Charm Cube (Shards/Wedges)

Rob Dennis

Rob Dennis

Developer
How This Started
It started with this tweet:
@klug_alters said:
If Shards taught me anything, it's that no matter how good the Charms in #mtgktk are, none of them with be good enough in cube. Sort of sad.

2014-11-08
Click to expand...

I also thought it was sad, and I was looking for a new project, so I started brainstorming some ideas.

@cuesbey said:
I think I'm making a new cube based around a single design experiment: What would be needed for all the shards/khans charms to be in demand?

2014-11-10
Click to expand...


I talked this idea over with a few folks, and "in demand" has been heard as anything from p1p1-windmill to "not sideboarded" so I've refined that down to this target:

"You'd feel happy to take 'your' charm pack 3, pick 1"
Click to expand...

This has a few implications:
  • if it's "your" charm, that implies that there were already incentives to be in that shard/wedge instead of a 2-color build
  • if you're in a shard/wedge, not only were there incentives, but mana fixing should be "good enough" but not so good that you can go 5-color "stuff"
  • charms are flexible, not rawly powerful, so the mean powerlevel of the rest of the pool has a maximum before flexibility stops mattering
Planning/Execution/The List

We're at 425 cards now, and the build was complicated enough I put in a bunch of extra work on Lambic (see signature) because there's no chance of making sense of this on a spreadsheet:
7frddWZ.png

(showing how the 458 cards I had at that moment broke down into a bunch of different buckets)

Let's go over how we got to the cards we chose:

General Design Philosophies
  • Multiples of a color (e.g. {2}{W}{W}) discouraged unless it's a unique effect and excellent late game topdeck
  • No Planeswalkers, and similar repetive play pattern things (e.g. Merfolk Looter or Recurring Nightmare) are under suspicion
  • As singleton as possible, but not when the cost is too high
  • An easy to cast p1p1 "to stay open" shouldn't always be the correct play
  • A step farther, going all-in on a p1p1 should be rewarded and not easily cut off
  • Prioritizing 5-6 pieces of fixing in your shard wedge should be enough
  • curve/creature size balanced around morph, which I wanted to keep as a fun way to subtly mana-fix and play to the board as consistently as possible. Morphs have been leaving over time, but there's still a sense that "2/3 for 3 with marginal upside is a C+"
Fixing
in a multi-color cube, your fixing is an important consideration. Here's where I landed
When 10 of the shock lands were something else, it had the effect of making all the landcyclers a LOT worse due to lower percentage of cards with the basic land types. I'd like to at least try them and see.

I added a second full cycle of the tri-lands due to consistent feedback from folks in this thread that they weren't able to find fixing after they were already in their colors. After some additional testing, it felt right without empowering a T1 5-color goodstuff deck.

Morphs

I'm following the KTK guidelines on morph costs. This means a morph can't just eat another morph until 5 mana. Additionally, cards that unmorph for multiples of a single color are under strong scrutiny.

Removal
Modern draft sets have been defined by the quality and types of removal being calibrated to support the set's themes. No more having a common Doom Blade in every set just because that's what's expected. For this cube, I'm taking some cues from KTK, because morph is such a big part of the format, and provided us with some good heuristics to follow:

  • 1 or 2 mana instants shouldn't be able to reach out and kill morphs
  • A toughness of 4 or greater is "pretty safe" and 5 or greater is "as safe as can be"
  • Better / full-value removal is generally multicolor (consistent with overall goals)
  • A larger than normal percentage of removal is enchantment based, to give the Naturalize modes of the charms a good chance at getting value and interact with bounce/blink/sacrifice effects.
Based on this metric, the removal modes found on all the charms compare very favorably, and I'd expect that contribute to them getting played.
 
Last edited:
Rob Dennis

Rob Dennis

Developer
Wedges/Shards
I built in some synergies to each shard/wedge, as a way to reward people drafting a deck instead of a bunch of cards. This has the nice side effect of presenting choices between a generically good card like Siege Rhino and an engine card like Abzan Ascendancy where the right pick for you depends on a lot factors. Since a guild is a member of 3 different shards/wedges, care and attention was needed to make sure a particular card was in demand by multiple drafters.

Let's go over each shards/wedges. Wedges first, because I believe they had more to work with than the shards. I'll mention some examples of the synergies we're mentioning, and an example deck. These have been drafted at cubetutor over a couple of different revisions and there will be cards that aren't in the latest list, but they do a good job of showing what I'm going for.
Wedges
Wedges were the lead groupings for me because they had relatively deep themes that played pretty well with other blocks/cards. Compare "attacking matters" to "artifact matters".

Abzan

abzan counter-lords from CubeTutor.com











Jeskai

jeskai token/aggro from CubeTutor.com











Sultai

sultai zombie triggers from CubeTutor.com











Mardu

mardu aggro from CubeTutor.com













Temur

Temur stuff from CubeTutor.com












Shards
Despite not being the lead, I fully expect people to go into shards, but the shard card pool is mostly split being supporting wedge themes or just being good value.

Bant

bant value from CubeTutor.com












Esper

control from CubeTutor.com











Grixis

grixis control from CubeTutor.com












Jund
Jund

jund death triggers from CubeTutor.com












Naya

naya tokens from CubeTutor.com









 
Rob Dennis

Rob Dennis

Developer
How You Can Help

  • Do some test drafts (here's a link, which is also in my signature, make sure to login and save your drafts so I can see them)
  • Opinions?
  • If you build something similar and have any play/draft reports, I'd love to hear them. I'm not going to get more than a few reps a month, and this kind of thing could be iterated over every day with new findings
 
Grillo_Parlante

Grillo_Parlante

Contributor
First, I think this is a really great cube. I love the approach that you took when designing it, and I can tell that you put a lot of energy in creating it. I took it for a spin, and here are the results:

Bant Tokens from CubeTutor.com










The draft itself felt great. I didn't feel overwhelmed, or without direction, I didn't feel that one color or color combination outpaced the others. The power level felt about right, with enough spikes to provide guidance and excitment, but nothing so clearly powerful as to be a dominating auto-pick. As with every draft I do of someone elses cube, silent departure showed up and made my deck, which is great.

I ended up drafting bant off of a first pick noble hierarch, and I was happy I got to do that as you mentioned your focus was on the wedge's rather than the shards. My shard deck, I feel, came out great, and was something I would enjoy playing.

When looking over the sample decks you provided, and the list itself, it seemed like there is a focus on two and three drops, and a lack of one drops, so my strategy was to try to go low and fast to race under those decks, while also having a strong long game plan (tokens + cruise). I really like that Khan's tension, between tempo and raw power, and I think I may have gotten a little bit too much of both world's, which might be something to watch. This also seems like the type of deck capable of producing board stalls.

I was also wondering if you felt that maybe some of the shards need more depth? For example, Grixis seems very focused on producing "fair" attrition-style control decks.

Really great list though, and I predict you are going to have a lot of fun with it.
 
Rob Dennis

Rob Dennis

Developer
Thanks so much for taking the time to draft, when I saw it on CT, I thought it was a really cool take. You're definitely right there are not a lot of 1 drops at all, and not as many 2 drops as a typical cube I'd build, this was a KTK influence, where a legit T1 play was laying tapland. I knew that there were some focused red/black "fun police" decks that'd try to exploit this, but didn't have a lot of info on what else was out there for people focusing on the cheapest possible stuff.

Here's an example of that Rakdos deck:

Rakdos aggro from CubeTutor.com











For the color-combos that are least distinctive, I have it as: Esper, Grixis, and Temur. These are kind of just value decks, and I predicted that some folks who enjoy playing, but in the draft are more "card pickers" than "deck builders" would feel most comfortable there. My theory there:
  • Esper was too parasitic in shards
  • Grixis is way too strong on color commitments thanks to the black influence
  • Temur's "4-power matters" theme doesn't have a super distinctive set of synergies and pushes a fair amount of "good stuff"
I think I could probably beef up one of these to be a bit more synergistic, but I'm not sure which is the most fruitful, and I'm not confident at all I could do all without getting overly proscriptive.
 
shamizy

shamizy

Love the idea behind this cube, was thinking about implementing a tri-color theme sometime this Winter into my own Cube. I didn't think to take it this far though, super interesting! This is what I came up with, wish I coulda had another fetch or relevant land-cycler come my way:

Jeskai Spellslinger from CubeTutor.com










 
Rasmus Källqvist

Rasmus Källqvist

Warwolt's draft of Charm Cube on 14/10/2014 from CubeTutor.com










Did another try with some more cyclers. I think you've done a good job at bleeding the archetypes together. Ended up with a blink / token deck this time, actually started as blue green after first picking a valley rannet but ended up green white after picking up a master biomancer mid draft and looking for ways to exploit it.

Edit:

Warwolt's draft of Charm Cube on 14/10/2014 from CubeTutor.com












Ended up running yet another. I've got to say it is really fun to draft the cube and I'm getting more confident in saying that the bleeding is well done. I'm looking into a pack and thinking "This could work well with this card" and it opens up my eyes for different kind of themes in later picks, the counter theme showed up as a way to bolster my tokens (WW by heart!) but I reached enough critical mass to dive into the counter matters theme and it payed off.
 
Goldenpineapple

Goldenpineapple

I don't know if this is very useful or not but I decided to not try to stick to an archetype and do what the monkeys around here did when my cube was very gold: Pick a land when possible, slam value, removal, and bombs if there are none:

I drafted Lands








The comp I'm doing this on has some weird formatting bugs so it's a nightmare to try to pick basics. I feel like I got a fair amount of premium removal and a fair number of powerful dudes that don't need any support at all in spiritmonger, ruric thar and aristocrat. I realize my blue is very loose but I think all of those cards are worth it.
 
Rob Dennis

Rob Dennis

Developer
wow, super grateful for everyone taking the time to draft and have such positive things to say :)

shamizy said:
jeskai spellslinger
Click to expand...

you built this a lot more defensively* than I might have expected, and curious how it'd play out. You mention missing 1 more piece of fixing, so I could see how going shields up gives you time to draw into your colors and fly over.

* e.g monastery flock over mystic of the hidden way

Rasmus Källqvist said:
bant blink counters
abzan counters/tokens
Click to expand...

I have to say, the abzan deck you have is SUPER SWEET. I'd love to see this go off with grim haruspex and maw of the obzedat or any early beastmaster ascension. I'm really thankful to hear your comments about the bleeding. This is definitely the area I've invested the most time, and everyone who designs a cube knows that "I hope they see what I built there" feeling.

Goldenpineapple said:
I don't know if this is very useful or not but I decided to not try to stick to an archetype and do what the monkeys around here did when my cube was very gold: Pick a land when possible, slam value, removal, and bombs if there are none
Click to expand...

it's super helpful since it's the testing area I've invested the least time. My local playgroup has a lot more of people running two colors in KTK, so I was more about encouraging them to branch out.

I three goals in this scenario:
  • the B+/A- version of a focused archetype is favored against 5-color stuff, but due to lack of synergy and not power/consistency
  • there's still enough fixing leftover to make that B+/A- version (if you prioritize 2 typed-duals and use landcycler I think it'd work in the pod you test drafted on)
  • the people doing the 5-color drafting still have a bunch of "oh man, this is a sweet one" moments
You definitely got a lot of good removal, but probably less than you could have expected if you went deep {B/R} or {W/B}. Which I think makes sense, since you prioritized lands and presumably were rewarded with more ramp and better finishers.

I'm curious if you thought the draft had those moments for you?
 
Dom Harvey

Dom Harvey

Contributor
From a quick glance, I'd worry about an imbalance between the Cube's best cards and some of the filler. In the Rakdos deck, your win percentage goes up so much if you draw Falkenrath Aristocrat over the likes of Ill-Tempered Cyclops.

Drafted a nice Esper deck that I forgot to save, no real synergies going on though.
 
Rob Dennis

Rob Dennis

Developer
that's certainly a risk area, and this is the first project I've gone in on that's not targeting a flat power level across cards so I don't have a feel from the draft what's the right balance.

Basically, KTK is a great draft set so far, and there's a huge difference in expected outcomes in drawing a butcher of the horde vs mardu roughrider. This is because the devs played it out and confirmed that a good deck with no rares can stomp a mediocre deck with some bombs.

Hoping that testing can feel that out, but I know that just based on my weekly game night, I'm not going to get enough reps to be sure.
 
Rob Dennis

Rob Dennis

Developer
shamizy said:
I just don't seem to see much on-color fixing in the drafts I've done. Weird.
Click to expand...

Where I end up starting is drafting the first two typed dual lands I see (typically by p1p8), then building whatever good in those colors. This has the nice side effect of giving me the most chances to find relevant land cycling. I am hoping that's not the only way to draft this list though.

edit:
commented on your abzan deck: http://cubetutor.com/cubedeck/207472
and I think could be remade as bant given your existing sideboard (I'd play the citadel either way because it give you GW)

and on the mardu deck: http://cubetutor.com/cubedeck/207477
I think there's a strong / consistent rakdos deck here with a very light splash (crackling doom off the warshrieker and 1-2 plains).
in general, I'd recommend the triland if you only use 2 colors in 99% of cases
 
Rob Dennis

Rob Dennis

Developer
For everyone who saves a deck on CT, I've gone though and commented on all over them. I'll continue when new ones show up, for at least a while
 
Rasmus Källqvist

Rasmus Källqvist

I just noticed that you are running only a singleton of the fetchlands, is this a concious thing or was there just not enough space? While the land cyclers are phenomenal, I think they are too slow for some of the faster decks. I saw grillo build that naya deck and exclude Thoctar and maybe including more fetches could help that? One obstacle is obviously size, I dont know if you really want to run 60 lands and if not you need to cut something and you probably need that saturation of duals to have the landcyclers be as powerful a choice. How fast and consistent do you want the two and splashed two colored decks to be?
 
Rob Dennis

Rob Dennis

Developer
It was a conscious decision in favor of the landcyclers, primarily because it was slower.

For some background, my main list was a tight 360 with 2x fetches and it definitely allowed people to hit their colors on turns 1 and 2, but that's a much faster speed than khans and I didn't feel like it was consistent with the slower/midrangey feel of a morph set (which is why there's a relatively small number of 1 and 2 drops that don't have late-game value). This is a similar reason that why I'm using 2 mana elves as the most common mana guys.

I expected a typical sequence for a normal deck was:
T1 land (maybe a tri-land), go
T2 land go, end-step land cycle
T3 land (BY NOW HAVE ALL YOUR COLORS)

Where an aggressive deck is probably going 2 colors (maybe with a very tiny splash off a fiery fall or an off-color dual)
and can try to do a T2/T3/T4 curve out.

I think this conversation has given me two thoughts:

I don't think woolly thoctar is very well-positioned because I don't really think it's an impactful T5 play, and I don't want to do what's necessary to guarantee that it's "always" a T3 play. As it is, Naya is in a decent spot to play it T3, but only if you get some mana creatures. Something like Rhox War Monk is a great T5 defensive play, so I think that's just on the right side of that distinction

I think I need to find room for Evolving Wilds and Terramorphic Expanse because that's a low downside effect I think I forgot about
 
Grillo_Parlante

Grillo_Parlante

Contributor
After doing another draft, I was wondering if you might consider running an additional copy of each of the tri-lands?

It seems like sometimes the packs break in a way where you get lots of on-color fixing, and other times where you get very little. I think part of that is caused by running only 52 land slots in a 425 tri-color list (12.2%), with the majority of your mana fixing lands geared towards bi-color fixing, meaning sometimes you just get screwed because you are in the wrong colors. For comparison, my cube is 11.6% mana fixing lands (40 color fixing lands in a 360 list), while not running a major tri-color theme. The tri-color lands not only fit better into your overall game pacing than shocks or ABU duels--due to being CIPT--but support a broader number of bi as well as tri color configurations.

For example, seaside citadel goes into a dedicated tri-color UGW deck, or could go into another dedicated tri-color deck, but as bi-color fixing for U/W, G/W or G/U.
 
Rasmus Källqvist

Rasmus Källqvist

I think there are some cards that are a bit filler in the wubrg so maybe you can find room by cutting some, and adding another round of tri-lands as suggested and fitting in the basicland fetches?
 
Rob Dennis

Rob Dennis

Developer
I saw this last night, and was originally going to approach this as quibbling over details:
- one thing to consider is that there are also 15 landcyclers which pushes it up to ~15.7% non-elf "fixing"
- this means that there will be 7-8 pieces of fixing opened in each seat, or 25 sources total in a grid draft, which were numbers I felt good about for supporting 3-color, but not 5-color. Especially if you pick your fixing first, then draft a deck.
- the commitment to not having duplicate mana symbols means the number of sources you need to cast all your spells are so much less than playing 3-color in a "normal" list (where you have costs like {2}{W}{W} and {2}{G}{G} plus a small splash)

BUT, the fact that there's "just enough fixing" to have a good experience if know the "right way" to draft it isn't promoting a great experience.

I think adding in more tri-lands is a good place to start, and if it really turns into 5-color everywhere, we can re-adjust.
 
Rob Dennis

Rob Dennis

Developer
have ~8 practice drafts after the tri-land change, and at first I wasn't super happy with it. It seemed like I didn't have to prioritize fixing, I made the landcycling part a little less important, and now it really seemed possible to try going five-color stuff.

But now, I really like we ended up:
- I have the option of drafting a sweet card first, instead of just opening on 3 fixers then seeing where I'm left
- landcycling is still important, and I realized that some of the graveyard synergies I built in really want to use these cyclers, so there's some extra value there
- I tried doing a five-color deck a few time, and there's basically no payoff. The decision to minimize onboard, continuous card advantage engines means there's no way to out power level an opponent who's leveraging synergies. This is consistent with my 5-colors goals I mentioned earlier
 
Rasmus Källqvist

Rasmus Källqvist

So if I understand you correctly the 5 color archetype is still viable in terms of a deck that can come together during a draft and be playable but the lack of synergies in comparison with the two and three color decks making it vulnerable and able to beat?
 
Rob Dennis

Rob Dennis

Developer
Rasmus Källqvist said:
So if I understand you correctly the 5 color archetype is still viable in terms of a deck that can come together during a draft and be playable but the lack of synergies in comparison with the two and three color decks make it vurnerable and able to beat?
Click to expand...

Yep, that's exactly it, and I've edited me last post since I forgot a few words in there.

I'm happy with that kind of setup. I don't like it when draft lands every time and scoop up bombs pick 13 is actually the best thing to be doing, because it feels like taking advantage of newer drafters who don't know the secret code.
 
You must log in or register to reply here.
Top