sided. It would be a fascinating project, but massively time-consuming: change one card in one deck, play 7 matchups a number of times to figure out the impact... rough balance would probably not be prohibitively hard to achieve, fine balance nearly impossible.

 

[sigh]

I think it depends what "this" is. If it's some kind of eternally cycling metagame, no, this is very probably mathematically impossible. If we make changes and see ripples out across the way other cards are valued in the changed format, and a period of players oscillating between over-valuing and under-valuing new and old cards: that is absolutely a thing and to be expected, but eventually though the new environment hits equilibrium in the same way the last did.

Rasmus and I made a pair of posts about exactly this. Maybe this is one major takeaway from this discussion so far: there is no way to have an imbalance create an eternal little mind game of sorts. I think most of us actually do agree on this. I'll make one semi-parallel. Wizards can't just make one new change to Type 2 (new set) and be done, they have to release stuff every now and again to keep the "freshness" in drafting and standard.
To bring it back to our environments, we can change one dimension (as I've argued is correct) and have things damp out. You'd then you'd have to change another to jump-start the transient again. Note that this transient can take a long time, depending on group dynamics and characteristics, so it's not like you have to change chunks of cards every week.

It's perfectly acceptable to take issue with the video, of course, but I think we are at a point where we can build on our own ideas without tethering ourselves to an initial example.

Unfortunately my cube is still getting finishing touches to an overhaul, my playgroup is fairly new and casual, and we don't meet as often as I'd find ideal. My experiment won't even be able to start yet until the storm has settled on the current changes .

 

[sigh]

The best way to go about this is have the having the implications of your decisions being non-obvious and far reaching as possible.

I like this, I think Ahadabans mentioned this in his "be as diverse as possible" post above. If anything, it makes the discovery phase intricate and interesting, and gives players and you more dimensions to work with.

 

[Quirk]

Sigh: completely in agreement.

The best way to go about this is have the having the implications of your decisions being non-obvious and far reaching as possible.

Now, to try and progress with this: hard-to-evaluate, complex cards make formats trickier to solve. Razormane Masticore has quite a tasty upside, a 5/5 first striker which bolts a creature every turn - but quite a cost, too - it's hard to evaluate, and so in many environments it doesn't get picked very much. Enduring Renewal must surely be completely broken in some board states, but how do you set them up? A format in which cards like these are stars is a format that's probably not quick to solve...

...but Pascal Maynard's recent article on draft fundamentals describes how even in synergy heavy environments, starting with flexible powerful cards is the way to go, and it's established everyone wants to first pick Flametongue Kavu. When do people move in on the more complicated cards? What motivates drafters to try and break Enduring Renewal? I see people drafting my cube frequently shying away not just from cards like Masticore, but picking up a Curse of Shallow Graves in the draft and then not putting it into reasonably aggressive decks - where my experience is that Curse can quickly tip into something of a beating. Maybe drafters only start to experiment when they're getting bored of the more mainstream options?

 

[ahadabans]

Here's a question I've always had in mind. Do you think it would be possible to build a static set of, say, 8 "duel decks" that are approximately balanced with interesting and compelling gameplay? It would have to be a set of deep and skill-intensive decks, that ideally take years to master, and need to be piloted significantly differently based on the matchup (in the way that Fox v Falco is super different than Fox v Jigglypuff).

To my knowledge nobody has ever attempted to create such a setting.

I love this question because it's exactly what my ideal meta would be. As Quirk said, can you do it and have rough balance? Definitely. Perfect balance? Not unless you spent hundreds of hours on the project and created custom cards that did exactly what you needed to shore up the balance problems you would inevitably find. And even then, a game with as much RNG in it as Magic is exceedingly hard to truly balance. How much of a decks success comes down to just timing (having or getting cards you need when you need them)? And how many games do you need to play in order to tell the difference between getting lucky and having a solid deck that's actually good?

More importantly though (and where I don't think a lot of people would be happy), you'd have to largely homogenize these decks. By that I mean they'd all have to work in roughly the same way and also have multiple win conditions. The easiest analogy I can make here would be something that was capable of putting early pressure but that also had some late game control elements (say a way to repeatedly draw cards and/or sweep the board). And which path you chose would be based on the evolving board state and the cards in your hand. In short, the decks could not be "gold fish" style decks where you just do your thing and ignore what the other guys is doing.

The key in my mind would be making highly interactive decks with minimal degeneracy (but enough synergy to where sequencing mattered but wasn't overwhelming) so that at all stages of the game you had open to you multiple meaningful play decisions and counter plays and it would never be obvious what the right move was.

To attempt to give a specific example... Say you had a two power one drop in your hand but not a lot of other support for an aggressive push. You also have a scry land. What is the right play? Play the creature and hope for more pressure or scry to setup later turns and work on a later game strategy? Ideally, the choice would be influenced by what your opponent did and not just the cards in your hand. If you were on the draw and they drop their own two power one drop, maybe you scry for answers with the assumption that you will be facing early pressure? Or maybe you race because you suspect that it to your advantage?

The problem though is that isn't how people generally build decks. Most people I think look for a specific theme and then look to maximize that strategy. People are looking for degeneracy, but degeneracy takes away meaning from in-game decisions. Soul Warden into ajani's pridemate into Path of Bravery nullifies a great number of viable strategies your opponent may have had before the game started. It also puts them on an immediate clock to find a way to undo your board state otherwise they can have an IQ 160 and still have no way to win this game of Magic. The problem is the most effecitive way to win at Magic is to find ways to break the rules. Playing fair is not generally an effective strategy, at least not at the power level most cubes operate.

These perfect sequences are not common though, especially in a singleton cube. Running multiples of cards makes them more likely, but still nowhere near to what is possible in constructed. And that is why I think most people look at their cubes and don't see balance issues or shallow game play or problems with too much power creep (even if all those things are there below the surface). The game of Magic is inherently random enough to where all these imbalances are masked.

 

[ahadabans]

Something else probably worth discussing is how cube has evolved over the years and the ways that has altered balance.

The ajani's pridemate into path of bravery example above... what did that look like 10 years ago? Cube was much less synergy based and more just a collection of powerful cards. 10 years ago, that same play was Soltari Priest into Glorious Anthem.

Let's compare the two and see how it alters the way the game of Magic functions. Both are the same speed out of the gate, representing an aggressive strategy that puts early pressure on the board. So from that standpoint nothing changed. Both swing for 3 on T3 (assuming no other interaction). But each turn, pridemate will get bigger and this effect will snowball the more life gain is present in your deck. Priest is always just swinging for 3 (outside equips or whatever). This is balanced though by the lack of evasion and protection on pridemate. He's much easier to fog and remove (although there is an ongoing cost for most decks when you chump).

I'm not suggesting either is necessarily more powerful in a vacuum. But the Pridemate deck gives your opponent less time to respond. It grows big very quickly and has to be answered in short order. The Pridemate deck also has more variance when it comes to match-ups because of the lack of evasion but it's also much less fair if well built.

Making this swap in our cubes increased the skill level of drafting. But did making this swap increase interactive gameplay? I've seen more than one person mention recently (different threads) that the most fun part about cubing is drafting and building the decks - not actually playing the matches. And I think that is a clear shift from the early days of cube that I believe is related to the example above. We say priest is a non-interactive card because of shadow and pro-red, but it also gives your opponent 5 turns to find a solution. Pridemate has a shorter clock than that even if it's easier to interact with, so I'm not convinced we improved overall game play with this change.

And that is a tame example. There are much stronger cards (creatures in particular) that take this much farther. Like Kalonian Hydra, even Managorger Hydra I'm finding pretty ridiculous (hardened scales into managorger hydra is every bit as much a removal check as Kalonian Hydra. I've often asked myself if these cards are helping make my cube more fun or just more fun to draft.

The more we move into a place where plays or a sequence of plays produce games states that require faster response (or have no response), the less fun the game will become IMO. And that is why I find the never ending power creep with many cubes disconcerting. I'm not arguing against running powerful cards or synergies, just that we have to be mindful about how far we go with them as there is absolutely such thing as too much of a good thing.

 

[Grillo_Parlante]

I like this, I think Ahadabans mentioned this in his "be as diverse as possible" post above. If anything, it makes the discovery phase intricate and interesting, and gives players and you more dimensions to work with.

There are ups and downs to this, too diverse environments tend to reward prioritizing assertive strategies and ubiquitous answers, making them fairly easy to solve. The benefit is that the sheer size of the sandbox provides a sort of artificial depth to play in, until you realize that a certain % of the cube doesn't actually matter.

You don't want to get carried away with build arounds or overly techy cards, as there is a danger you end up watering down your environment with narrow cards that don't fit at all with what the cube actually wants to do.

I had a lot more success with those sorts of metagame relationships I've been talking about, where slightly pushed strategies provide a baseline to reevaluate other cards, adding depth, and opening up the possibility for counter play where other cards get reevaluated. You can do that in a relatively small space, and it tends to be more in line with what the cube wants to be doing.

 

[Grillo_Parlante]

Fully aware of the risk of enraging people, I have decided to post the extra credits video where I saw the noob tube concept.

BEHOLD

What I further liked about this was where he discusses first order optimal strategies, which I find are a plague upon cube, and while this may not represent a format solution, certainly feels like it cheats a format. It also represents natural imbalance within the cube, creating a benefit to running tech that requires the player to adjust their tactics when a first order optimal strategy presents itself. This adds depth to a format both due to the changing card evaluations it demands, but also because it prevents players from experiencing a format through the lens of a rather shallow strategy.

For an (again simplistic) example, let us pretend that my formats first order optimal strategy is to draft hexproof creatures like



Who is again our element A, causing other plays to start thinking of ways to counter that strategy, perhaps by drafting our old standby element B



This causes the troll player to adjust their tactics by drafting



Which causes us to--ok, ok, I'm partially trolling here, but seriously, the bouncelands (of which there are 20 in the cube) kind of were a first order optimal strategy here (though actually transition to being really deep cards: an important distinction), which has triggered a pretty extensive meta strategy chain from the formats inception.

I think that was kind of the missing part of the puzzle from the OP, as far as cube was concerned: to have a meta react there has to be some strategic bulwark its responding to, which is unlikely to exist solely off of troll/edict style relationships (though those relationships are beneficial on a micro scale, just because they provide greater depth to individual cards, which cumulatively is positive for a format). However, when you have a real density of a card type that offers huge rewards, that changes the equation immensely.

Now help me figure out how to do this with higher powered formats

 

[Jason Waddell]

I think for a meta to be cyclical, it has to be able to return to where it started. I don't see that happening. You can argue that people learned to evaluate spider spawning properly, or that Troll players learned to play other creatures (lol), but these things aren't ever going in reverse.

Neither is the control player going to just stop playing edicts. With a fixed amount of removal, it's going to get in there.

I agree that these dynamics are nice, I just feel like a lot of things are being improperly described in this thread, or that we're aspiring for something that isn't attainable.

Like, if you let people draft a fixed version of your cube for 15 years, I don't think it's going to cycle. People will learn proper evaluation, will get the ins and outs, will improve and evolve, but it's moving in a certain direction.

 

[Grillo_Parlante]

I think for a meta to be cyclical, it has to be able to return to where it started. I don't see that happening. You can argue that people learned to evaluate spider spawning properly, or that Troll players learned to play other creatures (lol), but these things aren't ever going in reverse.

Neither is the control player going to just stop playing edicts. With a fixed amount of removal, it's going to get in there.

I agree that these dynamics are nice, I just feel like a lot of things are being improperly described in this thread, or that we're aspiring for something that isn't attainable.

Like, if you let people draft a fixed version of your cube for 15 years, I don't think it's going to cycle. People will learn proper evaluation, will get the ins and outs, will improve and evolve, but it's moving in a certain direction.

Right, we are all in agreement on this, its just a tremendously awkward concept to describe otherwise. Again, the troll/edict example is just a simplified example, and not to be taken literally. The idea is to have a change occur on a dimension of play which creates a play-counter play relationship.

 

[Quirk]

I'm not convinced the Extra Credits stuff is helpful, but I don't want to get dragged into another exploration of their inadequacies at length. Their lack of understanding of game theory means that even using their terms puts our discussion in an awkward place. We aren't really talking about metagame reactions. We're talking about players learning to evaluate the format better.

"First order optimal strategy" doesn't really apply to Magic in the way it does to fighting games with tiers of fighters where a newbie can pick a top-tier character and button-mash and give an expert with a bad character a torrid time. Your strategy isn't "pick Grave Titan", your strategy is all the stuff you put in your deck, its synergies and gameplan. The difference between the newbie and the skilled drafter comes in several parts, some of which are:
1) Evaluation of cards
2) Understanding how synergies between cards can change those evaluations
3) Attitude to drawbacks
4) Understanding of higher strategic concepts such as mana curve, card advantage, etc.

Your true newbie, sitting down to draft for the first time, may well see a card that reads "gain 4 life" and a card that says "deal 3 damage to target" and conclude the first is better than the second. Your true newbie rare-drafts because rares are better. Your true newbie is going to be confused by cube, where lots of cards have gold symbols, and may tie-break on favourite colour or best picture or whatever.

So where are we coming in when we start talking about newbie-friendly strategies? Do we want to give Timmy and Johnny lots of fun fuel? Timmy's cool with including dragons; if you give him some good ones and he manages to live long enough to cast them, maybe they'll win some games for him. Johnny'll be all over whatever wacky synergies you've stuck into your cube. We don't really need to provide tools against their strategies, because their strategies are incoherent and Magic has a learning curve.

Or are we talking about people who know the rough shape of an aggro deck, a control deck, a mid-range deck, a ramp deck, who tend to draft a reasonable looking mana curve, but don't know your format yet? 'Cos, yes, they'll reach for raw power first and foremost, seeing as it's comforting to know heading into pick 2 that however badly they manage to screw up drafting the rest of this complicated-looking cube, at least Batterskull's in their deck and that'll flat out win some games.

And here's where I think we're speaking about the same thing, really: whatever cool ideas you've stuck into your cube, if they don't look like they can compete with that Batterskull, people are probably going to be averse to explore them. Certainly pick 1, my thoughts tend to be "Oh yeahhhh, this looks powerful" or "Let's try building around this" but if I've struck out on the first and the "this" happens to be Barren Glory I'll move right on to the last option of "I guess there's some removal in this pack at least".

So maybe we're really looking for stuff that's competitive in raw power with the alternatives and that supports the synergies and cool ideas you're introducing to act as that crucial first pick; and that maybe means the alternatives need dialled back to the point that the synergy cards with the most raw power can fight for first pick when they meet in the same pack. That might be removing Batterskull, or it might be making it relatively bad in the format (hi, welcome to the Utility Artifact Hate subdraft - want Ancient Grudge? Master Thief? Carry Away?). But it's hard to do this without occasionally having to cut a fan favourite...