What Is Format Coconut? Lorcana’s Multiplayer Beta, Explained

Written August 23, 2026, last checked August 27. The format is in open beta and the coconuts get adjusted constantly, so parts of this page will go stale. I date everything and fix it when it breaks.

Format Coconut is Disney Lorcana’s multiplayer format, currently in open beta. Three or four players sit at one table and race to 25 lore. You build with one copy of each card, and the whole thing hangs off your “coconut”: an ability you pick in the official app that stays on all game and lets you break the one-copy rule for a single named character.

If you’ve played Commander in Magic, you already know the shape. A big pile of one-ofs, a free-for-all table, and one card that decides what your deck is about. If you haven’t, no problem. The rules fit on a napkin, and the rules aren’t the fun part anyway. The fun part is what a full table does to a game built for two people.

Quick context on why this page exists: I started playing the beta, wanted decklists, and couldn’t find anywhere that had them collected in one place. So I built my own, one per coconut, and the notes turned into a site. That’s the whole origin story. More on that on the about page.

The essential rules

• 3 or 4 players, free-for-all. First to 25 lore wins, not the usual 20.
• Minimum 60 cards. Sixty is a floor, not a target.
• One copy of each card, with one exception below.
• Up to 3 inks, and one of them has to match your coconut’s ink.
• Your coconut is a digital card in the Lorcana app. It never goes in your deck or on the table, and nobody can target or remove it. It’s just an ability that’s yours all game, and it steers everything you build. There are 18 coconuts in the beta, 3 per ink.
• The exception: each coconut names a specific character card, and you can run up to 4 copies of it. I read “up to 4” as including zero, so nothing forces you to play the named card, though most coconuts give you a good reason to.
• Everything else is normal Lorcana.

Two gaps the beta rules document leaves open, in case your table hasn’t settled them: I play with the first player skipping their turn-one draw, which is the usual multiplayer convention, and as far as I can tell the format has its own banlist, currently empty.

Also worth saying out loud: “Coconut” is a codename. The final name isn’t decided and the format is digital-only for now. I built a site around a name that might disappear, which is either early adoption or a mistake, and I’ll find out which.

Why a full table changes everything

The rules above look like small tweaks to regular Lorcana. In practice they rearrange deckbuilding from the ground up. Three ideas keep coming back in everything I write here.

Burst lore is the currency

At a four-player table, nobody can police everybody. In 1v1, a slow grind of lore advantage wins games. In multiplayer, the player who quietly assembles one 8-to-15-lore turn wins out of nowhere while the other three are busy with each other. So the question I ask about a deck isn’t “how much lore per turn on average” but “what does its explosive turn look like, and can the table see it coming.”

“Each opponent” effects roughly triple in value

A card that drains 1 lore from an opponent is fine in 1v1. The same card draining 1 lore from each opponent is a 3-lore swing. Anything that scales with the number of opponents gets repriced way up, and there are around a hundred cards in the pool worded that way. Single-target removal goes the other direction, because banishing one character answers a third of the table’s threats and costs you a whole card plus your tempo. Board wipes and pings that hit everything do the job that spot removal does in 1v1.

One copy of each card makes drawing your best resource

You can’t count on drawing your best card when there’s exactly one of it in there. What you can count on is drawing a lot of cards and running several cards that do the same job. Consistency comes from density of effects, not from playsets. That’s why my deck techs are obsessed with draw counts, and why I run the actual probabilities instead of eyeballing them. The math has embarrassed my first impressions more than once.

There’s also a softer thing that 1v1 players tend to underrate: politics. Some coconuts help the whole table. Some paint a target on your forehead. Evaluating a coconut means guessing how three other people will react to it, and that guess is part of the deckbuilding.

The 18 coconuts at a glance

The full roster as it stands in the beta, three per ink. This is just a sketch of what each one wants to do. The tier list is where they actually get compared and argued about.

Amber

• Ariel — Spectacular Singer: gain lore whenever a Princess of yours sings a song, equal to her lore value. A songs-and-Princesses engine.
• Pocahontas — Peacekeeper: once per turn, a chosen character gets +1 lore, can’t challenge and must quest. It works on anyone’s characters, including opposing ones, which makes it much more flexible than it first reads.
• Stitch — Rock Star: once per turn, play a character costing 2 or less for free; if it’s a Lilo or a Stitch, a chosen character gets +1 lore that turn. Adjusted in a mid-August update.

Amethyst

• Dumbo — Ninth Wonder of the Universe: your characters can use their exert abilities the turn they arrive. This unlocks a pile of cards that were simply too slow for 1v1.
• Snow White — Merry as the Morning: once per game, if you can show a Snow White plus 7 or more differently named Seven Dwarfs across hand, discard and play, all your characters get +2 lore for the rest of the game. A build-around puzzle, and a hard one.
• Winnie the Pooh — Hunny Wizard: pay 1 ink to draw a card whenever you play a character with no ability. Vanilla characters become a draw engine.

Emerald

• Donald Duck — Fred Honeywell: pay 1 less for Boost characters and locations and for Boost abilities. Buffed in a mid-August update.
• Robin Hood — Sneaky Sleuth: start with a free Robin’s Bow from outside your deck, and deal 1 damage whenever you play a character named Robin Hood. Reworked in early August.
• Ursula — Deceiver of All: your characters count as +1 cost for singing, +2 if named Ursula. A songs coconut coming at it from a completely different angle than Ariel’s.

Ruby

• Mickey Mouse — Brave Little Tailor: every Mickey Mouse card in your hand, deck and discard gains Shift 2. Cheap Mickeys turn into launching pads for expensive ones.
• Mr. Incredible — Super Strong: Super characters gain Rush the turn they’re played, and you may exert a weaker opposing character. The classification requirement drafts half your deck for you.
• Sisu — Emboldened Warrior: characters with more strength than each opposing character can quest the turn they’re played. The printed text says “all characters”, not “your characters”, and most tables I’ve seen play it as helping everyone. Check what your table does.

Sapphire

• Moana — Curious Explorer: with a Moana, Heihei or Pua in play, you may ink an extra card each turn. Plain, honest ramp.
• Mufasa — Ruler of Pride Rock: once per turn, pay 5 ink to put your top 2 cards into your inkwell. Explosive ramp.
• Nick Wilde — Wily Fox: run 4 copies of Pawpsicle, and once per turn banish 4 of your items to gain 4 lore. Items into lore.

Steel

• John Silver — Greedy Treasure Seeker: your locations gain Resist +1 for each character there. A locations deck whether you like it or not.
• Scar — Finally King: your first Ally character each turn costs 1 less. Another classification-driven build.
• Tinker Bell — Giant Fairy: whenever another of your abilities or actions damages an opposing character, deal 1 more damage to it. A ping amplifier that gets scarier with every extra opponent.

Reading down the list, you can see a tension the beta is still sorting out. Some coconuts are strong passives you’d happily play without ever drawing the named card. Others are mostly an excuse to run four copies of a great character. The recent adjustments look aimed at making you actually want the named card in your deck, and where each coconut sits on that spectrum is a thread that runs through all my deck techs.

This is a beta, and it moves fast

Format Coconut opened its beta in late July 2026, and the coconuts have already been through several rounds of changes: Robin Hood and Mufasa in early August, Donald Duck and Stitch in mid-August. That pace is the point of a beta, and it means any analysis of the format has a shelf life measured in weeks.

That cuts both ways for a site like this. Everything carries a date, and something written before an update may describe a coconut that no longer exists in that form. But the churn itself is worth writing about: what a rework fixed, what it broke, and what it says about where the format is going.

How I do this

Small operation, simple method, and I’d rather show the method than pretend to be an authority on a format that’s a few weeks old.

• Math instead of vibes. When I say a combo fires 43% of the time by turn 7, there’s a calculation behind it and I’ll show the assumptions so you can pick them apart.
• Dated guesses, openly revised. Tier placements are hypotheses with a timestamp. When an update or a better argument kills one, I say so and change it.
• Community chatter as a sample, not consensus. I read what beta players are discussing, but any one channel is a narrow window. When my reading differs from what’s circulating, I say why and what would change my mind.

From here, the natural next stop is the tier list, where all 18 coconuts get graded. After that come the deck techs, a couple per week: a full deck for each coconut, with a 4-player game plan and labeled flex slots. All 18 are already built, and I have played each of them at least once, so the order is up for grabs.

I’ve been playing and building since the beta opened, but the format is weeks old and nobody has it solved, me least of all. If you’ve played Format Coconut and something here doesn’t match what happens at your table, tell me. What did I miss?

Last checked against the official coconut roster and beta rules: August 27, 2026.