General [Design Discussion] Perfect Imbalance

[Grillo_Parlante]

The problem is the notion of metagames, of reacting to metagames, of treating the format as a metagame rather than a solveable game. It's absolutely true that people's evaluations of cards will change as they get used to them within the format, that their evaluations will grow nuanced as they build different decks and see them play different roles. But this is not a metagame thing. It's not a metagame reaction. It's a learning process. The concept of metagames they're talking about is not a sound concept. The models they draw their ideas from - LoL, Magic - are mostly kept interesting by injections of new stuff, by changes. When we talk about metagames, we draw away from the real models we actually have which give us some way to think about and talk about depth and unsolveability and all that kind of thing.

Thats great, except FSR already deconstructed the video in a way that was helpful many pages ago. I kind of want to move on at this point.

If you dislike the semantics thats fine, but please give me a sound term that I can use that communicates effectively the way people have been reacting to my environment: "learning process" is far to nebulous. Right now they are responding strategically to the way people have been drafting.

 

[Ferritovore]

Sorry everyone, I'm locking this thread now.

 

[Grillo_Parlante]

It does feel like its gotten to that point /sigh

 

[Suicufnoc]

Honesty I think learning process is pretty acurate wording. Solving process? That's what they're doing. Figuring out what the best strategy is. I'm not sure that makes it improper to call it a metagame, though.

But, what do you want to move on to? I don't think anybody's off topic, they just don't agree. I don't really think there's much else to discuss, unless someone has a groundbreaking idea.

 

[Grillo_Parlante]

I would like to not have a page of arguments over the correct usage of the term metagame, or whether the extra credits crew consists of idiots or not.

 

[Quirk]

If you dislike the semantics thats fine, but please give me a sound term that I can use that communicates effectively the way people have been reacting to my environment: "learning process" is far to nebulous. Right now they are responding strategically to the way people have been drafting.

Well, it's often easy to express the empirical details of what's happening, and maybe separate them from the proposed cause. People are playing bouncelands less - presumably because they've seen what happens when they get destroyed. People are playing edicts again because some decks give good value from them. Stating observed trends and suggesting causes doesn't have the same room to go wrong. Once you have a model, such as "metagame fluctuations", it becomes something to view the world through - but that can be a trap if the model doesn't fit, or close your eyes to ideas that conflict with that model. Really, your players are learning your format. It seems like the way they're playing your format is still evolving. That's good. It suggests you've made a deep format. I've tried some drafts of the Penny Pincher Cube. I like it, I think it looks like a deep format.

 

[Grillo_Parlante]

Thank you, thats very helpful.

I'm having a hard time, however, conceptualizing that depth in a way that I can port it over to other types of cubes, which is why I made the thread.

Do you have any thoughts on designing a format so it includes "metagame fluctuations"? Especially from the early stages, before you have observed trends.

 

[quanta]

You can make a format deeper, harder to solve, etc, but that does not get you away from player psychological profiles. Your players are going to roam across a format like a hill-climbing optimisation algorithm. Depending on their psychology, they may well get stuck on local maxima. You may well have to change things to shake them off it.

Basically, my frustration is that the terms being used are still terms from the videos which are not simplified, but flat wrong in almost every respect. We end up discussing real phenomena but the model is nonsense.
e.g.

Idea that metagames needs something to react to and counter-act to was introduced using a simple video and example.

This idea is bad and wrong and it needs to die.

Yeah, but your players hill climb by reacting to each other. Then they eventually they probably hit some sort of dead end, but maybe not. Maybe their strategies sort of wander around randomly because people are frankly terrible at implementing mixed strategies; even moreso when they aren't playing rock paper scissors and are unsure how good each strategy is vs. each other. The original model is mathematical nonsense, but that is a boring discussion to have.

Although one thing we need to do to make good draft formats is make sure no one strategy is the nuke to rock, paper, and scissors and part of how we manage this is adding complexity, it's probably worth remembering that wizards found that you can easily overload most humans with bad complexity so maybe it'd be just as useful to go over the other side of the coin and try to figure out seemingly obvious solutions that are actually bad.

 

[Quirk]

I think there's a lot of mileage to be had out of discussing what makes formats hard to solve - and what makes them enjoyable to play, and whether those are the same things. In theory, Cube should be starting at a huge advantage over other formats here, with so many highly interactive cards, but often the cards people most want to play are among the most broken things in the format. I think most Cubes can boast interesting interactions between groups of cards, but the questions I'm most interested in are how to keep players excited without dropping too much power in their laps, how to trade off the power of synergy with the ability to fill weaknesses... how to keep the spectrum of interesting decks that can be built as wide as it possibly can be, and how to keep them viable - and whether it's maybe even okay to have some decks that aren't very viable but that people like to play.

 

[Quirk]

it's probably worth remembering that wizards found that you can easily overload most humans with bad complexity so maybe it'd be just as useful to go over the other side of the coin and try to figure out seemingly obvious solutions that are actually bad.

I like this. Not unplayably bad, presumably, but something that looks first-pickable that you should be taking 6th? Players do come in with preconceptions of how good cards were in their original formats, and Wizards every so often does a sneaky reprint of a past great which isn't so great in the new format any more...

 

[Quirk]

Do you have any thoughts on designing a format so it includes "metagame fluctuations"? Especially from the early stages, before you have observed trends.

Well, one thing I think the Penny Pincher Cube is good at is providing a wide variety of cards which are hard to evaluate without knowing what else is in the format. A really obvious example is Trinket Mage, whose value fluctuates with the one-mana artifacts available, but cards with Heroic and Outlast are also hard to evaluate without understanding how many ways there are to trigger Heroic, how many creatures can be expected to have +1/+1 counters... It's fairly easy to evaluate the rough power level of an Elspeth or a Jace, but you can't work out how strong a theme is until you've had the chance to play with it a few times. As different players get different feels for the themes, they'll bring stronger implementations of these themes to the table, and themes that looked good early on may diminish in importance, and other themes come to prominence.

 

[quanta]

I like this. Not unplayably bad, presumably, but something that looks first-pickable that you should be taking 6th? Players do come in with preconceptions of how good cards were in their original formats, and Wizards every so often does a sneaky reprint of a past great which isn't so great in the new format any more...

Reminds me of something MaRo said how about in Scars of Mirrodin block, go for the throat is way weaker than it would have been in almost any other draft format and shatter is way better. So although sometimes old cards just fill holes in a nice way, it's even better if they interact with the rest of the format in a way that requires you evaluate them differently. I think this is an interesting line of thought to follow. It might be a good reason to rely more on removal that's relatively contextual which seems to be how wizard's draft formats are nowadays. It's also maybe another reason to favor cards that are synergistically powerful so to speak rather than just straight up powerful. Cards that require synergy will have a more interesting learning curve.

Originally I was going somewhere else with my thought more along the lines of "What ways look like they'd be a good way to introduce complexity and a fun learning curve into a cube, but are actually tedious or frustrating?" We should be able to glean something from past draft formats even though they are rather different from a cube.

 

[Grillo_Parlante]

Reminds me of something MaRo said how about in Scars of Mirrodin block, go for the throat is way weaker than it would have been in almost any other draft format and shatter is way better. So although sometimes old cards just fill holes in a nice way, it's even better if they interact with the rest of the format in a way that requires you evaluate them differently. I think this is an interesting line of thought to follow. It might be a good reason to rely more on removal that's relatively contextual which seems to be how wizard's draft formats are nowadays. It's also maybe another reason to favor cards that are synergistically powerful so to speak rather than just straight up powerful. Cards that require synergy will have a more interesting learning curve.

Originally I was going somewhere else with my thought more along the lines of "What ways look like they'd be a good way to introduce complexity and a fun learning curve into a cube, but are actually tedious or frustrating?" We should be able to glean something from past draft formats even though they are rather different from a cube.

Thats very good, so in terms of threat removal relationships we would be adding depth by making it harder to figure out how threats and removal are supposed to be interacting in the format. I like that encapsulation.

One of the interactions I run that occastionally comes up is



It doesn't have the density as a relationship to impact how people draft, but it is a cool moment when you realize that quirion ranger is a much deeper card than you had imagined. You could have been playing the format for months but still get that feeling of excitement from a new discovery.

 

[Quirk]

Reminds me of something MaRo said how about in Scars of Mirrodin block, go for the throat is way weaker than it would have been in almost any other draft format and shatter is way better. So although sometimes old cards just fill holes in a nice way, it's even better if they interact with the rest of the format in a way that requires you evaluate them differently. I think this is an interesting line of thought to follow. It might be a good reason to rely more on removal that's relatively contextual which seems to be how wizard's draft formats are nowadays. It's also maybe another reason to favor cards that are synergistically powerful so to speak rather than just straight up powerful. Cards that require synergy will have a more interesting learning curve.

I think there's a lot to be said for contextual removal. Most of us have more creatures in our cube than any other card type, and getting to a more nuanced understanding than "does this give me value as soon as it hits the board?" is I think a possible key to additional format longevity.

Originally I was going somewhere else with my thought more along the lines of "What ways look like they'd be a good way to introduce complexity and a fun learning curve into a cube, but are actually tedious or frustrating?" We should be able to glean something from past draft formats even though they are rather different from a cube.

Might depend on your players, but I seem to recall Odyssey block with Madness/Threshold/Flashback split player opinion quite sharply between those who loved the board complexity and those who were overwhelmed by it.

 

[Quirk]

It doesn't have the density as a relationship to impact how people draft, but it is a cool moment when you realize that quirion ranger is a much deeper card than you had imagined. You could have been playing the format for months but still get that feeling of excitement from a new discovery.

Quirion Ranger is a really cool card. I've seen it do so much work in Pauper Stompy. It's more interesting than something like Kiora's Follower because sometimes, returning that land is a real trade-off, and sometimes it's basically mana ramp.

 

[Alfonzo Bonzo]

Thank you, thats very helpful.

I'm having a hard time, however, conceptualizing that depth in a way that I can port it over to other types of cubes, which is why I made the thread.

Do you have any thoughts on designing a format so it includes "metagame fluctuations"? Especially from the early stages, before you have observed trends.

Is there perhaps something here about the way you draft your cube? I will almost always grid draft or Rochester/snake draft. If all picks are public information and if we refer to the contextual creature removal examples above: if your opponent drafts some bomby hexproof creatures you are more likely to prioritise edict/wrath effects?

Otherwise, to 'solve' the meta game you would need regular drafts with predominantly the same group of players without substantial changes to the card pool on a regular basis.

I have to admit some of this conversation has gone over my head, and there are clearly some tensions in this thread. The thing I love about this forum is how we generally are open to new and different ideas even when they are most likely terrible (in my case) and the spirit of at least trying things.

 

[Eric Chan]

Really late to the party here, and afraid I don't have anything useful to contribute, but I do want to put on my mod hat for a second. I'm alarmed at some of the mean-spiritedness in this thread - from one poster in particular - as well as the level of pedantry on display when it's clear that most folks aren't too interested in a discussion on semantics. If you're posting on Riptide Lab, we expect you to keep conversations productive without resorting to borderline personal attacks, and negativity in your tone. I'm not going to mod any posts that have been made thus far, but I will starting from here on out. Basically, watch your tone - I really shouldn't have to tell anyone this, especially not in such a public fashion, but given the contents of the last two pages of this thread, I feel I need to.

Anyways - now back to your regular scheduled programming.

 

[Quirk]

I've slept, I'm cooler-headed, and I would like to apologise for some of the more frustrated stuff I was typing last night, and try to lay out clearly the argument I've been developing and which I think Grillo and I now have some level of mutual understanding of despite my perhaps poor attempts to communicate it, and show that it really isn't about semantics.

What we all want is a format that's got a lot of "play" to it - a format that's hard to solve, in which it's hard to figure out the best decks and the best ways to build them.

We started from a place that talked about deliberately imbalancing game elements to create cycles in the metagame.

1) We rejected cycles pretty quickly.
2) "Imbalancing" things does not, according to game theory, necessarily make things any harder to solve. It just changes what the solutions are.
3) The "metagame", the game outside the game, where players try to gain advantages by determining what others will play, becomes strongly relevant only when the game itself is largely solved. From the beginning, we minimax: we don't draft narrow hate cards for decks people might play when we could be picking Jace, the Mind Sculptor instead which is good against everything. We only start to care about what other people are specifically playing when we can no longer get an edge by understanding which cards will generally perform best better than they do.

So, from here, there isn't really a fruitful place to go. It doesn't give us any leads on making a format deeper. And we returned here a number of times, and the conversation grew heated. I felt that I'd laid this out clearly enough that we should be moving on. Obviously, I was wrong, and a number of people thought that this was about semantic differences, and it didn't help that when I wrote a long post trying to explain, in particular, 2), that I got a sharp reply from Grillo confusing it with 1) and stating that I was arguing non-issues; I got quite sharp myself at that point.

However, there is quite a bit of stuff in game theory that does give us leads on how to make a format deeper. Basically, we want to make decisions more complex, and that means making game elements harder to evaluate. It's not about whether an answer to the threat exists in the format, it's about what proportion of the answers handle it. It's about making card evaluation hinge on understanding the rest of the format: if a card is good in a particular game state, how often does that game state arise?

 

[James Stevenson]

Every time anyone says "tempo" they should donate a ticket to the modo cube.

I'm adding "metagame" to the swear jar.

 

[FlowerSunRain]

Otherwise, to 'solve' the meta game you would need regular drafts with predominantly the same group of players without substantial changes to the card pool on a regular basis.

The irony of this whole conversation is that so much of the typical booster draft process depends on blind decision making and a lack of control as created by the splitting of the packs that much of the fine tuning we do on both the design side and also on the player's side is probably irrelevant. Combine that with multi-player chaos (amplified again by blind decision making) and the effects of personal skill in actually playing Magic, including how biases in that skill level may prejudice drafting towards a particular deck a given player is better at, and I would venture that the rough edges are largely meaningless compared to rougher edges of reality.

 

[Grillo_Parlante]

Is there perhaps something here about the way you draft your cube? I will almost always grid draft or Rochester/snake draft. If all picks are public information and if we refer to the contextual creature removal examples above: if your opponent drafts some bomby hexproof creatures you are more likely to prioritise edict/wrath effects?

Otherwise, to 'solve' the meta game you would need regular drafts with predominantly the same group of players without substantial changes to the card pool on a regular basis.

I have to admit some of this conversation has gone over my head, and there are clearly some tensions in this thread. The thing I love about this forum is how we generally are open to new and different ideas even when they are most likely terrible (in my case) and the spirit of at least trying things.

Our drafts are pretty regular and we have the same core of people, so pretty consistent. We do normal drafts of three, 15 card packs. That cube has been pretty heavily drafted between my drafts, Modins, and all of the "featured cube" drafts.

The issue I'll traditionally run into isn't so much anyone really "solving" the format as it is the format stagnating at a certain strategic point.

 

[ahadabans]

It's about making card evaluation hinge on understanding the rest of the format: if a card is good in a particular game state, how often does that game state arise?

What a great post. This statement in particular resonates with me, along with what FSR said:

The irony of this whole conversation is that so much of the typical booster draft process depends on blind decision making and a lack of control as created by the splitting of the packs that much of the fine tuning we do on both the design side and also on the player's side is probably irrelevant. Combine that with multi-player chaos (amplified again by blind decision making) and the effects of personal skill in actually playing Magic, including how biases in that skill level may prejudice drafting towards a particular deck a given player is better at, and I would venture that the rough edges are largely meaningless compared to rougher edges of reality.

I'm going to again bring up a point I made earlier because I believe it's critical we don't lose sight of it. At what point are we over-engineering things and ultimately making our format less deep by doing so? Consider how too much focus on synergistic effects can lead to just running too many conditionally playable cards, and unless you have them in the right combination (and sequence them appropriately in a game), you wind up with just a pile of unplayable garbage.

An example I think is Trinket Mage. Now, that is a card that really excites me because it's a tutor that supports a specific archetype I want to push (artifacts). But consider how narrow it actually is. How many 1 mana artifacts can I run in my cube even if I maxed out, and how many of them would be worth a slot in decks? Without fast mana, moxen, black lotus, etc. what exactly am I fetching? There are still some solid targets, to be sure but nothing truly unfair. So aren't I just better off running Vendilion Clique or something similar and not bothering with Trinket Mage? Now, with a lower power level this question is a lot less black and white. I get that. But at the same time, notice how many choices you remove by trading a card like Clique with Trinket Mage. So some extent, a lot of what we are doing is similar to this where we are trading more dynamic universally good cards for ones which are only good in certain situations or builds. Is that really increasing the depth of our meta?

I don't want to come off as fully supporting power max cube design, but I do believe there is something positive to take away from it. Specifically, that most powerful cards by design already possess deep game play potential. You don't really have to do much (just add water). So while we might dismiss the idea of just running the best cards available as a shallow way of building a cube, it can (and very often does) lead to a deep meta at by itself. And if you don't believe that, just ask yourself how cubing began. Were people trying to craft a specific meta? No. They just threw together a bunch of powerful cards and even doing that haphazardly started a whole new format that has lasted over a decade. My first cube was a mess. It also was the most fun I have ever had playing Magic. I'm only still playing this game because I found cube. That's a stone cold fact.

While I certainly like adding my own signature to my cube with some unique and maybe questionable card choices (Haakon for example) - and discussions like this are awesome, very interesting and I don't mean to be Debby downer - I also don't want to get too far away from what got me into cubing to begin with (which really was just running some powerful cards and seeing what people did with them). Again, the best things that have happened in cube are the ones I didn't plan for.

I don't mean to be dismissive of the discussion here, and rereading my post it comes off that way. I just want to try and encourage us to stay somewhere in the middle with these design ideas. Some level of manipulation is healthy. Going back to the previous edict example. If you are going to run hexproof creatures, you need ways to get rid of them. If that is a strongly supported mechanic, you probably need a higher than average number of edicts. This is a solid line of thinking in my mind. Tokens is another tricky one as any kind of token support (blink/anthems) really increases the need for ways to clear the board (via sweepers or whatever). Beyond these basic types of decisions though, I feel like getting too focused on "chains", "themes", "archetype support" - whatever we want to call this metagame manipulation we are doing - is not going to achieve what we want for a variety of reasons.

 

[japahn]

What a great post. This statement in particular resonates with me, along with what FSR said:

I don't want to come off as fully supporting power max cube design, but I do believe there is something positive to take away from it. Specifically, that most powerful cards by design already possess deep game play potential. You don't really have to do much (just add water). So while we might dismiss the idea of just running the best cards available as a shallow way of building a cube, it can (and very often does) lead to a deep meta at by itself. And if you don't believe that, just ask yourself how cubing began. Were people trying to craft a specific meta? No. They just threw together a bunch of powerful cards and even doing that haphazardly started a whole new format that has lasted over a decade. My first cube was a mess. It also was the most fun I have ever had playing Magic. I'm only still playing this game because I found cube. That's a stone cold fact.

Seconded, including the bold part.

The powermax cubes of 2008-09 had way more situationally good cards than the ones today. Wizards printed a lot of cards that have no weaknesses, no setup cost since then, which led to a stale midrangey meta in powermax cubes. Until then, cube had a lot of the dynamics we're describing in this thread. They happened naturally because the busted stuff tended to be not obviously good, all-around cards, but situational ones that played better than they read. Now enter Elspeth, Jace, Baneslayer Angel, Grave Titan, the SOM Swords - their power level is too high and consistent.

Someone (on this forum I think?) said that MtG is a game about solving and adapting to corner cases. Conditional threats/answers are better than generic ones.

 

[Grillo_Parlante]

In a lot of RT formats, trinket mage is pretty bad, because it just doesn't have a wealth of targets to hit. At lower power level, it has a ton of great targets, and at the higher power level (I would assume) it also has great targets (lotus etc.). You are better off running vend. clique in that scenario, because trinket mage isn't supported: the depth that it promises can never actually be realized.

With removal you are just making the card evaluations somewhat non-obvious: how deep you want to go with that is going to be up to the designer; but there certainly is an important side discussion as to at what point over-complexity hurts a draft format.

 

[ahadabans]

In a lot of RT formats, trinket mage is pretty bad, because it just doesn't have a wealth of targets to hit. At lower power level, it has a ton of great targets, and at the higher power level (I would assume) it also has great targets (lotus etc.). You are better off running vend. clique in that scenario, because trinket mage isn't supported: the depth that it promises can never actually be realized.

I wasn't suggesting that you couldn't make Trinket Mage worth it. Simply that it's a step down in power and (at least in this case) a step down in depth. Clique can disrupt your opponent, it can't shuffle away a card you don't need. And it beats down effectively in any kind of Ux deck. Trinket Mage is just a gray ogre that tutors an artifact. You can't run it in nearly as many decks and so the depth it adds to any meta is less regardless of how much you design around it.

This is somewhat an unfair example since Clique is probably the best designed Magic card in history. Trinket Mage is cool and I think can add to an environment. Don't misunderstand what I'm saying here. I want to specifically call out the difference between dynamic power cards that add to a meta (Clique) versus ones that don't (Jitte). Making your cube compatible with the first kind (and including as many as you can) will only do good things to how your cube plays IMO, and maybe more so than any kind of meta manipulation.