Commander Legends: Battle for Baldur's Gate Vision Design Handoff

Posted in Making Magic on June 6, 2022

By Mark Rosewater

Working in R&D since '95, Mark became Magic head designer in '03. His hobbies: spending time with family, writing about Magic in all mediums, and creating short bios.

As part of leading a Vision Design team, the lead designer must create what is known as the vision design handoff document, where they describe the vision of the set for the Set Design team, walking through the set's mechanics and themes. I've shown off several different vision design handoff documents in this column:

Today, for the first time, I'm going to show off a vision design handoff document written by someone else. What follows is the vision design handoff document for Commander Legends: Battle for Baldur's Gate (CLB) written by Glenn Jones, the lead vision designer for CLB. The lead designer has a lot of latitude in how they want to structure their handoff document, so you'll see that Glenn puts his together a bit differently than I do. Also, Commander Legends: Battle for Baldur's Gate was Glenn's first time leading a design team, and thus, his first time creating one of these documents. As with all my vision design handoff articles, this is the official document as it was turned in, offset with boxed commentary.

5XX "Pentagon" Vision Design

Vision Design Team

  • Glenn Jones (lead)
  • Corey Bowen
  • Ethan Fleischer
  • Michael Grothe
  • David Iezzi
  • Mike Mearls (D&D SME)
  • Chris Mooney
  • Jules Robins
  • Emily Teng (creative lead)
  • Chris VanMeter
  • Gavin Verhey

Goals

Mix D&D into Commander

Character-building: One of the loudest and most unique elements of the set should be how it manifests the character-building experience of D&D within the sphere of drafting Magic, setting it apart as a product and experience from anything Magic has done before.

Right now, the vision design handoff focuses on using the mechanic replacing partner to represent this dynamic, but this is essentially a single decision and doesn't apply to multicolor commanders, which are very popular to draft. It's possible the mechanic is not loud enough, and that there are ways to introduce more decision points or applications.

Adventuring companies: Similarly, the set cashes in on the sensation of using creatures to build a group of adventurers with a common purpose—winning this game of Magic. There are so many executions that multiple viable ones were never tested (or considered). The popularity of Zendikar Rising's party and the excitement about how it relates to D&D implies that this element of the set has the potential to create a lot of positive reception at reduced cost, given the density of class creature types already necessary.

Multiplayer: Goals for the Limited environment were to match the kind of experience offered by Commander Legends—games should last 60–120 minutes, tend to be one game per draft, and tend to have three to five players per game. Similarly, the set should be drafted in pods of four to eight and in a variety of styles. It's fine to deviate from some of these norms in service of a larger gameplay goal. As Commander Legends releases, monitor how its gameplay is received.

The vision design files may be underpowered relative to the final target. This is just an artifact of the team's focus on testing mechanics and iterating on their file impacts, not an indicator of baseline power level. The set is closer to Conspiracy: Take the Crown than it should be, on a strength scale.

Vision Design debated on cooperative elements to mix in, but nothing materialized that seemed ideal. "The torch" is the best existing house for that kind of gameplay—or more specifically, its rooms.

Awesome reprints: Like Commander Legends and other past multiplayer sets, the set should reserve room for awesome Commander reprints that are difficult to put in Standard and Masters sets. Unlike those sets, the D&D IP restricts the potential inclusions even further.

Complexity: Unlike Adventures in the Forgotten Realms, which had a pseudo-core set design philosophy so that it would bridge the gap for D&D players into Magic, "Pentagon" assumes its players either have a significant familiarity with Draft, Commander, or both. It's not as complex as a Masters set, but it's pushing into the higher end of typical Standard releases. The primary audience for the content is expected to be existing Commander players, and the unique ruleset of a Commander Draft format also caters most strongly to players interested in a novel Magic experience, not a first Magic experience.

Explore Baldur's Gate

Story and setting: The setting is the Sword Coast, and specifically Baldur's Gate. The set can use characters and stories from a wide range of time periods, as long as they're specific to this region of the Forgotten Realms. The "Time of Troubles" is one of the most major events in Forgotten Realms history and centered here. We're specifically leaning on it as a crossover-level event to incorporate characters from several different stories and settings. It's not a requirement that each character appear during any Baldur's Gate story—the city is so significant that any major character has likely passed through.

Where Adventures in the Forgotten Realms was "fantasy adventures and dungeons," "Pentagon" is "mysteries and monsters of the Sword Coast." The setting should feel urban and dark, like a fantasy version of Gotham City. The set has a dramatic ratio of sentient humanoids to "monsters" to portray that motif, with the expectation that non-adventurer/NPC creatures will probably be in the neighborhood of 15%.

There was text here, but you can't see it yet. You're just not ready for it. Maybe someday. But not today.

The Dead Three: These three evil gods are a huge part of the history of Baldur's Gate, and Bhaal has been a primary antagonist in the video game series, so we're using them as timeless antagonists, sort of how Phyrexia functioned in early Magic. Portraying these characters in the set has a variety of options, thanks to the Time of Troubles and Magic's various ways to mechanically manifest divinity.

Current Mechanics

Story and Backgrounds

TLDR: Something like this, but unsure about the execution.

Story appears on monocolor legendary creatures. It allows the player to add one color to that creature while using it as their commander by selecting a Background of that color. Backgrounds start the game in the command zone and have an emblem-like ability that costs mana to turn on for the rest of the game. This ability should only have functionality while the commander itself is on the battlefield.

Of all the partner variants, Background has been the most promising.

  • Reduces the anxiety of having a legal commander to functionally zero, with multicolor commanders covering any gaps not made up for by this high as-fan.
  • Adds agency to the draft and gives players a character-centric decision to make.
    • Related, it's flavorful. Backgrounds are a fun part of character creation in D&D.
  • Has a lot of executions, depending on how much it interacts with the game, deck, or draft.
  • Doesn't add cumbersome amounts of rules text to the commanders.

That said, it has concerns, too.

  • Introduces an additional object that must be accounted for in collation and/or accessories.
    • They could be draftable, but this threatens the "safe" sensation and adds way more of them to the draft than necessary, since ~6 will be used.
  • Multicolor legends tend to be the most exciting in the set. Commander Legends's partners could compete by going up a card, but Backgrounds will have to carve out their own space.
    • Some playtest games have featured no Backgrounds.

The file currently reflects a variant in which Backgrounds appear on the back of cards in the set as faux-DFCs that are only ever one face—front in the deck, back in the command zone. It's a haphazard implementation since time was of the essence to playtest this variant at all. To work on it further, the front and back of these cards should be adjusted to line up more closely in their effects. However, this path is a bit more fraught because the synergy-driven Background designs being reflected by the cards on their front faces will mean players draft the cards for their front face more often, reducing players' access to Backgrounds and risking trainwrecks with no legal commanders.

As I've continued to personally consider the mechanic, these are my two favorite executions:

  1. A list of five to ten Backgrounds exists outside the draft for everyone. A randomization method could determine which five are being pulled from a larger list for the draft, perhaps included in an app.
  2. A list of five "basic" Backgrounds exist outside the draft for everyone, but a few higher-rarity cards—perhaps five uncommon, five rare, five mythic rare—include stronger ones as faux-DFCs, or whatever achieves an interesting as-fan without filling the draft with too many Backgrounds.

"Pick up the Torch"

TLDR: Confident, but the exact composition of the dungeon merits iteration and testing.

Whenever a card instructs a player to "pick up the torch," it removes ownership of an outside-the-game status—the owner of the torch—from each other player and gives it to that player. This functionality is essentially the same as how the monarch transfers. The bearer of the torch similarly has two abilities:

  • At the end of combat, advance within the catacombs.
  • Whenever one or more creatures deal damage to you, their controller picks up the torch.

The "catacombs" is a single-card map shared by each player within the game. Its functionality is essentially that of a communal dungeon card from Adventures in the Forgotten Realms—players move through it, choosing rooms and triggering the effects within those rooms.

Playtests have positively received the mechanic in a general sense, but the individual rooms of each dungeon still merit work. Players are expected to complete a lap in a game, maybe two, so keeping the abilities useful is important. Gameplay suffers when rooms provide the occupant with particularly effective ways to protect themselves, because that makes the mechanic feel inaccessible to the rest of the table. It's the most fun when it's moving.

As mentioned above, room triggers could play up the multiplayer nature of games and perhaps include some elements that care about the relative positions of other players. Some random examples:

  • Draw a card for each player in this room.
  • Deal 2 damage to each player in an adjacent room.

Some unanswered questions in this vein the team did not explore to a conclusion:

  • Is this different enough from monarch?
  • How does this interact with monarch, if at all?
  • Is it fun to have this and monarch in the same game?

Class Party

TLDR: Creatively appealing, but the current implementation is a bit unsatisfying.

The file has many cards (~55) that care about controlling as many creatures as possible with different "jobs." This will be supported by the comprehensive rules, batching every job creature type in Magic under the word "class" and using that to count it up (approximately 40). Aaron is fine with this approach, provided an ad card in the product lists all the class creature types—not just the ones in the set, but the comprehensive rules entry.

Of the alternative ways to achieve this gameplay, my next favorite (untested) option is having "kickers" for controlling a creature of a class type to tell a story, for example:

  • A Naturalize that draws a card if you control a Druid.
  • A Knight that costs less if you control a Bard.
  • A Barbarian that gets stronger if you control a Warrior.

The most important thing for a mechanic in this space is rewarding diversity of classes, not tribal. It's fine if someone drafts a lot of non-Warrior cards that care about having a Warrior and a lot of Warriors, but not if the cards that care are also Warriors themselves. Ideally, the as-fan of cards like this in the set would help players care more about a lot of different types.

Adventure

TLDR: A popular returning mechanic with unique twists is a recipe for high confidence.

At their mechanical core, "Pentagon" adventurer cards are not unique from anything that Throne of Eldraine did—cast side A, then cast side B later, or just cast side B now. Creatively, the concept behind them has shifted. In Throne of Eldraine, each card told its own story about the creature on side B. In "Pentagon," each card tells a story about the player's commander.

To accomplish this, side A should always represent an inciting incident or interruptive event within the hypothetical life of a commander, while side B represents the antagonist, how the commander triumphed, or the reward received for their success. To accomplish those last two, many of the Adventure side B cards aren't creatures, using a variety of card types.

  • Artifact: Loot or a puzzle
  • Aura: Magical reward or a skill of the commander
  • Creature: A boss they defeat or an NPC they encounter
  • Enchantment: Magical reward or a skill of the commander
  • Equipment: Loot
  • Sorcery: A feat or skill of the commander

These could be culled. Equipment, artifacts, and Auras have felt particularly resonant, creatures work just as good as always, and sorceries might be next best for how they support spellcast synergies and feel like they're actually doing something. Instants are inherently problematic because they sit face up and demand opponents' attention.

Melee

TLDR: Handy ability for tokens, not great on cards.

It's very useful to be able to make meaningful tokens in Limited formats, but a token that's meaningful in Commander, with four players each sitting at double the usual life total, has some unique demands. Can't make huge tokens at efficient rates, since that would be weird next to Standard cards and possibly even impact Legacy. "Pentagon" has been using the melee keyword from Conspiracy: Take the Crown as a flavorful way to give simple Soldier tokens a little extra body.

Creatively, these tokens represent the Flaming Fist, which is technically a mercenary army employed by the patriars of Baldur's Gate to keep the peace. They could have the Mercenary creature type.

Goad

TLDR: Confident the set wants this.

Goad's gameplay in multiplayer Limited and Constructed formats adds so much fun and opens unique payoff space for traditional mechanical structures that it should be considered evergreen in any multiplayer set. I would protest its removal, and the set probably has room for more of it.

Prowess

TLDR: Take it or leave it.

Prowess was added to the set to supplement the spellcast and "surgecast" mechanics in blue. It's not doing anything particularly great for gameplay or draft strategy and could be removed from the set. However, it's nice to have a creature ability worth about half a mana for balancing purposes and simple enough that it's not costly to keep. A bonus is that the word itself can easily apply to most D&D classes to imply expertise and could possibly be spread across five colors with color pie approval.

Gods

TLDR: They're here, but what makes them divine?

We've always expected to have several gods relevant to the Time of Troubles in the set: Bane, Bhaal, Myrkul, Cyric, Mystra, and Lord Ao at minimum, with a variety of others also available to use (Torm at the front of that list). Standard has represented divinity a few ways, but always tends to use some persistent durability (Theros, Amonkhet, War of the Spark, Theros Beyond Death). They don't need to lean on that as heavily in Commander, since the command zone itself can provide the foundation for immortality. The current file has a simple cost-reduction mechanic trying to represent that dedicated worship makes these gods stronger and easier to summon over and over.

It's worth noting that during the Time of Troubles, all these gods (except Ao) become mortal, so they could all be represented as class creatures/demigods, but that would ultimately be less satisfying and likely mismatches expectation for the people excited to play as these characters.

Gates

TLDR: Might as well make a few, but they can't bear weight.

Baldur's Gate has Gates! However, the as-fan required for Gates and synergy cards to make "Gates matter" cards impactful in a 360-card set crowds out a lot of meaningful other cards. It's a nonstarter, beyond letting a few cards that care about basic lands also care about Gates.

Discarded Mechanics

Skilled/Feat

Skilled/feat was a Commander mechanic for enabling players to add a color to their monocolor commanders. The commanders had a keyword, and nonlegendary cards in the set had monstrous abilities that you could add to the commander and activate to give them new abilities.

The crux of skilled/feat's issues was that crossing the mechanical functionality of the feats across the expendable nonlegendary creatures and the persistent commander was a massive amount of rules baggage to store on both cards. From a templating standpoint, it wasn't viable to get the "once per game" feat for the commander, and the "once per instance" feat on the normal creatures.

Ultimately, playtesting feedback was neutral to negative on the mechanic, often comparing it unhappily to monstrosity, adapt, and the commander tax itself. The character-building element was received well but not actually enjoyable in the game.

Spell-Pair

Spell-pair was a commander mechanic for enabling players to add a color to their monocolor commander by starting the game with an instant or sorcery in the command zone that they could cast once per game, possibly under a specific condition.

The mechanic's biggest issues concerned the kinds of cards that could go into the command zone being very narrow. For example, almost every instant was off limits, since it would hang over the game when the player had open mana. Sorceries that traded for a resource were problematic since players would be reluctant to run their cards into the face-up trick.

The most promising version let the commander "pick up" the Adventure part of a card but places too large a constraint on Adventure executions and breaks the creative paradigm a little bit. It did let the mechanic hook into older cards, which is an admirable quality if balanced.

Partner

Partner works—Commander Legends proved it—but as the second set in this SKU, it's important to avoid setting the precedent that Commander Legends is synonymous with partner. It should be close to the last resort—even just trying ~70 multicolor legends might be a better route.

"Enter the Upper City"

The precursor to "pick up the torch," this mechanic was loosely inspired by King of Tokyo-esque battles for dominance and the three-tiered structure of Baldur's Gate itself. Through combat and keyword actions, players would move up or down a map of the city, accruing rewards for elevating their status to the top and maintaining it there at the expense of other players. However, the incentives proved too challenging and awkward to balance convincingly, and the enthusiasm for the torch makes it unlikely the set should return to this mechanic.

Monstrosity

Initially, the team was enthusiastic to pair Adventure and monstrosity as two mechanics emblematic of D&D. The mechanics for the set continued to play in space closer and closer to monstrosity—one-time activated abilities—and the creative for the set became more urban over time, reducing the amount of space monstrosity could occupy. In the end, more Adventures can serve this role, so monstrosity is redundant. Removing it also ensured the set had enough room to make classes matter.

Draft: Set Structure and Archetypes

Rarity Distribution

  • 141 Commons
  • 120 Uncommons
    • 30 Uncommon commanders
  • 77 Rares
    • 25 Rare commanders
  • 22 Mythic rares
    • 6 Mythic rare commanders

The legends are being called out separately because they're collated separately. Two legends appear in each pack with an uncommon appearing ~73%, a rare ~24%, and a mythic rare ~3% (per slot).

Collation Per Pack

  • 13 Commons
  • 3 Uncommons
  • 1 Rare or mythic rare
  • 2 Commanders (mixed rarity)
  • 1 Premium card (mixed rarity)

Additional fixed collation is an option, such as defining two or three cards in the set for each legend as "complements" that would always appear in packs containing that legend. The purpose would be to make it even easier to take the first two picks and begin drafting on rails. However, making Background draftable would mean players are less likely to take this path and more often pick a legend and a Background or a card that synergizes with one. Any Background implementation likely affects additional collation modification options.

Color Themes

Most sets use color pairs, but Commander Limited is a unique beast in managing color commitments. To that end, the set was structured around monocolor themes and how they intersect. The expectation is that players will most often commit to one color, then find their way to a second.

White

  • Creaturefall
  • Tokens

Blue

  • Spellcast
  • Cast two or more spells

Black

  • Sacrifice
  • Recursion

Red

  • Exilecast
  • Treasure

Green

  • Ramp
  • 4-Plus power matters

As the set has evolved to include more mechanics, especially class, the focus on some colors has shifted toward a more singular mechanic. White cares a lot more about just going wide for class and making tokens, blue isn't as focused on surgecast, black wants to sacrifice a bit less often, red still makes Treasure but cares less about it, and green doesn't have much going on with 4-plus power anymore. The set still works fine with this primary/secondary mixture, but there's room to make changes here.

This structure could be removed and go back to traditional color pairs, and the set would work. However, one element of Commander Legends that the set evolves upon is creating a draft format where the best strategy for the player isn't to just get on rails as fast as possible. The Background mechanic lets a player sit on one color and dabble in the rest quite easily, so if our color structure can support that, we're most of the way there. Most Commander Legends's color pairs don't have loads of crossover synergies, but ours can if every strategy can manifest differently for each pair.

Note: Commander Legends restricted people to two colors often. In playtesting, three colors hasn't felt problematic thanks to the set's emphasis on Treasure and its use across all five colors, as well as the Torch including multiple early rooms that fix mana. I personally believe drafting three colors is likely not very hard, and we could facilitate it more strongly if we wanted to do so.

Primary Class Colors

This is how the classes are "centered" within the set. This isn't a limitation—feel free to put classes in any color that's appropriate for the creative and mechanical identity—but it's important that these colors resonate as part of the identity for each class to feel diverse, or that the distribution be adjusted to do so another way. They've been selected to specifically minimize mechanical overlaps and provide enough diversity across colors for classes.

  • Artificer: Blue
  • Barbarian: Red
  • Bard: Red-green
  • Cleric: White-black
  • Druid: Black-green
  • Knight: White
  • Monk: White-blue
  • Ranger: Green-white
  • Rogue: Blue-black
  • Shaman: Red
  • Warlock: Black
  • Wizard: White-blue
  • Warrior: Red-green-white

These identities can be changed and don't even perfectly match the set right now; they're just foundations to help build equitable distributions. It will be worthwhile to map creature races the same way before concepting starts.

Commander Constructed Goals

Legends

The set will include 60–80 legendary creatures by the finish. The file currently supports the lower end of this number thanks to how forgiving Background is as a color-fixing mechanic relative to partner.

The current roster of legends has been sufficient for mechanical playtesting but aren't close to final and lack a lot of the novelty and charm that this product should aim for. They do include a handful of relevant characters from lore. Those names all include commas as a weak but simple distinction. The character grid developed by Mike Mearls should be used to continue filling out these slots, with a focus on characters related to the Baldur's Gate video game series.

The set will provide tons of humanoid characters to the game, but it will be important to preserve some room for monsters to be used as commanders, too.

Reprints

Commander Legends was supported by a variety of reprints, ranging from Constructed-unplayable to high-demand cards. "Pentagon's" new-to-reprint ratio is very different, and its Constructed goals have shifted to match. Ideally, the new cards in the set all have some applications in the larger world of Commander, even if it's just casual Treasure theme decks or whatever.

Dual Land Cycle

The set includes a ten-card cycle of new dual lands. Dual lands aren't necessary, and a full set of ten isn't either. However, they are an easy route to building desirability. To succeed in their goal, they should feel sideways from Battlebond lands—better lands are unnecessarily strong, but worse lands would upset players as a lost opportunity to get more Battlebond lands. The current set has weaker rules text, which is almost a necessity given the format, but land types to make up for that.

Again, there are some things you're just not meant to see yet. You're not ready for it. Maybe someday. But not today.

These slots could become Battlebond lands if creative thinks those names can be manipulated into viable concepts for Baldur's Gate.

5XX Set Guide in Excel

This is an Excel tool to help manage "Pentagon," specifically the structural elements and how they relate to as-fan. It has some color coding to avoid accidentally editing formulas.

Yellow: These cells need manually inputted data.

Green: These cells calculate their own data from other cells.

Orange: This is the information you want to know.

Some headers inform the calculations of other cells but aren't yellow because there's no need to change them. Here's what all the sheets are for!

  • [Rarity] sheets: Each of these is just a general look at cards of that rarity. It served as a guide for building the set, but the rules aren't hard and fast.
    • Adventure makes it easier to mess with the ratios.
  • Color pairs: Breakdown of the Limited themes and some evaluation of how well each color combines with another's theme.
  • Cycles: Just a list of how some stuff is cycled throughout the set. Nothing too exciting here.
  • As-fans: This sheet is where the sausage gets made.
    • Modifiable collation scheme that's used as a reference for the rest of the sheet. Note that it doesn't have a premium slot.
    • As-fan calculator to figure out that value for any given rarity distribution.
    • Percentage of commons/uncommons that are creatures, accounting for uncommon legendary collation and the modified target Commander Legends used. Each multicolor creature counts as 0.5 for each color.
      • The target for these ratios is based on what Jules did in Commander Legends; a 15% bump from typical booster sets. That modifier is in Y7.
    • Table for as-fan calculations based on the rules text of the cards (using hidden ISNUMBER(SEARCH functions in RAW) that also accounts for legendary collation.
    • As-fan calculator for every class in the set. It auto-populates every field based on the RAW and also accounts for legendary collation.
  • Class distribution: A color-based visualization of how classes are distributed across colors and commanders. No inputs on this page; everything is automated.
    • Nothing here accounts for the color identity of multicolor commanders, but the same ISNUMBER(SEARCH trick other sheets use could.
  • RAW: Pasting an export of the set with the correct fields here will populate the rest of the spreadsheet.
    • Be mindful of the ISNUMBER(SEARCH columns from O onward and stash additional ones at the end.
  • Class planning (defunct): Essentially the original scratchpad for planning how to introduce classes to the set. Not worth interacting with much, but a good reference tool for how it was done the first time.
  • Zendikar Rising party math: Just a slapdash reference for comparing the "Pentagon's" color distribution and as-fan to Zendikar Rising's party.
  • Old ideas: An old list of popular and usable mechanics, just jotted down.

Finally, we come to the end. I hope you enjoyed having a chance to see a different take on a vision design handoff document. A big thanks to Glenn for letting me show this off. As always, I'm eager for any feedback on this column or this document. You can email me or contact me through any of my social media accounts (Twitter, Tumblr, Instagram, and TikTok).

Join me next week for the beginning of Double Masters 2022 previews.

Until then, may you enjoy trying a new version of an old thing.

 
#937: CLB Vision Design with Glenn Jones
#937: CLB Vision Design with Glenn Jones

29:57

I sit down with designer Glenn Jones top talk about the vision design of Commander Legends: Battle for Baldur's Gate.


 
#938: Double-Faced Cards
#938: Double-Faced Cards

32:01

For this podcast, I look at the history and design evolution of double-faced cards.

Latest Making Magic Articles

MAKING MAGIC

July 4, 2022

More Life Lessons by, Mark Rosewater

Back in 2006, I wrote a two-part article called "Life Lessons" (Part 1 and Part 2) where I examined different lessons I'd learned in life and then explained how I applied those same life ...

Learn More

MAKING MAGIC

June 27, 2022

A Double Take, Part 2 by, Mark Rosewater

Last week, I started telling some card-by-card design stories about cards reprinted in Double Masters 2022, but there were a lot of stories to tell, so I'm continuing today. Body Double ...

Learn More

Articles

Articles

Making Magic Archive

Consult the archives for more articles!

See All