General [Design Discussion] Perfect Imbalance

[Grillo_Parlante]

So I stumbled across this old extra credits video

Which got me thinking about a few things. The basic idea is that games are made more interesting by having a meta game, and you create meta games by including a certain level of deliberate imbalance. I wish they had a transcript instead of a video, but the argument is presented in three parts:

A. Metagames are Good

1. introducing deliberate imbalance creates a metagame
2. metagames are important because they introduce an evolving state of play that keeps any one style of play from being definitively correct (E.g. MODO drafting/good stuff drafting/signet drafting)
3. This allows players to experiment with different approaches to the game.
4. Keeping the format fresh by providing an interesting and evolving problem; players at all skill levels change and grow as players

B. Perfect Imbalance isn't included Haphazardly

There is than a WOTC reference, which it would probably be healthier for all of us to ignore...but in sum:

1. You establish a baseline for balance
2. Allow certain game elements to deviate from that baseline to a reasonable degree. This creates a game around the game of trying to figure the system out, gaining an edge in a nearly but not perfectly balanced system.

C. Cyclical balance creates an evolving state of play that adds depth to a format

1. You include game element A that is a little better than average, which attracts a large number of players
2. Other players start thinking of ways to counter game element A, which they find in game element B
3. Game element B become popular, causing players to think of another counter, which they find in game element C.
4. This can continue onward, but at some point it will cycle back to game element A, and the system renews itself.

Begin the Arms Race

So this helped me conceptualize a few things. As much as pauper is a format plagued by problems, it does include a lot of interesting relationships that I ported over to the penny cube, which I was calling meta relationships, a term I made up that I don't think was helping anyone. The closest I could get to explaining how this worked was using clumsy language like "reasonable conditionality" on removal. I think I can do better now.

Lets begin with the simplified story of Troll Ascetic, to demonstrate the benefits of cyclical balance to trigger these strategic arms races.



Troll Ascetic is an above the curve threat over here. The combination of hexproof (in a format with ample auras), regeneration, and a 3/2 body makes this a very dangerous turn 3 play out of any green deck. This is our game element A, a better than average card sure to be first picked for its pushed power level.

For a week or two, decks running troll ascetic do very well, and players in the format start scratching their heads about how to deal with this threat.

Than one day, a player who had previously been drafting "good" removal like murderous cut, sees an effect type in the pack that they had previously scoffed at.



Edict effects are the perfect solution to the ascetic decks that want to get their troll out and voltron it up. This becomes our element B.

For a couple weeks the edict decks do very well, and the troll decks suffer. The troll players start to scratch their heads, wondering how they will get around this problem. Than they see some cards in the draft they had previously scoffed at.





When ran alongside the troll (or other hexproof threat) these cards become edict fodder, protecting the hexproof threats. Troll Ascetic decks return to dominance, and this is our element C.

The edict players go back to the drawing board, trying to figure out how they will overcome this. They start to look at wider or recursive removal options.





This is our element D. There are a few more iterations of this I could do, but I think you get the idea. The decks are evolving and contorting themselves in response to changing strategic needs, providing a continuous problem for players to solve, and making the format deep. Note that I could have started the chain with murderous cut (a pushed removal spell in its own right) and had Troll Ascetic be the strategic response to it. In the end, the decks may contort themselves so much that troll ascetic is no longer played, which will reduce the value of edicts, which will result in the chain restarting itself as the meta becomes soft to troll again.

This is one of the major chain types I took from pauper, though there are others involving damage based removal (and especially damage based sweepers) and growth or sacrifice effects. Pauper's best element is this dance between control tech and aggro tech.

The other major one (which I've talked about more extensively) are the effects of bouncelands on a format, which have started their own cyclical chain over here, due to their pushed nature but CIPT cost. Unlike removal sub-chains, however, the land based relationships are tied to the very nature and prevalence of the formats aggro, midrange, and control decks.

This is way more interesting than simply running ubiquitous good cards, and hope that at least it ends the discussion of whether a flat or broad power band is good: the answer is neither, you want a relatively close power band with some reasonable fluctuations.

Power Max Problems

And of course, I think this is the fundamental problem with any sort of raw power max approach to a format. If I have a glut of planeswalkers or other powerful cards that are above the curve, but don't invoke any element B, that strategy may just take over the format. In the absence of any metagame, drafting around those pushed cards becomes the defacto strategy, resulting in a shallow format.

This also impacts the aggro-midrange-control relationships we were discussing before, which should ideally be a cyclical relationship if we are going to run with it. Those relationships can be designed on paper, but without an actual active strategy-counter strategy relationship, they only exist so far as the players' whims allow them to exist.

Thoughts?

 

[FlowerSunRain]

I like this part and think it is 100% true:

This is way more interesting than simply running ubiquitous good cards, and hope that at least it ends the discussion of whether a flat or broad power band is good: the answer is neither, you want a relatively close power band with some reasonable fluctuations.

However, I hate that video and think the creator is dead wrong about a lot of the ideas in it. I think those ideas poorly translate to cube design.

 

[vennythekid]

This is a super interesting topic!
The big quesiton here is: what is a good baseline that players will enjoy playing at? I've considered dropping my cube in powerlevel considerably, but it already excludes lots of "cube cards" that my group enjoys playing with. So basically, what's the highest powered baseline that can be acheived, and what do those cards look like? What sorts of cards sit right above this band? What types of counter/meta play should be introduced?

I personally think Riptide cubes have been succesful thanks to a few defining cards which should set the tone:

 

[ahadabans]

Nice post. As always, very well thought out and insightful.

I agree with the message. Especially the part about raw power and how there can be a defacto (at a minimum shallow) strategy if this is the only consideration. At the same time, I struggle to find the sweet spot, power level wise.

My cube is still pretty high powered all things considered. And there's appeal with my players when it comes to powerful cards. It's part of what guys know cube to be (right or wrong). And it's fun opening packs and being visually assaulted by 15 cards that could all conceivably be all-star cards in casual kitchen table Magic decks. This idea was what drew me into cube and what got several of my friends fired up to play.

I guess where I'm going with that is at what point can you still craft a deep meta and still be playing with "broken" (or powerful) cards? Can that even be done?

I wanted to talk about the chain you described (troll ascetic, edict, etc). I sometimes run into scenarios during testing that make me question includes in my cube. I'll give a specific example. Abyssal Persecutor. I love the design of this card and I appreciate the fact that you have to couple it with sacrifice outlets to win with it. That said, it's really not a hard mini-game to navigate (so it strikes me as a minimal drawback truth be told). And 6/6 flying tramplers are exceedingly difficult to deal with (especially ones that only cost 4 mana). I'm honestly not sure this card is good for my environment even though I really like it.

Just the other day, I found myself staring down this card and it occurred to me that the deck I was playing had no way to deal with it. I couldn't burn it, block it, destroy it or race it (OK, maybe I had one answer in my deck, but it wasn't in my hand and the odds of me drawing it before I died was remote). As soon as persecutor resolved, I had essentially lost the game. Now, it's not to say there aren't answers in my cube for this card or other decks would have been less screwed. There are plenty of answers to cards like this. But you won't always have them in your deck. And sometimes cards like this are "silver bullets" in match-ups. I don't know that I really like that. We can say theoretically that you should value unconditional targeted removal more highly to deal with things like Persecutor, but that isn't always practical. How much of that removal can you run in your cube before there's over saturation and how many decks can run that removal anyway (not all colors have these types of cards even if you ran as many as you could)?

At lower power levels, this feels like less of a problem. Primarily because one card is very unlikely to auto-win games the way something like Persecutor can. I sometimes wonder if the game of Magic simply works better at a power level closer to Pauper than at your typically rare cube power level. But it could just be inherent in the game of Magic itself (after all, it's essentially just a game of threats and answers). Sometimes you lose because your opponent had the right cards and you didn't. Still, this is my biggest argument against truly pushed power cards that immediately demand answers or you lose or cards that generate obscene value for minimal cost (particularly ongoing value that snowballs over time). It's also why I like synergistic mechanics so much (at least with that kind of power you had to put the pieces together in draft and during the game - it feels more earned that way).

I've tried to look at cards I add to my cube with that question in mind. Is this a card that must immediately be answered or the opponent will just lose the game? And if the answer is yes, how easy is that card to answer and is there a proper cost associated with playing it or unlocking the cards power (via synergy or whatever)?

 

[Grillo_Parlante]

You can theoretically go pretty high power, and the original foundation for those riptide style cubes was based on a cyclical relationship from legacy.



I've been hating on wasteland lately, but maybe the question is more about finding ways to better implement existing cyclical relationships in cube, and expand on new ones. The thing that makes this challenging is that Legacy is the only high power format that really does this well (which is why I love you legacy).

One from Modern that I like

 

[Face]

This is definitely an interesting principle to apply to limited. If drafters are informed of these relationships in your particular format before the draft, or simply catch on to it after the first couple, just the presences of element A/B/C/Z will add that extra dimension to their pick decisions. I do see, however, that players need to be aware of this specific tone of metagame for it to show any significant returns.

Constructed formats are supposedly built on this style of metagame, but historically have not been executed so well. Fortunately, the recent standards have been shaping up interestingly what with the volatile rotations. It still feels a little like a midrange slug-out right now, but I'm confident it'll get there eventually.

I feel the best vehicle for us to weave this into design is through duplicate sealed. It's a nice halfway between limited and constructed with a focus on "build-your-own-metagame." Analyzing constructed formats and your options can be overwhelming at times, but when you look at a pool of 100 cards at most, it's a little easier to see what synergies are available to you and to each of your opponents, leading to much more interesting deck construction. I feel that in terms of the situation you've described, duplicate sealed would play out closer to it than a draft would.

I feel this would also be a lot of fun to try designing. You'd have free reign with running multiples, build-arounds, bombs. This also opens up to a lot of asymmetrical design choices in the kind of fixing you'd include, or the kind of curve each color implements. There's also the aspect of crafting the relationships between A/B/C/Z a lot more intimately.

But then again, opening this up to a playgroup of 8 would result in a collection just under 800 cards, so it's probably not an undertaking for the faint of wallet.

 

[Grillo_Parlante]

I wanted to talk about the chain you described (troll ascetic, edict, etc). I sometimes run into scenarios during testing that make me question includes in my cube. I'll give a specific example. Abyssal Persecutor. I love the design of this card and I appreciate the fact that you have to couple it with sacrifice outlets to win with it. That said, it's really not a hard mini-game to navigate (so it strikes me as a minimal drawback truth be told). And 6/6 flying tramplers are exceedingly difficult to deal with (especially ones that only cost 4 mana). I'm honestly not sure this card is good for my environment even though I really like it.

Just the other day, I found myself staring down this card and it occurred to me that the deck I was playing had no way to deal with it. I couldn't burn it, block it, destroy it or race it (OK, maybe I had one answer in my deck, but it wasn't in my hand and the odds of me drawing it before I died was remote). As soon as persecutor resolved, I had essentially lost the game. Now, it's not to say there aren't answers in my cube for this card or other decks would have been less screwed. There are plenty of answers to cards like this. But you won't always have them in your deck. And sometimes cards like this are "silver bullets" in match-ups. I don't know that I really like that. We can say theoretically that you should value unconditional targeted removal more highly to deal with things like Persecutor, but that isn't always practical. How much of that removal can you run in your cube before there's over saturation and how many decks can run that removal anyway (not all colors have these types of cards even if you ran as many as you could)?

They had another video, which is connected with this idea, but it was basically amounted to it being healthy to have some sort of "noob cannon" in a game, and that is sort of the standard I've been using for questions like this. The actual argument is kind of awkward to apply to magic, but one of Alex Ullman's articles provided a pretty good intellectual bridge for me.

For those that don't know, the "noob cannon" is any sort of overpowered weapon in a FPS that first time players relay upon to get kills, but which isn't actually very good. The under slung grenade launcher in COD or the shotgun in counter strike are the best examples. For a new player coming onto a server filled with sharks, it provides enough punch so a bad player can learn the game against experienced competition and not become discouraged and quit.

While I was just praising the way pauper encourages cyclical exchanges between threats and answers, one of the reasons that the format has become sick is that its taken that concept to an extreme and now there is no "clear best answer" in pauper: everything gets trumped by something in someway, which is terribly awkward for answer based decks that are being pressured to have the right cards in hand or lose. The result is one part of the meta collapses, and the format coalesces around narrow assertive decks that want to minimize interaction.

As a result, I don't think you want to have everything strongly interacting with these meta chains: its ok, even desirable, to have some number of ubiquitous powerful cards: the "noob cannons" in a sense. New players or casual players are going to be overwhelmed if they have to just figure these relationships out, and there has to be some sort of gateway tech to help them survive while learning the format. This is also kind of like good bomb rares in limited, which should be beatable by more experienced players, but allow less able players snag some wins and feel good before becoming better.


You just don't want those gateway cards to take over the format, you want people to grow out of their crutches, and transition to better strategies. An easy example of that danger from cube is when you see players relying heavily on planeswalker strategies or other good stuff or artifact ramp strategies: thats something you might draft your first time and have fun with, but the cards are so powerful you end up never gravitating away from (for good reason).

So how good is Persecutor? Is it so strong that people are going to cling to it or a strategy it represents? Or does it provide a welcome rush of power that a player will enjoy for a while, before transitioning to more advanced strategies? If good players and well drafted decks can generally deal with Persecutor, its probably ok to keep.

 

[Jason Waddell]

I don't think cyclical balance exists in a fixed-resource setting. In constructed, players choose which resources to bring to the player, and on a given day you can have, say, 50% of the players wielding a certain aggro variant. This doesn't exist in cube. An "evolving meta" isn't a very prominent aspect of cubing in my opinion.

 

[Quirk]

While I was just praising the way pauper encourages cyclical exchanges between threats and answers, one of the reasons that the format has become sick is that its taken that concept to an extreme and now there is no "clear best answer" in pauper: everything gets trumped by something in someway, which is terribly awkward for answer based decks that are being pressured to have the right cards in hand or lose. The result is one part of the meta collapses, and the format coalesces around narrow assertive decks that want to minimize interaction.

You disappointed me there. I remembered Pauper being Delver-dominated, and this paragraph made me go and look in hopes that that had changed. It hasn't, Delver's still a third of the metagame. The counterspell is still pretty close to the "clear best answer", and Delver runs about ten (six unconditional), plus bounce. And on top of that, Delver lists haven't changed very much the last few years.

What do you mean by narrow assertive decks that want to minimise interaction, anyway? Every aggro deck is happy to goldfish. Every combo deck just wants to go off. Both these styles of decks in Pauper are playing permanents to the board that can be interacted with, though. The old Grapeshot/Empty the Warrens storm combo didn't, but it's banned; and there's Burn, obviously, but Burn's not even all that good in Pauper.

It is a high-pressure format, and card advantage seems sort of out of vogue for the time being (though some of the lists running Ichor Wellspring obviously still have intentions in that direction). Still, MBC was very much a card advantage deck, and UR Cloudpost before Cloudpost's banning was so pure a control deck that many versions' primary win condition was rebuying Rolling Thunder with Mnemonic Wall to throw at the opponent's face.

As a result, I don't think you want to have everything strongly interacting with these meta chains: its ok, even desirable, to have some number of ubiquitous powerful cards: the "noob cannons" in a sense. New players or casual players are going to be overwhelmed if they have to just figure these relationships out, and there has to be some sort of gateway tech to help them survive while learning the format. This is also kind of like good bomb rares in limited, which should be beatable by more experienced players, but allow less able players snag some wins and feel good before becoming better.

I think an important part of the "noob cannon" is that it's seen as suboptimal by skilled players. This isn't true of cards like Batterskull, which a pro will happily take out of a non-powered cube p1p1, or even Thragtusk. And this is where it gets hard - people play cube to play with powerful cards, but the most powerful cards are broken even in unskilled hands.

A while ago I had a conversation with a colleague who played a little Magic. His assessment of the strategy was: "it boils down to whoever plays the biggest creature wins". Vizzerdrix as noob cannon? Pretty much any format will sort that level out. A somewhat more skilled player of my acquaintance still dislikes cards with drawbacks and symmetrical effects. Maybe this points to some kind of guidance: are the cards which force you to make trade-offs and make you care about sequencing actually more powerful in your format than the "bombs" that look obviously above the curve? Is there really an edge to be had from careful play, or is the ultimate skill test on your skill-testing cards whether you understand the meta enough to ignore them and pick Bloodbraid Elf?

 

[Jason Waddell]

I think an important part of the "noob cannon" is that it's seen as suboptimal by skilled players. This isn't true of cards like Batterskull, which a pro will happily take out of a non-powered cube p1p1, or even Thragtusk. And this is where it gets hard - people play cube to play with powerful cards, but the most powerful cards are broken even in unskilled hands.

A while ago I had a conversation with a colleague who played a little Magic. His assessment of the strategy was: "it boils down to whoever plays the biggest creature wins". Vizzerdrix as noob cannon? Pretty much any format will sort that level out. A somewhat more skilled player of my acquaintance still dislikes cards with drawbacks and symmetrical effects. Maybe this points to some kind of guidance: are the cards which force you to make trade-offs and make you care about sequencing actually more powerful in your format than the "bombs" that look obviously above the curve? Is there really an edge to be had from careful play, or is the ultimate skill test on your skill-testing cards whether you understand the meta enough to ignore them and pick Bloodbraid Elf?

Short (possibly not useful) answer: it depends on the bombs in question. But I think as a designer you're in control of that, via your selection of cards, and how you weight raw power in your card selection, relative to cards with sequencing options.

As usual, I think a lot of this happens naturally when you keep your curve low. The more cards in your hand are competing to be cast, the more sequencing depth there is in the format (and more possibility for misplays, as CML loves to point out).

 

[Quirk]

Short (possibly not useful) answer: it depends on the bombs in question. But I think as a designer you're in control of that, via your selection of cards, and how you weight raw power in your card selection, relative to cards with sequencing options.

Raw power unfortunately makes for easy first picks though. It doesn't have to be Grave Titan; Consecrated Sphinx or even Flametongue Kavu will do. I happily grab the most powerful thing I see in the first pack - and I'm stoked about it, because the draft is going well, in a way I probably wouldn't be if I'd picked up a Wild Mongrel or Bloodghast. As a designer it's easy to slip in cards you're excited to play, which make you feel good about your deck when you pick them, but which make games less interesting than alternatives would.

That said, sometimes the card you're excited to first pick is broken only if you build around it right. I think this is less unhealthy. It takes you out of a raw power mentality from the start.

As usual, I think a lot of this happens naturally when you keep your curve low. The more cards in your hand are competing to be cast, the more sequencing depth there is in the format (and more possibility for misplays, as CML loves to point out).

I agree. I think it's still easy to stick in absurdly pushed low drops though - Brimaz is a case in point. Attractively pickable, but unlikely to lead to many hard decisions.

 

[Grillo_Parlante]

Thanks for the responses everyone.

I don't think cyclical balance exists in a fixed-resource setting. In constructed, players choose which resources to bring to the player, and on a given day you can have, say, 50% of the players wielding a certain aggro variant. This doesn't exist in cube. An "evolving meta" isn't a very prominent aspect of cubing in my opinion.

I think thats more a reflection of density issues with certain cards. The easiest rebuttal I can offer are the 20 bouncelands (40 CIPT lands) from the penny cube: thats not a fixed resource, but at that density its pretty close to it, and is probably the most prominent aspect of that cube. You don't need to have 20-40 of an effect to create a meta relationship, but you do need more than 4 if you want it to feel significant to a format.

You disappointed me there. I remembered Pauper being Delver-dominated, and this paragraph made me go and look in hopes that that had changed. It hasn't, Delver's still a third of the metagame. The counterspell is still pretty close to the "clear best answer", and Delver runs about ten (six unconditional), plus bounce. And on top of that, Delver lists haven't changed very much the last few years.

What do you mean by narrow assertive decks that want to minimise interaction, anyway? Every aggro deck is happy to goldfish. Every combo deck just wants to go off. Both these styles of decks in Pauper are playing permanents to the board that can be interacted with, though. The old Grapeshot/Empty the Warrens storm combo didn't, but it's banned; and there's Burn, obviously, but Burn's not even all that good in Pauper.

It is a high-pressure format, and card advantage seems sort of out of vogue for the time being (though some of the lists running Ichor Wellspring obviously still have intentions in that direction). Still, MBC was very much a card advantage deck, and UR Cloudpost before Cloudpost's banning was so pure a control deck that many versions' primary win condition was rebuying Rolling Thunder with Mnemonic Wall to throw at the opponent's face.

Alex Ullman has a very good article on just this topic that I think you might enjoy. It used to be that there was a clear best spot removal spell in the format in the form of flame slash, but with the printing of gurmag angler thats no longer the case*. Counterspell and daze are the only good general answers in the format, which is partly why the only deck that can efficiently run them is 1/3 of the metagame.

Those factors, among others, warped the entire meta, condensing decks towards the narrow and assertive. Some of those changes are pretty dramatic, such as MBCs focus on supporting and resolving gray merchant, and others much more subtle, such as the stompy changes he addressees. There is also a nice bit at the end about some of pauper's meta relationships.

*This is another easy example of how you can create these sorts of cyclical relationships. If almost all of my damage based removal spells deal, say, 3 damage, and I offer a selection of pushed threats with 4 toughness, that can be enough to trigger a chain.

 

[Jason Waddell]

Thanks for the responses everyone.



I think thats more a reflection of density issues with certain cards. The easiest rebuttal I can offer are the 20 bouncelands (40 CIPT lands) from the penny cube: thats not a fixed resource, but at that density its pretty close to it, and is probably the most prominent aspect of that cube. You don't need to have 20-40 of an effect to create a meta relationship, but you do need more than 4 if you want it to feel significant to a format.

Perhaps I was unclear. By fixed resources, I mean, there is a finite supply of cards in your environment. The players can "want" to play more edicts (or whatever), but there are a finite number of edicts. The "meta" so to speak is static. This is not a bad thing, or a good thing, it's just a thing.

I agree that bouncelands create a *different* meta, but that it's an independent static meta. Things will change in drafting strategy based on your players (e.g. how heavily you have to prioritize fixing depends on the people at your table), but that's not a cyclical thing.


Follow up: when I'm drafting I am super happy if I can start with some powerful card and adapt from there. First-pick FTK is perfectly fine, in my opinion. Part of my beef with Modern Masters was that there was so little design space for playing a non-pigeonholed archetype.

 

[ahadabans]

I love these kinds of discussions. Not sure about everyone else, but I personally get a lot out of them.

A couple themes I am hearing which feel like valuable tools (ones I want to make sure I remember when this tread ends):

1. Lower curve means more opportunity for sequencing and decision points. And that's generally desirable at pretty much any power level. I think that ties in perfectly to the discussion about balance since it essentially boils down to the whole threat/answer mechanic which is one of the foundations of this game. For me personally, I often forget about trying to lower my curve as my cube is largely midrange by design. But that still doesn't mean I can't be running a lower mana curve. Generally, every time I replace a high cost spell with something lower cost, it tends to open up more early play decisions where most games are won or lost. It increases interactivity. And that is a good thing.

2. Finding the right balance between your "noob cannon" power level and your more skillful ("synergistic"? for lack of a better word) card power level. You don't want your easy-to-play cards to be so powerful that there is no incentive to "graduate" to more advanced deck building. Talking about cards like Batterskull, Wurmcoil, etc. that just make all decks which can cast them better.

That last topic has probably been beat to death in other threads, but have we ever really put together (definitively) what we believe that this list to be? I'm certain it's relative to the overall power level of your cube so this may not be something easy to define that will be applicable to everyone. But there are certainly going to be obvious candidates for such a list, along with gray area type cards. Would a sticky thread about this be useful to people? I think the vast majority of the people here run less than power max lists, so it feels widely applicable.

Speaking for me specifically, I know I don't get to cube with friends as much as I would need to really get solid feedback on my meta. So I spend a great deal of time on my own just testing. But that means I only get input from my own frame of reference, which is where forums like this are really valuable. I like to play certain types of decks. I am good at building certain types of decks. I can do really cool things with some cards, but other cards I do not use well. So testing decks that maximize those other cards I'm clumsy with is very hard for me to do. A card that is actually broken but that I have never really played or exploited can very easily just wind up misused by me in testing and appear totally benign when in reality it should be on a banned list (or worse is just poisonous and adding zero value to my list). It's very hard to figure some of that out on your own.

 

[Grillo_Parlante]

Perhaps I was unclear. By fixed resources, I mean, there is a finite supply of cards in your environment. The players can "want" to play more edicts (or whatever), but there are a finite number of edicts. The "meta" so to speak is static. This is not a bad thing, or a good thing, it's just a thing.

I agree that bouncelands create a *different* meta, but that it's an independent static meta. Things will change in drafting strategy based on your players (e.g. how heavily you have to prioritize fixing depends on the people at your table), but that's not a cyclical thing.


Follow up: when I'm drafting I am super happy if I can start with some powerful card and adapt from there. First-pick FTK is perfectly fine, in my opinion. Part of my beef with Modern Masters was that there was so little design space for playing a non-pigeonholed archetype.

The way it works over here is that players change their draft strategy based on perceived strategic needs. There is a finite supply yes, but people are competing over that finite supply based on their card evaluations, which reflects the metagame if one exists. If a player "wants" more edicts, they just have to draft to reflect that, and if they are consistently successful with that draft strategy, eventually the meta will have to respond, card evaluations will change, draft strategies will adjust, and we will have entered another iteration of the cycle. I think this is especially helpful in cube, as its a given you will get enough playables.

Take for example the bouncelands. Over here they were the default non-aggro strategy for a long time, with basically every deck running 15 lands in a bid to abuse them. In response, decks started to prioritize more disruptive tools to prey upon them. We're now in a phase where the players who had previously heavily prioritized bouncelands in the draft are looking at augmenting or shifting their strategies. The gainlands, for example, which they had previously scoffed at, look like a pretty reasonable compromise, and some of the traditionally bounceland obsessed players have been augmenting more and more of their mana base with them. Eventually people (such as myself) will stop prioritizing the disruptive tools as much, and this allows a shift back towards bounceland abuse, in a very fluid cycle.

 

[Rasmus Källqvist]

Perhaps I was unclear. By fixed resources, I mean, there is a finite supply of cards in your environment. The players can "want" to play more edicts (or whatever), but there are a finite number of edicts. The "meta" so to speak is static. This is not a bad thing, or a good thing, it's just a thing.

But isn't part of what makes this metagame happen the fact that you get different strategies, depending on what priorities (what picks) the players make with the resource pool your cube represents during a draft? Kind of like how a "highlander" moba (no double of the same hero on the same team) can create a meta game based on what team compositions are favored. It's the same pool of resources, but you're limited to what you use of that pool for any given round.

You also get this effect of, if a piece of the resource pool gets picked off its done, you don't get any more of those. So in that way its unlike the standard meta game where every participant can use the exact same part of the resource pool.

I also like that this sort of ties into negative space, which is something I used to neglect but have started to think more of. The idea that if you limit a certain effect to a single color (one section of the resource pool) you'd get people gravitating towards that. But that was more about doing that with the card types and different strategic options, not with power level, but I think doing that sculpting with the power level is even better.

You can sort of create this designer-metagame that way both by limiting what a color can do, and where the powerful cards are. I also think that helps clear up some of the overwhelmingness of opening that first booster in someone elses cube. If the booster itself presents a sort of map over the metagame it's much easier to start figuring the cube out.

Some cards will stick out, and when you start moving into a certain direction you'll notice that the packs follow suite and guide you a bit simply by offering you the cards you want in only one section (ie green and blue won't have good ways to deal with on board creatures, which gravitates you towards mardu for removal cards).

 

[Chris Taylor]

So, this topic exploded before I could comment
Big fan of extra credits. Game design is one of my passions that I don't actually get a lot of exposure to, since cube design is kinda like game design, but with a LOT of stipulations.

I like this part and think it is 100% true:

This is way more interesting than simply running ubiquitous good cards, and hope that at least it ends the discussion of whether a flat or broad power band is good: the answer is neither, you want a relatively close power band with some reasonable fluctuations.

However, I hate that video and think the creator is dead wrong about a lot of the ideas in it. I think those ideas poorly translate to cube design.

Can you elaborate FSR? I'll bite that a lot of these ideas translate poorly to cube design, but if something in here is noticeably incorrect, I'd love to know

 

[ahadabans]

The way it works over here is that players change their draft strategy based on perceived strategic needs. There is a finite supply yes, but people are competing over that finite supply based on their card evaluations, which reflects the metagame if one exists. If a player "wants" more edicts, they just have to draft to reflect that, and if they are consistently successful with that draft strategy, eventually the meta will have to respond, card evaluations will change, draft strategies will adjust, and we will have entered another iteration of the cycle. I think this is especially helpful in cube, as its a given you will get enough playables.

Take for example the bouncelands. Over here they were the default non-aggro strategy for a long time, with basically every deck running 15 lands in a bid to abuse them. In response, decks started to prioritize more disruptive tools to prey upon them. We're now in a phase where the players who had previously heavily prioritized bouncelands in the draft are looking at augmenting or shifting their strategies. The gainlands, for example, which they had previously scoffed at, look like a pretty reasonable compromise, and some of the traditionally bounceland obsessed players have been augmenting more and more of their mana base with them. Eventually people (such as myself) will stop prioritizing the disruptive tools as much, and this allows a shift back towards bounceland abuse, in a very fluid cycle.

At some point though, one of these cycles will prove to be the more efficient (most likely to work against the field) though right? They can't perfectly be balanced to the point where it will just flow like rock/paper/scissors endlessly. So from that standpoint, I agree with Jason that to some extent this is pretty static. In some more obvious scenarios, good players could work the cycle out in their minds and jump to the most efficient end point bypassing the cycle entirely. Take your Troll Ascetic example. If I'm putting my eggs in a hexproof creature, I'm already thinking about edicts since that is my only weakness. I would have skipped the first two phases of that cycle knowing that this play ruined me and knowing that there are enough edicts in the cube that it's probable I may face a deck packing one. It's also logical in my mind to have multiple threats because sometimes you have to go wide (edict or no).

With that said, I think what you are saying about an evolving meta DOES actually happen. But maybe not exactly for the reasons you state. I am referring specifically to how often we change our lists (tied specifically to how often we are receiving new cubeable cards). I can tell you that I have been replacing cards in my list that I just put in not too long ago, many times with very little performance feedback on those cards. That is how fast my list is changing now. Probably not a good thing, but that is where we are at I feel. And as the list changes, these cycles we are talking about end up evolving. So it might feel like a cycle but really it's just parameters altering how the cube functions as a whole.

 

[Grillo_Parlante]

In some more obvious scenarios, good players could work the cycle out in their minds and jump to the most efficient end point bypassing the cycle entirely. Take your Troll Ascetic example. If I'm putting my eggs in a hexproof creature, I'm already thinking about edicts since that is my only weakness. I would have skipped the first two phases of that cycle knowing that this play ruined me and knowing that there are enough edicts in the cube that it's probable I may face a deck packing one. It's also logical in my mind to have multiple threats because sometimes you have to go wide (edict or no).

You have to keep in mind that the troll example is a very simplified example for illustrative purposes, but to stick with the example, the durable threat deck you are drafting now has to balance between buffs, fodder, and threats. If edicts are not being prioritized in the draft environment, you may be drafting an objectively worse deck. In addition, you haven't broken the chain in the example you gave, just moved the point where we are starting, and this is because its a well designed chain.

I draw a distinction between this type of approach and just changing mass numbers of cards. One is about an evolving problem, the other feels more like supplanting old problems with new ones.

 

[sigh]

Hello Riptide! First post here after lurking for awhile. This is the sort of thread that really gives me life. I'm an analytical engineer by trade, so this sort of discussion gets the juices flowing.

There's a lot that's been discussed above that I like:
-A Meta can be static (mostly agree, see below)
-Metas can also change dynamically
-Specialization of colors can help in guiding a direction for colors. This is how the color pie works.
-changing the available resources will force a change of Meta.

I think an important thing to keep in mind is how *knowledgable* the crowd is about the meta. A cube that has been drafted a thousand times by experienced drafters may very well have been "solved", but a newer cube with newer players isn't "solved" yet. In this regard, someone could play a Troll deck for the very first time, and this in return increases the desirability of Chainer's Edict or fleshbag marauder by several picks. Now this one interaction has been solved, and the meta progressed as Grillo and others have discussed.
In short, the meta inputs are resources, player's present, and knowledge. Player's can only learn so fast, so these chains are a necessary by-product of solving the environment.

On that note, it's probably almost impossible to fully solve a meta, of any sort (look at modern lantern control popping out of a relatively sluggish meta, for instance). So if anything, the rate at which these loops slows down, but never stops. It becomes more optimizing on the solutions that have appeared and less finding the solutions. And then changing the meta through resource-change can actually speed it up by far more than one might anticipate. Same goes for basically any input change.

In any case, super sweet thread! This is exactly the sort of discussion to have with yourself and amongst others to progress your design.

 

[Rasmus Källqvist]

I'd also like to note that these meta game cycles as described here (A->B->C->A) can be seen as an oscillation.

An oscillation is either stable and continous to cycle indefinitely, or there is some damping that causes the oscillation to slowly die out.

Now, oscillation is usually describing some kind of shifting between amplitudes of a signal or mechanical movement so it's not a perfect analogy, but my idea is that just as an oscillation is dependent on certain criterias to be met (the troll and edicts if you will) you can alter parameters to get another behavior of the cycle.

In less analogous terms; you can stir up a solved meta by changing the contents of the resource pool (the cards in your cube). We all know this as the de facto method of keeping a format alive, both for us as cube designers and for WotC with ban lists.

 

[FlowerSunRain]

Can you elaborate FSR? I'll bite that a lot of these ideas translate poorly to cube design, but if something in here is noticeably incorrect, I'd love to know

It doesn't address the reality of the situation. The meta-shifts as described in the video are rarely cyclical without outside intervention, cyclical design creates very frustrating situations for players and most well designed games would be significantly less interesting if they had cyclical balance.

1) League of Legends is an atrocious example of an exemplar of cyclical balance. Most meta-shifts in the game are triggered not by the fluid nature of a well designed metagame, but instead by direct intervention from the designers. The metagame looks like it is constantly changing, but not because players are figuring anything out. Its changing to respond to changes to the game itself. The flow is meta develops -> developer changes something -> meta adjusts. It is linear. Once in a blue moon the meta changes on its own as players figure dtuff out, but because the developers are CONSTANTLY changing things its not the primary driving force. Cyclical balance is a farce. If the designers stopped tinkering, the metagame would eventually settle, with MAYBE a linear shake up when new tech is discovered. See: basically every good fighting game ever.
2) Players have preferences and enjoy the game more when they can express them. In a cyclically balanced game, if you are a dedicated rock player, it sucks when paper is in fashion. It kinda makes you not want to play.
3) The situation of "How can I beat X with Y?" is way more interesting then "I can't beat X with Y, let me play Z!" I disagree with the premise that playing a different character is "exploring a new strategy". Gaining superior mastery in a game to overcome a challenge is a wonderful thing. Picking a different box on the character select screen isn't engaging, exciting or skillful.

 

[sigh]

I'd also like to note that these meta game cycles as described here (A->B->C->A) can be seen as an oscillation. An oscillation is either stabile and continously cycles indefinitely, or there is some damping and the oscillation slowly dies out. Now oscillation is usually describing some kind of movement between amplitudes of a signal so it's not a perfect analogy but my idea is that just as an oscillation is dependent on certain criterias to be met (the troll and edicts if you will) you can alter parameters to get another behavior of the cycle. In less analogous terms; you can stir up a solved meta by changing the contents of the resource pool (the cards in your cube). We all know this as the de facto method of keeping a format alive, both for us as cube designers and for WotC with ban lists.

An excellent way to describe it! And to further ourselves down the technical rabbit-hole, our damping could be attributed to the "knowledge" variable (to a degree). If the player's never really progress in their understanding of the optimum, the cycle will continue (A B C A) etc. Only very experienced players can probably "damp" it down to a negligible amount. At this point, a change in resources can kickstart the whole process again. To continue using our little Troll, taking out all of the edicts would stir this little relationship up a lot. Now maybe the best way to combat the Troll is with token strategies that can chump block. From there, board wipes may become more valueable, and then back to the Troll. The cycle is reborn.

It doesn't address the reality of the situation. The meta-shifts as described in the video are rarely cyclical without outside intervention, cyclical design creates very frustrating situations for players and most well designed games would be significantly less interesting if they had cyclical balance.

A good point, and cyclical is indeed simplistic at best (and yes frustrating). But a cube is far too multi-variable to analyze as-is. So cyclical ABCA is a good model for understanding. XvY is another. And any new meta will almost guaranteed have to be multidirectional to a degree. Before people even know what's what about X v. Y, they have to play around with the options out there, and this can be kinda "circular". Maybe more experienced players will finish this almost immediately and move to XvY, or ForceThisThingAlways, or whatever. Almost no model is a complete farce, just more accurate in some situations and at some times.

 

[Lucre]

This has got to be the best magic forum out there.

 

[Grillo_Parlante]

To continue using our little Troll, taking out all of the edicts would stir this little relationship up a lot. Now maybe the best way to combat the Troll is with token strategies that can chump block. From there, board wipes may become more valueable, and then back to the Troll. The cycle is reborn.

In real life thats actually something that happened. Troll (and lumberknot) found themselves on the short end of more than a few goblin bushwhacker charges. Another cute play against durable threat decks has been chump->sacrifice with tymaret, the murder king to slowly bleed them to death. It turns out having an awesome hexproof dude doesn't do you much good if you are falling behind.