In the beginning...
When I first designed my cube, I had an idea that it would be empirically determined. I used eidolon232's list and information, wtwlf123's list, and then, later, cubetutor stats to try to get the most commonly played 450 cards. I called it the "Blue Chip Cube" after blue chip stocks. Each card was supposed to be a conservative choice in that it was widely playable in as many decks as possible.
However, recently I've done a 180. I now think that while you need a good foundation of widely playable cards, a healthy amount of narrowly playable cards can add a lot of interesting texture. I think it was kojima here who made a comment, like four years ago, about how he or she was bored when they opened a pack and say all the same cards over and over, and how they really enjoyed cubes with many off-beat choices. At the time, I felt like, "Cube is so interesting and deep though! How can you get bored of this!" However, I'm beginning to feel the same way. Drafting a cube where Sulfuric Vortex is good has been done, I'd be excited to draft one where Corpsejack Menace was good.
The problem
The problem, as I see it, is that cube drafts aren't as fun as they could be. Because all the cards are good, you usually end up with more than enough playables. This is good on one hand because, as a designer/curator of a cube, you want your players to be able to play the games. However, I think it makes the drafts less challenging. I've tried a few solutions, like reducing the size of the packs, but that kind of mucks up the draft and the strategy of tabling cards. Official limit formats often have a bunch of blanks in the packs (Abzan Runemark and company), and I don't think this is an accident. While I don't think it's an optimal solution, reducing the viable cards in the draft makes players have to make harder choices. It's not optimal because it also creates other problems, like there just not being enough playable cards in a given color. A much better solution than just straight up bad cards, to me, is having more situational cards. These cards can be quite narrowly playable, serving the dual function of forcing players to be creative in the draft and giving interesting decks support.
The benefit
Although I think this idea could be applied to cubes of any power level or theme, I am going to talk about it in the context of regular unpowered cubes. At least for me, the goal of my cube is to create an environment where players can have fun. I think in terms of Timmy/Tammy, Johnny/Jenny, and Spike; I want each of them to enjoy drafting my cube.
I want Timmy/Tammy to be able to feel powerful. That means there are big creatures. At least one badass dragon (Atarka), one badass angel (Baneslayer), and a lot of individually powerful cards (Jace, Jitte, Wrath of God). I want to have room for stuff like Craterhoof Behemoth, even if it means cutting one of the good, but lesser, efficient cards.
For Johnny/Jenny, drafting a crazy or off-beat deck is super fun. I want Johnnys to be able to try to do all sorts of "achievement unlocked" type decks, like reanimator, or Life from the Loam.
Spike loves a challenge, and making the draft more difficult or nuanced is likely to keep Spikes engaged.
Examples
I recently added the Splinter Twin combo, Griselbrand, and a few other, more speculative cards to the cube. I know these aren't real surprising or weird, but this is a big change in philosophy for my cube and so I'm shifting slowly. Still, Griselbrand was a card I didn't play before because I think that most of the time, he's going to be cheated in somehow, and only a small number of decks can do that.
I still want to play the nostalgic cards that have kept me interested in cube like Sulfuric Vortex, but I don't really feel like I need to play every Searing Spear that's printed. Some of these totally playable, even good, functional reprints are some of the cards I'm looking to cut to try out stuff like Molten Vortex to see if some crazy Life from the Loam deck happens. Maybe some of those cards will be complete misses, but there have been other cards I was skeptical of, like Pox, that ended up being pretty sweet.
Anyway, I'm still working out some of these ideas and just wanted to share my thoughts with everyone here. My list for reference: http://www.cubetutor.com/viewcube/43967
I think an important factor is your draft system. We Rochester full table which means that narrower archetypes are more playable compared to 2 player Winston for example. Adding narrower cards is fine if they get played oftne enough. Otherwise they end up being a trap and might leave players annoyed at having been tricked in drafting unplayable cards. For us this a reason to simulate 8-man draft by playing four players each. 2 player formats end up being dry and too zero- summy, especially if you play the same players all the time.
We do support fringier archetypes like reanimator, Stax, storm, spells matter,burn... But the line between too little and too much support is tricky. Also you have to make sure that the fringe archetype is actually powerful enough to win against 'normal'archetypes ( bread and butter aggro, control, midrange...). Landfall/Titania was a failure for us. It was synergystic as hell, fun to play, but lost too many games agaisnt simpler game plans.
What helps is too get as much dual purpose cards in your cube. Gitaxian Probe for example works in Storm, but also in spells matters and gets played in mono as 23rd cards once in a while. Guttersnipe works wonders in spells matters decks, but also is a boss in pure burn. Tempt with Vengeance works in Token, aggro, ramp and inifinite mana combo-decks.
I generally err on the side of Wide Playability, because it leaves players with the flexibility of doing whatever they want. With too many Narrow (archetype-specific) cards, drafters can get pigeonholed into playing particular archetypes, and it eliminates a lot of the variance between drafts that makes Cube such an attractive format. So I prefer to only play the most critical Narrow Playability cards that are essential for enabling/encouraging specific archetypes, and fleshing out the rest of the cube with cards that have as Wide a playable range as possible.
That said, I prefer a card that has enough flexibility to go into different archetypes. But I'm not willing to sacrifice a slot if an existing card is too weak. For example, I used to have all the guildmages in the Cube. Gradually I began to cut most of them because their abilities are too weak/expensive. But they definitely pass the stats test.
I generally err on the side of Wide Playability, because it leaves players with the flexibility of doing whatever they want. With too many Narrow (archetype-specific) cards, drafters can get pigeonholed into playing particular archetypes, and it eliminates a lot of the variance between drafts that makes Cube such an attractive format. So I prefer to only play the most critical Narrow Playability cards that are essential for enabling/encouraging specific archetypes, and fleshing out the rest of the cube with cards that have as Wide a playable range as possible.
I think my ideal list probably will overlap with yours in 80-90% of the cards. So I'm not talking about drastic changes, more just giving certain archetypes more tools. I think the biggest place I'd want to update things is in the gold sections. I used to want to play powerful, widely playable cards in the gold sections, but it seems like the color pairs fall into fairly well defined archetypes, and I'd rather give just that deck a good tool rather than a card that's merely good in that deck but good elsewhere too.
For example, Kolaghan's Command is a widely playable card, but just good in a RB aggro deck. Tymaret, the Murder King gives RB reach, and plays well with the pox/stax cards in black. So even if Tymaret is the worse card overall, it's better than K Command, or something like Grenzo, Dungeon Warden in the most commonly played RB archetypes. Still experimenting with this, but I feel like I can walk it back if certain cards don't pan out.
I think people underestimate how radically different two cubes can be from one another with only 10% of the cards being different. In a 360 cube, 10% of a section might be 5 cards. If those 5 cards are -5 aggro creatures +5 wraths, the entire color plays completely different than it does in the other list, despite only being 10% off from one another. People talk about how dramatically different powered cubes and unpowered cubes can play, and those differences are way less than 10% of the overall cards.
I get what you're saying in your example, but even if I was playing a dedicated Pox/Stax deck, I'd rather have Kolaghan's Command than Tymaret, the Murder King any day of the week. And by changing Command to Tymaret, you're giving an extra tool to a commonly played deck, but removing a tool that can be playable in all the other ones. It hurts BR midrange, control and all the other archetypes more than it helps BR Pox/Stax ...and I don't think it's particularly close.
So I don't think cutting a universally playable card for an archetype-specific card isn't a good idea unless that narrower card is a "must have" for the way the archetype functions. It needs to be a card that either pulls me into the archetype, is essential to making the archetype function, or is completely broken in a specific deck. If it's a card that gets slightly better in a deck with the right composition of cards, I don't think it's worthwhile. Like you could play Shrapnel Blast instead of Lightning Strike and Thoughtcast instead of Impulse because they're playable in the artifact.dec, but they're not better enough from their competition to make it worth it, and I wouldn't take them unless I was already in their given deck. Unlike something like Tezzeret the Seeker or Tolarian Academy, which are either essential or broken, and make me want to move into the archetype in the first place.
For example, Kolaghan's Command is a widely playable card, but just good in a RB aggro deck.
Sorry to detract from the intent of the OP but my experience is the exact opposite. I like this card better in every other archetype than I do in aggro.
I built on theme first, then diverged from there. For example, I started with "multiplayer artifact cube" and "enchantment cube." mine look different than other people's here. The theme allows for mechanics to be worked into the system that are no longer good in anything but their own block or similarly designed block structure.
Private Mod Note
():
Rollback Post to RevisionRollBack
Life is a beautiful engineer, yet a brutal scientist.
I think people underestimate how radically different two cubes can be from one another with only 10% of the cards being different. In a 360 cube, 10% of a section might be 5 cards. If those 5 cards are -5 aggro creatures +5 wraths, the entire color plays completely different than it does in the other list, despite only being 10% off from one another. People talk about how dramatically different powered cubes and unpowered cubes can play, and those differences are way less than 10% of the overall cards.
I get what you're saying in your example, but even if I was playing a dedicated Pox/Stax deck, I'd rather have Kolaghan's Command than Tymaret, the Murder King any day of the week. And by changing Command to Tymaret, you're giving an extra tool to a commonly played deck, but removing a tool that can be playable in all the other ones. It hurts BR midrange, control and all the other archetypes more than it helps BR Pox/Stax ...and I don't think it's particularly close.
So I don't think cutting a universally playable card for an archetype-specific card isn't a good idea unless that narrower card is a "must have" for the way the archetype functions. It needs to be a card that either pulls me into the archetype, is essential to making the archetype function, or is completely broken in a specific deck. If it's a card that gets slightly better in a deck with the right composition of cards, I don't think it's worthwhile. Like you could play Shrapnel Blast instead of Lightning Strike and Thoughtcast instead of Impulse because they're playable in the artifact.dec, but they're not better enough from their competition to make it worth it, and I wouldn't take them unless I was already in their given deck. Unlike something like Tezzeret the Seeker or Tolarian Academy, which are either essential or broken, and make me want to move into the archetype in the first place.
First off, I think there's merit to both approaches. Every cube has a mix of both narrow, archetype-specific cards and wider, more general cards. Your comment here helped me realize that I think multicolor and monocolor cards work differently, or I think about them differently. The downside to picking a multicolor card is big enough that I almost never will take one unless I am in those colors already. In other words, they work sort of like Thoughtcast in that I am only picking them if I am in a relatively narrow deck. In addition, I think it depends on what decks actually show up in the draft. If RB midrange is commonly played, maybe it doesn't make sense to cut K Command for Tymaret, or maybe it makes sense to rotate them in and out every few drafts. Control rarely plays RB together, for example, so I'm not too worried about hurting potential RB/x Control decks because they don't show up anyway.
I wish all the multicolor cards were bombs in their color combos. Totally playable role-players like K Command are perfectly good, but boring. I'd rather the multicolor cards help define what the archetype is doing. Another example: Mystic Snake is certainly a good UG card, but UG is almost always a ramp deck that is tapping out each turn for me since I cut green aggro. In that deck, Mystic Snake is fine, but there are many mono-G or mono-U cards I'd rather have. In other words, there is little to no payoff for the cost of being multicolor.
I'm certainly not cutting something like Ponder for Thoughtcast, but multicolor cards that are just meh at best and traps at worst aren't too crucial.
For example, Kolaghan's Command is a widely playable card, but just good in a RB aggro deck.
Sorry to detract from the intent of the OP but my experience is the exact opposite. I like this card better in every other archetype than I do in aggro.
I agree. I meant, it is merely good, not excellent, in aggro.
I think multicolor cards are narrow enough without limiting them to a specific deck on top of their other restrictions.
For example, your quote here:
"I meant, it is merely good, not excellent, in aggro."
Change that to "I meant, it is merely good, not excellent, in its respective archetype..."
It's twice as important when you're talking about an archetype specific card. Unless a card is godly or critical in a specific deck, I think cutting a universally playable card for it is a mistake. Especially in the multicolor section.
Also, I disagree about Mystic Snake. I guess it's too bad that all of your Simic decks are just ramp decks. I think it makes the drafts a lot less interesting when certain guilds are always doing only one thing. Part of the reason why universally playable cards are so important, IMO. Allows decks trying to do different things with enough tools to do them. One of the things I like the most about the metagame in my cube is that I can play aggro/tempo, midrange or control in any color combination in any given draft. If that wasn't the case, I think the meta would get really stale.
I think multicolor cards are narrow enough without limiting them to a specific deck on top of their other restrictions.
For example, your quote here:
"I meant, it is merely good, not excellent, in aggro."
Change that to "I meant, it is merely good, not excellent, in its respective archetype..."
It's twice as important when you're talking about an archetype specific card. Unless a card is godly or critical in a specific deck, I think cutting a universally playable card for it is a mistake. Especially in the multicolor section.
Also, I disagree about Mystic Snake. I guess it's too bad that all of your Simic decks are just ramp decks. I think it makes the drafts a lot less interesting when certain guilds are always doing only one thing. Part of the reason why universally playable cards are so important, IMO. Allows decks trying to do different things with enough tools to do them. One of the things I like the most about the metagame in my cube is that I can play aggro/tempo, midrange or control in any color combination in any given draft. If that wasn't the case, I think the meta would get really stale.
Fair enough. Obviously you are a respected figure here because a lot of people agree with you. I think that the critical point of disagreement is just how diverse the decks are within each color section. If multiple archetypes see play within a color pair, then wider playability is best. If only one or two archetypes see play then I think narrow cards are better because there aren't decks being played to justify having wider playability. However, even when the decks aren't that diverse within a color section, widely playable cards can still be played.
So, it would be logical to conclude that since widely playable cards are good if decks in a given color pair are diverse or not, it would be best to have widely playable cards no matter what. I can see the logic in that. Maybe my experiment will fail. The big advantage of narrowly playable cards is that they create good signals, especially for less experienced cubers. Like in recent draft formats like Origins, the multicolor cards can help signal what a color pair should be doing.
I also don't really get what you're saying here. Can you explain this idea a bit more?
For example, your quote here:
"I meant, it is merely good, not excellent, in aggro."
Change that to "I meant, it is merely good, not excellent, in its respective archetype..."
It's twice as important when you're talking about an archetype specific card. Unless a card is godly or critical in a specific deck, I think cutting a universally playable card for it is a mistake. Especially in the multicolor section.
In order for a card to justify only being valuable to one deck, it needs to be critical to the way that deck functions or completely busted in that deck. Otherwise, you're much better off playing a card that can contribute to the success of multiple different decks.
Your experiment won't be a failure if your goal is to have players consistently within the same decks. If you want players in the same 8 decks every time you draft, including the more narrow archetype cards will make each individual deck stronger and more consistent. But if you want players to draft a variety of different decks, you want to limit the number of archetype-specific cards so that players aren't pigeonholed into playing only one deck with the pile of cards they wound up with.
When I first designed my cube, I had an idea that it would be empirically determined. I used eidolon232's list and information, wtwlf123's list, and then, later, cubetutor stats to try to get the most commonly played 450 cards. I called it the "Blue Chip Cube" after blue chip stocks. Each card was supposed to be a conservative choice in that it was widely playable in as many decks as possible.
However, recently I've done a 180. I now think that while you need a good foundation of widely playable cards, a healthy amount of narrowly playable cards can add a lot of interesting texture. I think it was kojima here who made a comment, like four years ago, about how he or she was bored when they opened a pack and say all the same cards over and over, and how they really enjoyed cubes with many off-beat choices. At the time, I felt like, "Cube is so interesting and deep though! How can you get bored of this!" However, I'm beginning to feel the same way. Drafting a cube where Sulfuric Vortex is good has been done, I'd be excited to draft one where Corpsejack Menace was good.
The problem
The problem, as I see it, is that cube drafts aren't as fun as they could be. Because all the cards are good, you usually end up with more than enough playables. This is good on one hand because, as a designer/curator of a cube, you want your players to be able to play the games. However, I think it makes the drafts less challenging. I've tried a few solutions, like reducing the size of the packs, but that kind of mucks up the draft and the strategy of tabling cards. Official limit formats often have a bunch of blanks in the packs (Abzan Runemark and company), and I don't think this is an accident. While I don't think it's an optimal solution, reducing the viable cards in the draft makes players have to make harder choices. It's not optimal because it also creates other problems, like there just not being enough playable cards in a given color. A much better solution than just straight up bad cards, to me, is having more situational cards. These cards can be quite narrowly playable, serving the dual function of forcing players to be creative in the draft and giving interesting decks support.
The benefit
Although I think this idea could be applied to cubes of any power level or theme, I am going to talk about it in the context of regular unpowered cubes. At least for me, the goal of my cube is to create an environment where players can have fun. I think in terms of Timmy/Tammy, Johnny/Jenny, and Spike; I want each of them to enjoy drafting my cube.
I want Timmy/Tammy to be able to feel powerful. That means there are big creatures. At least one badass dragon (Atarka), one badass angel (Baneslayer), and a lot of individually powerful cards (Jace, Jitte, Wrath of God). I want to have room for stuff like Craterhoof Behemoth, even if it means cutting one of the good, but lesser, efficient cards.
For Johnny/Jenny, drafting a crazy or off-beat deck is super fun. I want Johnnys to be able to try to do all sorts of "achievement unlocked" type decks, like reanimator, or Life from the Loam.
Spike loves a challenge, and making the draft more difficult or nuanced is likely to keep Spikes engaged.
Examples
I recently added the Splinter Twin combo, Griselbrand, and a few other, more speculative cards to the cube. I know these aren't real surprising or weird, but this is a big change in philosophy for my cube and so I'm shifting slowly. Still, Griselbrand was a card I didn't play before because I think that most of the time, he's going to be cheated in somehow, and only a small number of decks can do that.
I still want to play the nostalgic cards that have kept me interested in cube like Sulfuric Vortex, but I don't really feel like I need to play every Searing Spear that's printed. Some of these totally playable, even good, functional reprints are some of the cards I'm looking to cut to try out stuff like Molten Vortex to see if some crazy Life from the Loam deck happens. Maybe some of those cards will be complete misses, but there have been other cards I was skeptical of, like Pox, that ended up being pretty sweet.
Anyway, I'm still working out some of these ideas and just wanted to share my thoughts with everyone here. My list for reference: http://www.cubetutor.com/viewcube/43967
We do support fringier archetypes like reanimator, Stax, storm, spells matter,burn... But the line between too little and too much support is tricky. Also you have to make sure that the fringe archetype is actually powerful enough to win against 'normal'archetypes ( bread and butter aggro, control, midrange...). Landfall/Titania was a failure for us. It was synergystic as hell, fun to play, but lost too many games agaisnt simpler game plans.
What helps is too get as much dual purpose cards in your cube. Gitaxian Probe for example works in Storm, but also in spells matters and gets played in mono as 23rd cards once in a while. Guttersnipe works wonders in spells matters decks, but also is a boss in pure burn. Tempt with Vengeance works in Token, aggro, ramp and inifinite mana combo-decks.
I feel compelled to repeat everything I hear
My 630 Card Powered Cube
My Article - "Cube Design Philosophy"
My Article - "Mana Short: A study in limited resource management."
My 50th Set (P)review - Discusses my top 20 Cube cards from OTJ!
That said, I prefer a card that has enough flexibility to go into different archetypes. But I'm not willing to sacrifice a slot if an existing card is too weak. For example, I used to have all the guildmages in the Cube. Gradually I began to cut most of them because their abilities are too weak/expensive. But they definitely pass the stats test.
UR Melek, Izzet ParagonUR, B Shirei, Shizo's CaretakerB, R Jaya Ballard, Task MageR,RW Tajic, Blade of the LegionRW, UB Lazav, Dimir MastermindUB, UB Circu, Dimir LobotomistUB, RWU Zedruu the GreatheartedRWU, GUBThe MimeoplasmGUB, UGExperiment Kraj UG, WDarien, King of KjeldorW, BMarrow-GnawerB, WBGKarador, Ghost ChieftainWBG, UTeferi, Temporal ArchmageU, GWUDerevi, Empyrial TacticianGWU, RDaretti, Scrap SavantR, UTalrand, Sky SummonerU, GEzuri, Renegade LeaderG, WUBRGReaper KingWUBRG, RGXenagos, God of RevelsRG, CKozilek, Butcher of TruthC, WUBRGGeneral TazriWUBRG, GTitania, Protector of ArgothG
I think my ideal list probably will overlap with yours in 80-90% of the cards. So I'm not talking about drastic changes, more just giving certain archetypes more tools. I think the biggest place I'd want to update things is in the gold sections. I used to want to play powerful, widely playable cards in the gold sections, but it seems like the color pairs fall into fairly well defined archetypes, and I'd rather give just that deck a good tool rather than a card that's merely good in that deck but good elsewhere too.
For example, Kolaghan's Command is a widely playable card, but just good in a RB aggro deck. Tymaret, the Murder King gives RB reach, and plays well with the pox/stax cards in black. So even if Tymaret is the worse card overall, it's better than K Command, or something like Grenzo, Dungeon Warden in the most commonly played RB archetypes. Still experimenting with this, but I feel like I can walk it back if certain cards don't pan out.
I get what you're saying in your example, but even if I was playing a dedicated Pox/Stax deck, I'd rather have Kolaghan's Command than Tymaret, the Murder King any day of the week. And by changing Command to Tymaret, you're giving an extra tool to a commonly played deck, but removing a tool that can be playable in all the other ones. It hurts BR midrange, control and all the other archetypes more than it helps BR Pox/Stax ...and I don't think it's particularly close.
So I don't think cutting a universally playable card for an archetype-specific card isn't a good idea unless that narrower card is a "must have" for the way the archetype functions. It needs to be a card that either pulls me into the archetype, is essential to making the archetype function, or is completely broken in a specific deck. If it's a card that gets slightly better in a deck with the right composition of cards, I don't think it's worthwhile. Like you could play Shrapnel Blast instead of Lightning Strike and Thoughtcast instead of Impulse because they're playable in the artifact.dec, but they're not better enough from their competition to make it worth it, and I wouldn't take them unless I was already in their given deck. Unlike something like Tezzeret the Seeker or Tolarian Academy, which are either essential or broken, and make me want to move into the archetype in the first place.
My 630 Card Powered Cube
My Article - "Cube Design Philosophy"
My Article - "Mana Short: A study in limited resource management."
My 50th Set (P)review - Discusses my top 20 Cube cards from OTJ!
Sorry to detract from the intent of the OP but my experience is the exact opposite. I like this card better in every other archetype than I do in aggro.
Modern
Commander
Cube
<a href="http://www.mtgsalvation.com/forums/the-game/the-cube-forum/cube-lists/588020-unpowered-themed-enchantment-an-enchanted-evening">An Enchanted Evening Cube </a>
First off, I think there's merit to both approaches. Every cube has a mix of both narrow, archetype-specific cards and wider, more general cards. Your comment here helped me realize that I think multicolor and monocolor cards work differently, or I think about them differently. The downside to picking a multicolor card is big enough that I almost never will take one unless I am in those colors already. In other words, they work sort of like Thoughtcast in that I am only picking them if I am in a relatively narrow deck. In addition, I think it depends on what decks actually show up in the draft. If RB midrange is commonly played, maybe it doesn't make sense to cut K Command for Tymaret, or maybe it makes sense to rotate them in and out every few drafts. Control rarely plays RB together, for example, so I'm not too worried about hurting potential RB/x Control decks because they don't show up anyway.
I wish all the multicolor cards were bombs in their color combos. Totally playable role-players like K Command are perfectly good, but boring. I'd rather the multicolor cards help define what the archetype is doing. Another example: Mystic Snake is certainly a good UG card, but UG is almost always a ramp deck that is tapping out each turn for me since I cut green aggro. In that deck, Mystic Snake is fine, but there are many mono-G or mono-U cards I'd rather have. In other words, there is little to no payoff for the cost of being multicolor.
I'm certainly not cutting something like Ponder for Thoughtcast, but multicolor cards that are just meh at best and traps at worst aren't too crucial.
I agree. I meant, it is merely good, not excellent, in aggro.
For example, your quote here:
"I meant, it is merely good, not excellent, in aggro."
Change that to "I meant, it is merely good, not excellent, in its respective archetype..."
It's twice as important when you're talking about an archetype specific card. Unless a card is godly or critical in a specific deck, I think cutting a universally playable card for it is a mistake. Especially in the multicolor section.
Also, I disagree about Mystic Snake. I guess it's too bad that all of your Simic decks are just ramp decks. I think it makes the drafts a lot less interesting when certain guilds are always doing only one thing. Part of the reason why universally playable cards are so important, IMO. Allows decks trying to do different things with enough tools to do them. One of the things I like the most about the metagame in my cube is that I can play aggro/tempo, midrange or control in any color combination in any given draft. If that wasn't the case, I think the meta would get really stale.
My 630 Card Powered Cube
My Article - "Cube Design Philosophy"
My Article - "Mana Short: A study in limited resource management."
My 50th Set (P)review - Discusses my top 20 Cube cards from OTJ!
Fair enough. Obviously you are a respected figure here because a lot of people agree with you. I think that the critical point of disagreement is just how diverse the decks are within each color section. If multiple archetypes see play within a color pair, then wider playability is best. If only one or two archetypes see play then I think narrow cards are better because there aren't decks being played to justify having wider playability. However, even when the decks aren't that diverse within a color section, widely playable cards can still be played.
So, it would be logical to conclude that since widely playable cards are good if decks in a given color pair are diverse or not, it would be best to have widely playable cards no matter what. I can see the logic in that. Maybe my experiment will fail. The big advantage of narrowly playable cards is that they create good signals, especially for less experienced cubers. Like in recent draft formats like Origins, the multicolor cards can help signal what a color pair should be doing.
I also don't really get what you're saying here. Can you explain this idea a bit more?
Your experiment won't be a failure if your goal is to have players consistently within the same decks. If you want players in the same 8 decks every time you draft, including the more narrow archetype cards will make each individual deck stronger and more consistent. But if you want players to draft a variety of different decks, you want to limit the number of archetype-specific cards so that players aren't pigeonholed into playing only one deck with the pile of cards they wound up with.
My 630 Card Powered Cube
My Article - "Cube Design Philosophy"
My Article - "Mana Short: A study in limited resource management."
My 50th Set (P)review - Discusses my top 20 Cube cards from OTJ!