Judge's Tower Design

If you're unfamiliar with the Judge's Tower format, read this first.

People often ask me how to build a Judge's Tower.In fact someone messaged me to ask this exact question while I was in the middle of typing up this article. What cards should you include? Is there an agreed-upon list? Is anything banned? Well, no. Judge's Tower is an informal format, and you can do whatever you want with yours.

If you want to use someone else's list, you can find some here, here, here, here, here, here, here, here, here, here, and here. But I think it's a lot more fun to build your own. You can tweak it to match your personal playstyle, you can put it pet cards, and you can avoid shelling out any significant money by using cards from your existing collection.

Different people are designing their Tower for different playgroups, so ultimately the answer to "how do I built my Judge's Tower" is "however is the most fun for the people you play with". That said, most people who play Judge's Tower want similar things out of it, and there are some best practices that we can use as a starting point.

Before designing anything, we want to know what our goal is. What do people want out of a game of Judge's Tower? Fun, obviously, but there are many different types of fun. Magic, Chess, Candyland, and Baseball are all games played because people find them fun, but that doesn't mean their players want the same things out of them.

What do most people want from Judge's Tower? Ultimately, Judge's Tower is for people who know a lot about the game and want to improve and/or show off that understanding. It's mostly played by judges and judge-adjacent players, and is sometimes advertised as a training tool for judges. As such, it should reward people who have a good understanding of the rules on a technical level and who have the ability to mentally keep track of a complicated game state.Note that these are different things. There are plenty of people who are great at rules knowledge but terrible at Judge's Tower, and vice versa.

Judge's Tower also helps judges internalize that they're going to be wrong, they're going to make mistakes, and that's ok. The unforgivingness of Judge's Tower is a feature, not a bug. Having someone excitedly yell at you "you messed up, you lose!", followed by you sitting out for a few minutes and then jumping right back into the game can do wonders for your self-acceptance and ability to handle failure. As a judge, we will screw up, and very few things will demonstrate that better than Judge's Tower.

Giving newer judges the opportunity to call out the error of a higher level judge is also immensely valuable. We don't want blind deference to authority; newer judges should feel comfortable challenging much more experienced judges, and believe me, even veteran Judge's Tower players will still make mistakes all the time.


One of the nice features of Judge's Tower is that there's a huge pool of "playable" cards. With such a wide selection of options and no strong pressure pushing you into using the same cards as other people, each deck can be wildly different from the next. The lack of maximum deck size means individual Towers can be quite large, ensuring that no two games are similar either.

The average Judge's Tower game will go through around 8 cards per person, though this varies wildly based on the deck composition and skill of the players.If you have lots of "look at the top X cards of the library" effects, you might average as many as 30 cards per person. I would recommend aiming for a minimum of 200 cards in your tower, in order to allow for several games in a row without having to shuffle the Tower in between.

There's no particular reason to shoot for a certain size. My personal tower has several thousand cards in it; I just add in any new cards that seem interesting, and take out any that prove to be unfun.I only bring a small portion of it with me to events, and then I shuffle the whole thing back up at home.

If you included any particular themes in your Tower, or have added specific cards for balancing purposes, remember that adding in more other cards will lower the fraction of your Tower that consists of those cards.

Tricky cards

Many of the best cards to include in Judge's Tower are those that allow players to do things that they would never actually want to do. As such, players tend to not even realize that something is an option, and they play the card the "normal" way in Judge's Tower, making them loseOr get a shame counter, if you're playing that version. For simplicity I'll ignore that variant in this article, it doesn't change any of the reasoning.. A few examples:

These types of cards are some of the best to include, because they're the ones that actually teach people things. Judge's Tower games often revolve primarily around who will be the first to forget about the existence of some permanent on the battlefield and fail to do what it says. This is fun, but gets old when that's the only factor. These sorts of "trick" cards can actually teach people rules they didn't know and help them become a better judge.

Delayed Problems

Another staple of Judge's Tower are cards that set up something you have to remember later. As these pile up, it becomes harder and harder to keep everything straight, and players will eventually forget something.

Anything with restrictions on when it can be cast or activated is great, since it has to stay in the player's hand up until the moment they can cast it, at which point they have to play it immediately. Hex, Even the Odds, Flash Foliage, Rack and Ruin, etc.

Same for activated and abilities. You don't want everyone to only be doing stuff in the upkeep; look for cards that will keep them paying attention for the whole turn. Ancient Hellkite, Arch of Orazca, Maze of Ith. Look for anything that can only be activated under certain conditions.

Look for rare trigger conditions. Will you remember Captain Ripley Vance when you're in the middle of casting your third spell? How about Soulherder? Erebos's Titan? Sandstorm Eidolon? Task Mage Assembly?

You also want on-board effects that change the rules of the game. Dosan the Falling Leaf will be completely irrelevant, right up until it isn't. Monstrous Hound effectively says "this can't attack", until 3 turns later when suddenly it can. Anything that fades into the background only to suddenly change later based on same other card that was played will trip people up.

The more confusion you can sow, the better. Mana Maze, Ashling's Prerogative, Void Winnower will tie people's brains in knots.Make sure your rules cover who is responsible for these. If one player controls Trove of Temptation, another player attacks someone else, and no one notices right away, which of them loses?

There are even cards that more explicitly make you remember to do something in the future (Undiscovered Paradise, Mishra's Bauble, Thawing Glaciers.) Do you remember enough about your turn to know if Aid from the Cowl should trigger? Good luck.

Cards that function from the graveyard are a must-have. Exactly which ones are best will depend on whether you're playing with a shared graveyard or individual graveyards, but anything like Grim Harvest, Anger, or Nether Shadow is likely good regardless.

Unfun "Gotchas"

Judge's Tower is all about "gotchas". A player forgets about something that wouldn't really matter in a normal game of Magic, and they immediately lose. That's pretty much only how you can lose a game of Judge's Tower.

Judge's Tower aficionados tend to lean into this trope. The Reddit post I linked above is titled "A Magic format for judges and people who hate themselves". A group chat I'm in for Judge's Tower players is named "Judge's Tower Support Group". The introduction post to the Judge's Tower codification project a few years ago was titled "Welcome to pure misery!" We enjoy how harsh and unforgiving the game is, and we know it's all in good fun.

An important aspect of this is that people need to have the ability to avoid the "gotcha". When someone slips up, they lose, but if their "slipup" was due to no fault of their own, that's no fun at all, that's just demoralizing. Judge's Tower should be a game of skill on a fair playing field, not a place for the owner of the Tower to embarrass other people.

So firstly, never make someone lose for not knowing a rule of Judge's Tower that wasn't explained to them. If someone is new to Judge's Tower and activates the first ability on a card instead of the last, don't tell them they lose, just inform them of the rule. The point of Judge's Tower is to test their understanding of Magic's rules, not their understanding of the Judge's Tower rules. (And especially not their understanding of some unique rule that's different in this Tower from other Judge's Towers they've played in the past.)Of course, once they do know the Judge's Tower rules, don't let them get away with mistakes they could full well have avoided.

Secondly, don't include cards that do the same thing. The big one here is Panglacial Wurm. If someone searches the library and fails to cast a Panglacial Wurm therin, they've lost. This is no fun. They had no way to know it was there; they haven't even seen the card! And once players do know it's in there, this adds no additional challenge. It just becomes a part of searching that you spend some time finding and casting the Panglacial Wurm.Not to mention how much time they'll waste searching if it's in someone's hand.

Miracle cards have this issue too. The player has to reveal them as they draw them, which is not how players have to draw any other cards in the game. They'll draw their card for the turn, put it into their hand, and then look at it and realize they've lost, without ever having had the opportunity to know what was coming. This is also not very fun; just like Panglacial Wurm, the only way to avoid this is to either know what's in the deck in advance, or just assume every Judge's Tower contains miracle cards and look at your draws before they touch your hand.

And this includes cards that care about something you'd never have thought to track, like Serra Avenger, Arboria, Premature Burial, or Sproutback Trudge. These are less of an issue, but it's not really fair to expect everyone to constantly remember what turn they're on, just in case they draw a Control Win Condition.Of course there's no problem with cards that only require you to start tracking that information after you've drawn the card.

In general, don't design your Tower such that new players will feel they're being punished just for not having played your Tower before. If you are going to include any of these things, let people know about them in advance. That way it goes back to being a test of skill and memory, not just dumb luck.

Lastly, avoid anything massively unbalanced. If you draw for turn and see Future Sight: congratulations, you've just lost the game. People don't join a game of Judge's Tower in order to play Russian Roulette.I know one person who uses cards in the ante zone as their shame counter, and has Contract from Below in their Tower. Anyone who draws it has to draw 7 cards and automatically gets a shame, even if they've done nothing wrong.

People Want to Play the Game

Like any game, Judge's Tower players want to have a chance to actually do stuff. While it can be fun to watch other people's turns like a hawk, waiting to pounce on any mistake they make, everyone wants their own time in the sun. When one player's turn takes so long that other people are getting bored, something has gone wrong. So we want to design our towers to be balanced. We don't want one player to take up the whole game and deprive everyone else of their chance to have fun.

By far the easiest way to do this is to simply avoid excessive card draw. I said above that Future Sight was an auto-lose; that's not quite accurate. There's another option, which is "you spend 20 minutes slowly and carefully playing through your turn, while everyone else stares at the ceiling." Anything that draws a player a large quantity of cards immediately unbalances the game; either their turn takes forever or they lose.

Even if you restrict yourself to effects that draw 3 or fewer cards, having too many of them results in the same issue. The more cards there are that make their controller draw cards, the more likely it is that each draw spell draws them into more draw spells. These "forced storming off" turns can go on for even longer than the turns from a single "draw 7" effect.


Even if you take my advice and limit your Tower to a small number and size of card draw effects, there will still be a compounding effect; the person who draws the first Divination is slightly more likely to draw the next. Permanent draw effects like Angelic Sleuth are worse, since even if the criteria under which they draw a card are rare, they stick around indefinitely, allowing the player to amass more and more of them until they're drawing several cards a turn.

The core problem is not about the size or frequency of card draw effects; it's about the fact that card draw is fundamentally a positive feedback mechanism. If the rate at which players draw new cards is proportional to the number of cards they've already drawn, this means that players who get slightly unlucky early on will have a very hard time reversing that trend. After several turns, you'll get to a point where one player's board state is growing much faster than the others.

This doesn't mean you shouldn't include some small card draw effects; far from it! Cards like Drogskol Reaver are a lot of fun. A big part of any tower is players having to watch for specific conditions to occur, and drawing a card is a much more interesting result than something mostly irrelevant like gaining life or putting a counter on a creature.

No, the way to fix this is by making sure that your Tower has a linear growth rate. I aim for a rate of around 1.5 cards per turn in my Tower, but you can go higher or lower depending on how quickly you want your games to increase in difficulty.You don't want your rate to be too high, because then the game slows down. If everyone is drawing 5 cards a turn, that's "fair", but I'm still going to have to wait quite a while in between each of my turns.

How do you avoid the positive feedback cards leading to exponential growth? Simple: just include cards that provide negative feedback. Any mechanic that preferentially removes cards from the player who currently has the most will take your growth rate downwards. You need simply tune the numbers of positive feedback and negative feedback cards in your Tower in order to have the desired average growth rate.

Some examples of negative feedback cards are:

Be careful not to go too far and make your growth rate negative. An average rate of 0 or lower means people don't amass significant board states, and the game doesn't feel like it's progressing over time, which can be unsatisfying to the players. And since board complexity isn't increasing, it'll take a long time before anyone loses. Cards like Endless Swarm or Mind Over Matter basically read "you win the game", which is boring for everyone.Unless someone else can destroy the Mind Over Matter. Making sure your Tower has enough removal helps to balance out cards that are too good as well as cards that are too bad.

A convenient way to avoid this is to make the cards replace themselves. Removal spells with cantrips like Rending Vines, Dregs of Sorrow, Altar's Reap, and Polymorph are absolutely wonderful. These cards still help balance the Tower, since they allow the player to remove their worst permanent and replace it with a random one. But the card draw ensures that the overall growth rate remains positive, since the player can't actually go down in cards.Specifically, they cancel out exponential growth without affecting linear growth.

You can also make your negative feedback cards conditional on someone doing poorly. Phantasmagorian and Voldaren Pariah do nothing until that player has a bunch of cards, at which point it lets them get down to a more manageable hand/board size. One with Nothing gives a player a get-out-of-jail-free card if they draw it as part of some large draw effect. Bolas's Citadel is brutal for a little while, but if you survive up until you have ten permanents, you get to wipe your whole board.

Cards that give more cards to the players who currently have the least also provide a balancing effect, like Heartwood Storyteller or Shah of Naar Isle. Anything that lets players choose who gets the cards also falls into this category; Cards like Inspiration, Battle Mammoth, and Curse of Verbosity are usually going to get pointed at whoever is currently winning. But be careful with these, since all they're doing is lowering the difference between the growth rates of different players. Overall, these types of cards are still creating positive feedback, and lead to exponential growth in board complexity. Use with caution.

Avoiding Tedium

Sometimes the game will get into a state where people just have to do the exact same things every upkeep. They can just say "ok, activate all these abilities, resolve all these triggers" without having to think about it much. This slows down the game without adding much of a challenge. Avoid having too many cards with abilities that will happen every turn and do nothing interesting, like Agent of Masks or Glittermonger.

If you can make the abilities matter, that's more interesting. Adarkar Windform might trip up an opponent in combat if they forget a creature has lost flying.

"Do nothing" abilities can still be interesting if there are enough cards that interact with them. If a player has gotten into the habit of saying "Activate Glittermonger, sac the treasure, go to draw step" every turn, Stony Silence will give them something to think about.

Also watch out for anything that could cause a loop of the same game actions. Academy Ruins might lead to a player drawing and replaying the same card every turn.

Repeatable removal will make the game boring for everyone. Cards like Hammer of Bogarden, Avatar of Woe, and Prognostic Sphinx all risk taking your growth rate negative, preventing board complexity from building up.

Also, avoid things that take a while. Anything that makes a player search their library, or even just shuffle it, will take a long time. So far I have yet to find any search effect that actually seemed worth the downsides.And remember that if something instructs you to shuffle when the library is already random, you can just skip it. Cards like Warp World or Thieves' Auction are less of a problem, since everyone gets to participate.

Allow People Choices (If Desired)

Some people want Judge's Tower to be as automatic as possible, such that the players are just executing a deterministic algorithm. These towers tend to include rules on what you have to target, what you have to find with tutors, etc., to take away as many choices as possible from the players, and have them focus entirely on performing the necessary actions.

Other players like having some choices to make. Judge's Tower can actually have some pretty interesting strategy decisions. I have 1 removal spell; which permanent is most likely to cause me to lose? I can create a delayed trigger for one of my opponents; who's most likely to forget it? I personally enjoy these sorts of decisions, and many other players do too.

Which way you want to go is up to you, but once you decide, go all the way. If you want your players to have choices, make sure you're really giving them that. Look for cards that don't have a single required functionality, and let players have some agency over the game.

Learning New Rules

While people do want to test their knowledge of Magic's rules, they usually don't want to test their knowledge of esoteric Judge's Tower rules. (Especially since those rules will be different from playgroup to playgroup.) The more time you have to spend explaining how your tower works, the less time you have to play it, and the more energy people have to spend remembering how this form of Judge's Tower works rather than actually playing the game. So try to avoid weird rules and cards that do things that don't fit well under the basic Judge's Tower rules. For example:

Stuff like this leads to confusion, where everyone looks around and says "uh, I'm not sure how this works in Judge's Tower". You'll have to come up with ad hoc rules to handle whatever happened, and it's not really worth the additional overhead unless it's a particularly cool card.

If there is a card you really want to include that has an unfortunate rules issue, you can consider just errataing it yourself. I stay away from large changes, but I'll sometimes tape a slip of paper on top of a card's ability that I don't want it to have, and we just play as though that ability doesn't exist.

Card Versions

Judges should know that looking up the oracle text of cards is important. But doing so for every single card is not fun. It's definitely ok to have a few older cards in the Tower, but if people are having to pull out their phones every few seconds, that's going to get old really fast. So I prefer to use the most recent version of cards whenever possible, and I'll never use cards in a language that the players don't speak.

Especially stay away from recent cards with functional errata, like Hostage Taker or Yotia Declares War. The first time someone loses because they didn't look up the current text of a card in Standard, everyone at the table is going to get paranoid and start looking up every single card they draw.

I will however try to trip people up with reminder text. Judges should know better than to rely on that for their rulings, so I will absolutely include cards that have incorrect or misleading reminder text, such as anything with soulbond or cipher. When bad reminder text isn't available, I avoid it entirely; judges should know how the rules work. (The foil 10th Edition version of Time Stop is great; everyone always forgets to exile it.)

The reason I think this is less of a problem than for oracle text is that once someone looks up how a rule works, they won't need to look it up again. They're actually learning something useful by looking up rules, whereas it's completely infeasible to memorize every card's oracle text.

And of course I'll absolutely include cards with confusing oracle text, like Dead Ringers or Takklemaggot.

Rules Were Made to be Broken

All of these guidelines are just that: guidelines. It's perfectly fine to break of a few of them if you have a good reason, and indeed, I've broken several in my own tower. I mention them as things to keep in mind. If there's a particularly interesting card you want to include, but its inclusion would violate one of these guidelines, go for it! Your Judge's Tower should maximize the amount of fun people have playing it.

Judge's Tower is an extremely forgiving format. Don't spend excessive time stressing over how to build the perfect Tower; just throw something together, try it out, and change it as necessary.And if you discover a new guideline that you think it's a good idea to follow, tell me about it!

Further resources

Some other articles I've found that talk about deckbuilding are Fringe Format: Judge's Tower by Chris Cornejo, From the Brewing Board – As Soon As Stacks by Chris Lansdell, and this forum post by an anonymous MTG Salvation user.

User Avalonians on Reddit has started posting Judge's Tower set reviews:

These are awesome and I hope they continue.