I guess I'll create discussion questions like this, grouping questions of the same type together but not having all ten questions in the same group because that's just nonsense.

Existing keyword into evergreen: Bloodthirst/Raid

Both of them teach good play habits and slightly change the style of play of both players by existing. With Bloodthirst, players are more inclined to block, while with Raid, players are more inclined to attack. Ideally I'd have another fusion of the two words into something like Bloodraid (If a creature you control attacked this turn, this enters the battlefield with a +1/+1 counter.) or something.

Existing evergreen keyword into not: Hexproof

Unlike other keywords, hexproof doesn't change how a player controlling a hexproof creature behaves in regard to that creature. Flying and Vigilance all encourage the player to make safer attacks, because there's less risk in attacking since you have less blockers to worry about/no swingback to worry about. Hexproof doesn't add any risk value to the creature (the same way hexproof until tapped/untapped does), and overall changes the play patterns of the opponent, not themselves. New players learn nothing when playing a hexproof creature.

Received most poorly but deserves a second chance: Affinity

People are too scared of affinity. The only issue with affinity is that it has a huge design space and it was mostly seen on colorless creatures that could eventually be cast for free if you had enough. If it was seen on more creatures like Qumulox, or didn't have "affinity for ARTIFACTS" and instead something else like "Affinity for Goblins", I'm sure it would be better received.

I opted for Unearth to be elevated to Evergreen. I was rambling, but basically I think Unearth has plenty of design space, and (barring the creature being flickered). Since it's a one-time use, Unearth should suggest the creature attacking. I also mentioned it could potentially be used as a shared mechanic for blue and black due to its Grixis history. I noted that white could alternatively be tertiary for unearth instead of red, since white has some background with reanimation.

I chose hexproof as the evergreen mechanic to remove. My response had to do with the creature being uninteractive for the opponent, and specifically mentioned how people found Invisible Stalker donning powerful equipment as a problem (although I left out the creature being unblockable). I added that because the creature is uninteractive for the opponent, there needs to be caution around the mechanic in a way that doesn't affect true evasive creatures.

I'm glad I didn't finish the essays, because I had intended to write about the ability word Radiance. I think radiance has a lot of design space left and could be appear in any color without breaking pie. Radiance could thrive in a multicolor or color-heavy set. I could go on, but I didn't get that far, and Radiance isn't a keyword.

1. An evergreen mechanic is a keyword mechanic that shows up in (almost) every set. If you had to make an existing keyword mechanic evergreen, which one would you choose and why?

I would turn "whenever this creature deals combat damage to a player, draw a card" into an evergreen keyword mechanic. This resolves several positive consequences.

First, this mechanic already appears in almost every set, except not as an official keyword. Turning it into a keyword will make it easy to identify for both designers and players.

It will become a mechanic that players will value and like. Everyone likes to draw cards. This form of "sabotage" gameplay that triggers on dealing combat damage to an opponent will provide novel directions in terms of combat strategies and design space.

This keyword will become the primary keyword for the color pair that values knowledge, yet still lacks a common characteristic, featured keyword: blue and black Dimir. It perfectly fits the colors and the representative guild to a T. It grants blue a combat-oriented ability, which the color sorely lacks. Black returns as a favorite for card advantage next to its ally. Green henceforth will be relegated to secondary or tertiary for this ability.

1. If you had to remove evergreen status from a keyword mechanic that is currently evergreen, which one would you remove and why?

I would remove double strike from evergreen status.

Double strike is redundant. The similar result can be accomplished by raising the power of a first striker, or doubling the power of a creature.

Double strike rarely appears at common. It fails the NNWO automatically.

The primary colors for double strike are red and white. These two colors are also thematically very focused on combat. Many new sets will very likely come with new mechanics, and they tend to be put on red and white cards. So the omission of double strike will not be missed, as it will be easily replaced by the myriad of other combat-oriented mechanics.

The time to officially be able to comment on this challenge isn't here yet (if you wish to avoid getting disqualified that is.) But I can comment on other people's answers, sure!

On Evergreening

Bloodthirst/Raid: Good. I think a lot of people would have chosen this, but that's because it's rather reasonable.

Unearth: Not as common a choice, but reasonable enough. I'm a fan of any mechanic that has workarounds against counterspells.

The Ophidian ability: Good catch. I don't know if it warrants keyword status, since I don't know if you could stick five creatures with this ability in the same set. And sometimes the Ophidian ability loots, both blue and red style. But it's been in so many sets, at this point I'd like to see it keyworded too.

On de-evergreening:

Hexproof: I wouldn't have chosen it. Hexproof is a mechanic that forces creatures to engage in combat, instead of only using spells to eliminate threats. And Wizards wants Magic to be a game about creature combat. Invisible Stalker was a problem, but the mechanic, as a whole, increases interactivity, not decreases it.

Double Strike: Fair enough. This question was pretty hard because each of the keywords are doing a specific role. None of them are 'bad'. But Double Strike is just more of First Strike, so it is a little loose.

On returning mechanic:

Affinity: You're probably right. I mean, if I was judgeing I'd put this in the 'pass' category. There are 'always safe' spells that work well with Affinity. For example, Murder. No matter how cheap it gets, it's still just Murder. A Murder that costs is unfair, but it isn't unbalancing.

Radiance: You'd need to work for this one, but I'm sure one could make a good case for it. The biggest problem working against it is that cards that care about color haven't been seen... since there were core sets, maybe? But a block where color matters in a sort of one-shot fashion sounds like a good idea to me.

Hmmm. Maybe I'm post my answers tomorrow. I will say you all had very different answers from mine.

Double strike I can agree with being rather unfit for evergreen. It has inherit "flashy" value though so I would rather tackle with hexproof first.

Defender seems to be a popular choice for this question and is only "tolerated" within WotC and is the least liked evergreen among players. So it seems entirely reasonable at face value. As I've been thinking of it for a while now, it's kind of a safe, non-choice since removing it will do so little and you hardly have to justify it if you don't really spend much thought on it. Obviously this would indicate it being a prime candidate for de-greening, but actually you would have to focus really justifying its removal since in general it seems just easier to keep it. It's very flavorful as seen in that from the very beginning of the game it has had a "tribe" backing it up. Recently it has had one or two limited archetypes centered around it as well.

On the other hand, something like hexproof is actively keeping the game from having some hexproof/shroud-esque mechanic that would actually be usable at much higher frequencies. Hexproof does change your playstyle in that you can be much more carefree about enchanting your creatures. That unfortunately also plays into its noninteracting nature. Instead of increasing play options, it removes them completely from the opponent.

Unearth? It seems too complex to me. It has a nested keyword (haste) hardcoded in it and is an activated ability that both players would need to keep track of. Even mentioning exile is a bit questionable IMO. Beyond that, it automatically asserts a certain tendency toward graveyard themes into every set it's used in. As far as I understand it, having plenty of design space is rather low on the priorities of evergreens. I mean, just look at the current evergreens.

Bloodthirst would now demand that +1/+1 counters are the default thing and would make using -1/-1 counters invalid. If in a -1/-1 counter set bloodthirst would be "halted" it would automatically mean it's not really an evergreen.

Raid is a pseudokeyword so I'm not even sure if it qualifies. It's more of a condition rather than an ability itself. Given the all the possibilities you can do with it, I don't really see how you are supposed to sell it as an evergreen. Maybe if there was a restricted version of it...

Not to be a douche, but choosing raid and/or bloodthirst as answers to this question would indicate to me a bit that you don't really understand all the requirements that an evergreen ability has to meet.

Affinity I don't know what to think about. We haven't had cost reducing abilities as evergreens before, but being first at something didn't stop prowess. I guess one could try to make a compelling argument for it.

Radiance is pretty bad. It's has been stated on being at 8 on the storm scale so that alone should you give you some perspective on it how unlikely of a keyword it is to be used again, let alone evergreened.

At common, "curiosity" keyword would go out of hand too easily with its infinite card advantage potential. Also, I don't think there's a single keyword that would constantly keep you drawing more cards or even draw cards. Even with cycling you just kind of throw the card itself away.

EDIT: Derp, I totally missed the "What keyword was received the most poorly but has the most potential?" line.

I noted on these quickly myself here where there's a similar thread:

MTG Salvation "The Great Designer Search 3" thread

Make evergreen: Afflict has been something I have been suggesting as a blue-black evergreen, and I would at least look very hard on the reception among the public and the test data.

Remove from evergreen: Double strike could easily be demoted to deciduous since it already is a mechanic not for common use.

Received poorly with potential: Haunt would do better the second time around if one resisted the urge to put it on permanents and nonpermanents in the same set.

For the last one I might want to have a second look though at which mechanics were actually received poorly since I am not aware of the public reception of some of these.

@SecretInfiltrator:

The evergreen variant of afflict would likely require it be non-parametrized such as being tied to the creature's power. The mechanic has been stated multiple times by MaRo as being limited in its design space and non-parametrizing it would make it even more so. In "State of Design 2017" article, relating to how it functioned with the block/set it was introduced, it's noted that the players weren't sold in its necessity and that MaRo also wouldn't have made it if given the choice again.
As for mechanic, it would obviously be 'another (pretty-much-)evasion' mechanic, which is quite bland.

tl;dr I don't get what you see in it.

@Tahazzar on Hexproof: I still don't understand how you think a mechanic that forces players to engage with it in combat is 'decreasing complexity'. Combat is pretty complex. Murder is not.

Personally, I'd consider to be the most important keyword Magic has, after Flying/Reach. But I'm cool with disagreeing. The point of the excercise isn't who's wrong or who's right. It's more about being able to back up an argument.

jmg: What does the embargo say? Everyone seemed to be assuming it was ok to talk about it elsewhere.

I liked a combat-relevant mechanic for new evergreen, as I think those are often most useful. Even if it only goes on a few creatures a set, if it can make interesting commons, that's good. People suggested several good options here.

But I gave the nod to totem armor instead. I think if most positive auras had totem armor, simple auras would be a lot more usable. And "flying, totem armor" aura is playable without needing a P/T boost as well, which is pretty much always required on current auras if people are going to actually use them.

Also, totem armor auras reward awesome-ing up your fun creature, whereas P/T boost auras reward spreading the risk to create multiple mid-size creatures, which is less fun.

But simultaneously, a P/T boost makes a creature a lot harder to handle for your opponent. I think the original totem armor creatures were too good for having totem armor AND a P/T boost. Having totem armor makes a trade feel more reasonable to both players.

I said, replace hexproof with activated-only hexproof, which may not need to be keyworded.

Now I read the other comments, I'm not sure that solves the problem by itself. I think jmg is right that you want some creatures where "removal" isn't a single answer to them. But the problem is when the creature gets out of hand in other ways as well.

Maybe we should go back to shroud after all. At the time I agreed with the argument that all-upside mechanics were more interesting. But at the time I didn't appreciate how miserable it was when someone put equipment or aura on a hexproof creature.

@jmgariepy:

Whether to play that Murder (during combat possibly) is a choice. A very interactive choice in fact, since it can't really be even played unless there is something for it interact (target) with it. Not being able to play it at all is the very definition of non-interactive.

Just recently, there was a period in standard where only one or two hexproof cards existed at common. I think at one point it was that only one card that granted hexproof temporarily was at common. Even from the standpoint of modern, hexproof hasn't existed as a keyword for that long so I really don't get where you have gotten the idea that it would be one of the most important keywords. As far as I know, it's whole existence is riddled with people complaining about it and/or some super non-interactive deck forming around it.

And yeah, these questions are probably like 50 % about estimating your argumentation skills. I'm pretty sure it was also stated as being the case in retrospect to one of the earlier GDS.

@Jack: I just looked at it, and it looks like it's fine to publish. I thought one had to wait until Wizards made an official decision about stage one, but I guess its kosher to talk about as long as the deadline is in.

I went with make bloodthirst evergreen, strip defender, and bring back bushido. Bloodthirst is pretty simple and encourages combat, which is what you want in an evergreen ability. +1/+1 counters are good actually, because lots of mechanics interact with them so it opens up interesting design space when interacting with set mechanics. And you just need to be smart about it and not put it in -1/-1 sets, like they didn't put prowess in Ixalan.

Defender would not be a mechanic if it hadn't been grandfathered in. Did you know that there are more creatures with daunt in standard than defender? But daunt doesn't qualify for evergreen status. Defender lets you make cards like Hightide Hermit, but you don't need to put it on every 0/4. I vote move it to deciduous.

And bring back bushido. It got dinged for being in a terribly received block, but it's a solid mechanic and there's plenty of design space in afflict-adjacent space, plus they did some wacky stuff the first time that can be revisited (bushido X and "points of bushido").