Hording all the Fun

Ever want to sling some cardboard but don't have anyone else to play with? Want to play with a bunch of friends instead of against them? Did you enjoy the Face the Hydra decks from Theros? If you answered 'yes' to any these questions then continue reading, my friend!

Introduction to the Horde Format

Horde is a variant of Magic in which a group of players team up against an autonomous deck (from here on out referred to as the survivors and the horde, respectively). It was first introduced here, and had a follow-up article here. Additionally, the Serious Fun column on the Wizards of the Coast site did an article about the Horde format here.

In general, the horde deck consists of a fixed number of cards (typically between 100 and 200), with a 60%/40% split of tokens/non-tokens, although many players adjust their decks to suit their individual needs. Zombies are the most commonly used horde tribe, but many people use other tokens, such as Dragons, Angels, and Slivers. You're limited only by your imagination (and Magic tokens, of course). This deck plays against the survivors in a two-headed giant fashion where the each team takes their turn as a group, so the horde takes its turn then all the survivors take their turn at the same time. Losing the game is easy - drop to zero life or any alternate win condition you've set up in your horde. Winning the game is a bit trickier: you have to mill the horde's library, empty their hand if they have one, and clear its board of all creatures.

The beauty and appeal of the format is that because the Horde is not controlled by a person, everyone is either a winner or a loser. Additionally, it can be played as a one person solitaire deck, or as many as you want (although much like regular Commander, 3-4 people is generally the perfect number).

The Basic Rules

These rules are taken from the above-mentioned articles, as a default rule set, along with the changes that I've made when building and playing my own deck. They are presented as a basic guideline, and you are encouraged to change them as you see fit.


Each survivor contributes 20 life, and 25 cards are removed for each player under four. (So with a typical horde deck and three survivors, they would start at 60 life and face a horde deck with 75 cards.)

(Authors note: I do not like this rule as I have found that it makes it easy for all but the weakest of Commander decks. I prefer using a life total of 30 life +10 life per survivor beyond the first, which caps out at 60 life. A 100 card horde deck should not be lessened below 100 cards, as I've found that even goldfishing a 100 card horde deck against 2 of my own decks has proven to be easily doable with a starting life total of 40 life.)


  1. The survivors always go first, and take the first three turns before the horde gets a turn. The survivors do not get to draw on their first turn.
  2. At the beginning of the horde's turn, it reveals cards from the top of its library until a non-token card is revealed. First, the revealed card is cast, then the revealed tokens are cast. (Author's note: I change it so that the tokens enter the battlefield as a special action not using the stack, so that the players can't interfere.)
  3. All creatures have haste, and must attack each turn if able. The horde has infinite mana, and will pay all costs for cards such as Propaganda or Rhystic Study.
  4. If the horde would lose life, it instead mills that many cards from its library into its graveyard.

Winning the Game (all conditions must be met):

  1. The horde must have no cards in its library.
  2. The horde must have no cards in hand.
  3. The horde must have no creatures in play.

There are some other things you should take into consideration when building a Horde deck or playing against it as a survivor. The horde is supposed to be as autonomous as possible, so try to limit the number of cards in the deck which require decisions or dexterity (this will also help make the game go faster). Examples are cards which require targets such as Aether Adept, cards which require specific choices such as Grave Pact, and cards which cause you to repeatedly mess with the board state such as Cathars' Crusade. Optionally, you can create custom rules to deal with these types of cards, but you should limit them because the added rules to track can become cumbersome. Additionally, because the horde generally only has the option to attack, as a survivor you shouldn't play "cheating" cards, such as instant win combos or effects which prevent attacking, as the horde probably will not be able to interact with these types of cards and effects. Lastly, certain cards can become "oops I won" cards for the horde under the right circumstances, most notably Coat of Arms. These types of cards should be used sparingly because the goal is for the horde deck to be a challenge, but still overly difficult to beat.

You also want to have an idea of the challenge rating for your horde deck. A deck that is too difficult to defeat or one that is easily defeated is not fun and won't lead to repeat usage. There are four main factors to consider when determining the difficulty of your deck:

Deck Size: A larger deck requires more damage to defeat, and will be able to go for many turns, dropping creatures each turn without a moment for the players to catch their breath. This can be adjusted by making a larger or smaller deck, or by removing cards to fit the number of players.

Ratio of tokens to creatures: Since you will continue to flip tokens until you hit a nontoken creature each turn, the higher the ratio, the more creatures (and the more total power) you'll have attacking every turn. You can adjust this by changing the ration either higher or lower as desired.

Difficulty of killing the monsters: If you have mostly 1/1s and 2/2s then you can easily absorb a few hits in order to save your smaller utility creatures, but larger creatures with higher toughness are harder to kill off. Similarly, if you include lots of creatures with keywords such as flying, regeneration, menace, or indestructible, you will have a harder time killing the monsters. Think of it like the mana curve of your deck: you want the bottom end to be heavy rather than your top end.

Power level of creatures: This was touched on above, but worth stressing again. If your deck contains a high ratio of large powerful creatures then it makes it more likely that an unlucky flip could see lots of damage coming at the party all at once. IT is for this reason that you want a wide spread of creatures with differing power/toughness, focusing on the lower end rather than the higher end.

Building a Dungeons & Dragons themed Horde deck

I previously built a zombie horde deck (found here), which is themed around a Night of the Living Dead-esque zombie outbreak. For this horde deck, I wanted to simulate playing a D&D campaign where the survivors are an adventuring party partaking in a quest, battling monsters, finding treasures, and running into traps along the way. Before I began building the actual deck, I wanted to create some custom rules to help guide the flavor. Here's what I came up with:

  • There are two decks: a main horde deck, and a treasure deck, which you draw from when a treasure card is acquired through the horde deck (think drawing a Community Chest card in Monopoly)
  • On the horde's turn it reveals cards until a nontoken creature card or a trap card is revealed.
  • Treasure cards drop attached to the last creature to enter the battlefield (unless they are the first card revealed for turn, in which case continue to draw until a monster is revealed). Similarly, creatures which create Gold tokens upon entering the battlefield have those tokens "attached" to them. They remain on the battlefield once the creature has died.
  • Players can skip either their draw step, main phase, or combat phase in order to perform a special action: move a gold token from the horde's battlefield to their own, take a treasure from the battlefield (and draw from the treasure deck), or transfer a treasure to another player.
  • Gold can be split among the players as desired
  • "Spells" go into one player's inventory to be used later. They have levels and can only be used once the party is that level or higher. Casting a spell is free, you don't need to use mana to cast them. On the off-chance that the cmc matters due to a player's deck, use the printed cmc.
  • The party gains a level each time an experience counter is revealed in any zone.
  • Ignore printed text on treasure and trap cards, using modified text as applicable.
  • Trap cards all have a DC check of 15 minus the party, dealing half of the effect rounded up on successful saves. They ignore hexproof and shroud.

My first step was to search Gatherer by creature type to look for creatures that could have come straight from a Monster Manual: we're talking skeletons, oozes, ogres, and of course, dragons, just to name a few. I automatically disregarded anything which had enter the battlefield, activated, or triggered abilities which I wasn't going to want to track. This proved to be a very limiting factor, as I also didn't want to include too many of any one particular card. After all, an encounter with the same monster over and over quickly gets boring.

Next, I conducted a similar search looking at all the available tokens, again looking specifically for creatures associated with Dungeons & Dragons. When multiple options were available, I tried to get a mix of different power and toughness stats, focusing more on weaker tokens. After all, multiple tokens can enter the battlefield at the same time, and it would be a very short encounter if the party ran into five 6/6 dragons.

Finally, I searched for flavorful trap cards, treasure for the adventuring party to find, and spells for the party to cast. This was more difficult, because there wasn't one magic keyword to search for. For the traps, I started with the Trap subtype cards introduced in Zendikar. Most of them were not what I was looking for, but there were a few usable ones. There were a couple of old artifacts that were traps, so those naturally had to be included, and a few spells rounded out this area. For treasure I started looking at equipment and picking out a variety that a beginning party could expect to find or have. As much as I wanted to include cards like Sword of Fire and Ice and other great equipment, I was trying to keep this deck on a budget. These were rounded out with some gold tokens which I wrote 1d4, 1d6, 1d8, or 1d10 on so the party could randomly determine how many gold tokens they received. Lastly, for the spells I went through the Dungeons & Dragons 5th edition spell list to see what popped out as Magic cards, either directly in name or in flavor, then searched to see if they would be appropriate for the deck. I also wrote a spell level on each one, one through four, to denote when the spell could be used.

What I should have done (and highly recommend for a horde project and building any deck in general) is compile a full list on paper, and THEN order cards. That makes sense, and it helps save on shipping costs. But of course, that's not what I did. Searching for cards was a drawn out process and I did it over the course of several days. As I was going along I was adding cards to my cart and at each stopping point I'd place the order. Because of this I wasn't really paying a lot of attention to the quantity or distribution of the cards, and would just throw everything into one large pile until I had everything. Once I was ready to build I did an inventory of how many cards of each type I had. I ended up with roughly 120 tokens, 80 creatures, 10 traps, and 35 pieces of equipment. I decided to make my first draft a ratio of 100 tokens, 75 creatures/traps, 21 treasure cards (which draw from the treasure deck), and four experience cards since the spells in the treasure deck go to level four. The treasure deck itself was 40 cards. This gave me an average of between one and two creatures revealed per turn. That seemed low, but testing would determine how to make changes.

The Decks

Individual Card Rulings and Errata

I took a sharpie to each card in the deck in order to make the following changes to these cards. It should also be noted that any creature which can regenerate will use that ability.

Alabaster Potion: X= 8d4 + party level
Booby Trap: Deals 10 damage to each Survivor
Brass's Bounty: Each Survivor gets the effect
Brutal Hordechief: Ignore the activated ability
Disintegrate: X= 1d6 + party level
Divination: Each Survivor gets the effect
Fireball: X= 8d6 (party can divide X as played normally, when cast as trap make one save throw)
Gold Tokens: 2x 1d4, 2x 1d6, 1x 1d8, 1x 1d10
Golgari Grave-Troll: Ignore dredge
Healing Hands: Each Survivor gets the effect
Hideous Laughter: Only affects Survivors
Mossbridge Troll: Ignore the second ability
Needlebite Trap: 5 damage to each Survivor
Pit Trap: Affects all Survivors' creatures without flying (half of them rounded up on successful save, Survivors get to choose)
Ravenous Trap: Affects each Survivor
Runeflare Trap: Affects each Survivor
Three Wishes Each Survivor gets the effect
Wall of Fire: Any Survivor may spend R to pump wall

Traps I Encountered

During the building process I came across a few concerns which I will note as I test the deck a few times, and for you to take into consideration as your build your deck. First, as I already mentioned, the token to creature ratio is very low. This means that on average I am expecting to only flip one or two creatures a turn. This will probably make for a boring dungeon crawl because the threat of a large ambush is low. I will most likely lower the creature count a bit more. Second, the equipment seems to range from mediocre to decent, with not much in the way of anything very exciting. This was somewhat expected because of budget constraints, but I will keep an eye out for better equipment that still conveys the flavor. I have a similar concern for the spells. Lastly, the spell levels feel arbitrary and uneven. I began using the actual spell level from the Players' Handbook, but halfway through I realized that D&D spell level didn't correlate into the spell level rating I was using, making spells with small effects have higher levels. I think I will redo the spell levels based more on the overall impact on the game and expected time they will be used while playing.

Play-testing Feedback

I played the horde deck using my Marisi, Breaker of the Coil deck, along with two other players running Grumgully, the Generous and The Ur-Dragon, respectively (Ur-Dragon switched to Syr Gwyn, Hero of Ashvale for the second game). We used the full horde deck and started at 60 life. You can watch the a quick introduction to the deck and the game play from both games here. In both games my prediction about the creature to token ratio affecting the difficulty rating played out. We never got any back-breaking flips, with the worst turn being a handful of small creatures and a 5/5 Giant or 6/6 Dragon. Booby Trap hit us and with two of us making our save throw and the third failing we decided that taking a 20 hit point blow was pretty brutal and changed it to just one instance of damage. (In hindsight, I think leaving it as I had originally hitting each player is correct because the rest of the deck was so weak and the odds of hitting it are low anyway.) In the first game Ur-Dragon got a rocky start and missed a few land drops, but Grumgully and Marisi were able to hold their own just fine - we probably could have defeated the horde without help from Ur-Dragon, which would have made for a more balanced challenge.

Overall, both games played similarly - we were able to get chump blockers online early on to absorb some damage during the first few turns and then once we each had board presences we could safely hit the horde for anything from 20 to 50 points of damage a turn. In both games Scourge of the Throne proved to be game ending with its dethrone trigger giving us an extra combat step, but we concluded that our board states were such that even without the extra combat step we still would have beaten the deck on the following turn regardless. Because of how slow the deck played the correct play was usually to just swing, occasionally leaving behind a 1/1 to chump block just in case.

The other thing we discovered is that no one really wanted to take the treasure tokens as they fell onto the battlefield, since the correct play was usually to just build up your own board state. I took the only one that dropped on the first game, which was a useless Death Ward, and on the second game we actually had one creature get two treasures, so Syr Gwyn and I split them, Gwyn getting an aptly flavorful Hero's Blade and me with a Behemoth Sledge. We both skipped our main phase so we couldn't equip them, and the following turn we were swinging for lethal, so the extra equipment was overkill at that point. Flipping experience counters was cute and a flavorful way of adjusting saving throws, but it was also additional mental baggage to track for little reward. Perhaps the correct thing to do is to make the entire treasure deck aligned with experience (including the equipment) and to add more traps.

If/when I go back and make changes, I want to do a few things before playtesting again. The first is to alter the token to creature ratio. I think I'll make the actual creatures more powerful and fewer, while focusing on tokens to help flesh out the swarm that you get in a D&D encounter. Secondly, I'll edit the treasure deck to make the equipment and spells more desirable to hopefully give more of an incentive to take them. Lastly, I want to make the traps more potent, since most of them got milled. I also think that ending the horde turn only after flipping a creature will make its turn play out stronger. We had a couple of turns where the only thing that flipped was a trap, and while getting hit for an average of more damage than normal hurt, it made it easier to defeat the horde because milling the deck from damage was stronger than worrying about a creature in play.

In Conclusion

Horde is a fun and relatively inexpensive variant deck to build and break out from time to time. Much like Commander, it can be tweaked and customized to suit your goals and power level as well as have its own personal theme. And since there is no metagame to consider, you don't have to constantly make changes once you are satisfied with the deck (although there is always the joy in finding a new card hidden among the latest set release). It's something I would encourage everyone to give a try at least once.

Thanks for taking the time to read this, and please leave your feedback in the comments. I'd love to hear about the horde deck you built or want to build. Are there any amazing (and cheap) cards that I completely missed for this deck?

User avatar
hmm, interesting concept.
User avatar
Oh my, I played this like an year ago, it was ultra fun! The Sliver variant was insane.
User avatar
This format needs to take off. I need to post my list, too.
We play Horde all the time... we changed from these rules to the WotC version...
Flip N+1 cards a turn (where N is the number of players), play all tokens and cards flipped.
Players have 20 life EACH, unblocked creatures do damage to ALL players. (we found lifegain to be too strong with a unified life total)
we also added a bunch of custom cards with Heroes' rewards on them (see the WotC challenge decks for examples) and find it works MUCH MUCH MUCH better.
I would encourage you to either switch to the N+1 method or to declare 'all commons are tokens' and then use that to flesh out the decks. Dragon Fodder for example makes a great token (individual goblins are too high variance, but small packs of 2 or 3 are better).
You must Login or Register to comment.