When I first saw Apex Devastator, I remember seeing the card say Cascade, Cascade, Cascade, Cascade, thinking that had a very clean yet Mythic feeling. When developing a new ability or using an ability in a new way, how do you determine when it's correct to write out the keyword four times versus saying "Cascade four times". In a similar way, how do you decide, for instance, in the case of Gitaxian Spellstalker, whether it should be prowess, prowess, as opposed something like Prowess 2?

Templating has an emotional impact component. While clarity is usually the driving factor, excitement is also important.

Countably infinite does NOT mean finite. If you shuffled an infinite number of Clown tokens into your library, you would never draw another kind of card in that game. Infinity plus however many cards were in your library is still infinity. Imagine if you didn't shuffle. How many cards would be on top of the first real card? Infinite cards are on top, so you'd never draw the card. If you shuffled the deck, that would still be the case. There would be an infinite number of Clown tokens before and after each real card, because the chance of getting a Clown token between any two cards while shuffling is 100%. Not 99%, not 99.999...%, just 100%. Infinity divided by the cards in your library is infinity. You can never draw any real cards, because even if you took off the top card, infinity minus one is still infinity.Grenzo, Dungeon Warden would always hit, since there are an infinite number of Clown creature token cards on the bottom of your library. Since tokens have 0 mana value, you can't cascade into any real cards. You could mill out if you got hit by Tasha's Hideous Laughter. If you chose Land for Abundant Harvest, you would get one if you had a land in your library. However, if you got hit by Cut Your Losses, you would only mill Clown tokens, and no cards from your library. Infinity plus the number of real cards in your library divided in half is infinity, so you'd never get to the first real card, and you'd still have an infinite number of Clown tokens on top. Finally, if you happened to play Braingeyser with X equal to infinity, you would STILL not draw any real cards. Infinity minus infinity is infinity. You would have an infinite number of Clown tokens in your hand, and an infinite number of Clown tokens on top of your library.It's easier if you think of infinity as a description or a quality rather than a number. Treating it like a number infects whatever math you're doing with odd-looking results.

FYI

"Part of what makes Cascade fun also makes it problematic." I get part of your job is to protect the future of the game. On the flipside I would prefer if designers prioritizes stuff that was fun even if it is problematic. I think the responsibility of game designers should be to maximize fun when playing a card and manage problems in a way that doesn't compromise the intention of the design. With that said, I get that this is me saying something in a vacuum.

Cards being problematic, with time, make them unfun. One does not deal in fun without also confronting problematic designs.

Follow up to my recent question about upgrading mechanics: like I said, it's become more of a recent trend to either introduce new mechanics as upgrades of older ones (Disguise, Cascade, Blitz) or even just tweaking old mechanics to make them a bit more intuitive to use (Hideaway, Amass, Phasing).Is this a conscious and active effort on R&D's part? Or is it more a product of the game's long lifespan? I.e. you can only create so many new mechanics before you have to start reinventing the wheel, so to speak.

Design technology keeps improving and the game keeps adapting. Good mechanics can fall out of favor due to changes, so updating them is just good design practice.

Most of my issue with UB, particularly the commander decks, is their lack of availability on MtGO, as well as their lack of cohesion at some points. Individually, the cards are absolutely flavor wins, but as part of a precon they really don't work cohesively. Take the Timey Wimey precon for example. That deck was absolutely pulling itself in different directions by trying to be time counters and investigate, and then several cards that had nothing to do with either strategy. I would have been more satisfied had the deck been all in on the time counters, rather than trying to split the deck's strategy.Another example is the Ruinous Powers 40K precon. It had a very heavy demon theme, but it was split between demons-matter/aggro and cascade/spellslinger. Flavorfully each card is an absolute banger. But the decks as a whole really lack a real focus on one or the other. An example of UB doing well is the LotR set as a whole. It allowed for flavor wins while still keeping themes appropriate for its strategy, with blue/black/red being heavy on amassing orc armies and buffing orcs, while green/blue was focused on scrying and payoffs for scrying. Flavor wins, while still keeping a clear focus for deckbuilding.

You’re comparing apples and oranges. A preconstructed deck has different limitations than a randomized product.

With keywords being remade, old cards that care about them could become obsolete. Why can't those old cards be erratad to batch the new and old parts? It's barely even a functional errata since they'd work identically as before, but with the new keywords included. Four cards from the Onslaught block cared about creatures with morph. They were changed to batch morph and megamorph as "morph abilities" when Tarkir came out. With disguise as remade morph and the implication that it's replacing morph entirely, why not add it to the batch and have cards care about the batch instead the single keyword? Same with discover and cascade, or how you've said a remade modular is more likely than modular returning. If fan favorite keywords are going to be remade and replaced, could they least be done in a way that's backwards compatible with the old support cards?

It’s a slippery slope. When is a mechanic a redone version and when is it just a tweak? What if we try something and then go back? The errata now possibly prevents us from reprinting the card. Functional errata can cause as many problems as it solves, so we have to be very careful with it.In the case of morph, the vast majority of the cards that interact with it say “face down” or “colorless” and don’t need any errata.This isn’t so say we can’t consider it, but it comes with baggage.

The difference between Cascade and Discover seems about the same as the difference between how Hideaway used to work and how it works as of NEO. Why did you update Hideaway but replace Cascade? Is it because Cascade is a more popular mechanic than Hideaway was? Is there some mechanical reason? Did the Hideaway update go poorly?

It has to do with what the exact changes are and how many cards of the original version get used. Also, the change could happen if the original flavor wants a tweak.

How likely would you to be to re-use an existing keyword in a set but call it something different strictly for flavour reasons? Would you ever do it? (I ask because there's a commander-set card that used a face-down creature to represent someone mysterious, but it still uses the keyword "manifest," which is fine for a one-of card but might not be the best fit for a Murder Mystery set, as an example.)

It’s not something we’ll do without a great deal of thought, but the option isn’t off the table. Note that our current strategy is to update the mechanic when we give it a new name (cascade vs. discover), so usually they play differently.

Such a shame about cascade… was my favourite mechanic

Discover plays pretty similarly.

Hey Mark, with discover being a better cascade mechanic but pretty much all metrics do you think we will see any more support for cascade? Sad to think my Averna the Chaos bloom deck wont be getting any new cascaders in the future.

We might do the occasional one-of’s in supplemental sets, but yeah, I don’t see cascade likely showing up en masse in future premier sets.