Ban the ban: essential game design advice (with examples)

no bans!

If you’ve read the rules to more than 3 board games, you’ve probably read a rule phrased like so: “You may not [take some action].

Why do such rules exist? The most common reason is, during play-testing, the designer discovered players want to take an action that would hurt the experience – for example, a too-powerful action every player would take every turn if allowed. The most obvious fix is to ban it.

Banning an action is usually a bad solution. Here I explain why and what to do instead.

A player’s actions never come from nowhere – they usually originate in the rules. What I mean is: rules create a “space of permissible actions” in players’ minds, and the actions they want to take are inside it. When a player wants to do something bad for play, it’s because the rules suggest she can.

As evidence, consider: for any game, there are countless not-explicitly-banned actions players could take but don’t. For example most games don’t ban switching sides whenever you want, lighting the board on fire, or flipping it. But no one does (except 10-year-olds and the guy below), because nothing in the rules suggests them.

I will keep putting this same gif in my posts over and over until I’m dead.

So. If players want to do something they shouldn’t, it must be inside the space of permissible actions. And when you ban it, you send a mixed signal: that the action should be inside the space of permissible actions, except it isn’t. This makes a game hard to learn (when new players make the same illegal move over and over) and hard for players to find flow during play (because they’re always checking to see if what they want to do is legal).

If bans were the only solution, they’d be a necessary evil. Happily, there are usually other solutions to not only fix the problem, but improve your design even further.

The exact solution depends on the problem’s specifics, but often it’s about incentives. Specifically, an action is so powerful, all players would take it all the time if allowed. This common case is an especially good opportunity to ban the ban, because it gives you a chance not only to allow more options and cut cognitive overhead, but to create an effect essential for a great strategy game: speciousness.

Speciousness is a term I use to describe the way good strategy games trick you into thinking bad moves are good. Speciousness means apparently good or right though lacking in real merit. You can read a discussion of speciousness here, but here’s the short version:

If you’re trying to design a strategy game of any depth, you face a dilemma: you don’t want players to feel lost in a strategic fog (because that I-have-no-idea-what-to-do feeling is discouraging), but nor do want want them to easily know what makes for a good move (because then the game will lack depth and surprise). How do you avoid both problems at once? Answer: you tempt players into making moves that feel more right than they actually are – that’s speciousness.

One way to do this is to include turn options which feel natural and powerful to the players, but turn out to have some cost.

You see where I’m going. If players repeatedly want to make a too-powerful move, rather than ban it, impose a cost. You’ll not only fix the problem and expand the space of options, you’ll also create speciousness (even if you can’t create speciousness, it’s still a good idea to avoid bans for the reasons given above, but speciousness is icing on the cake).

To make these ideas clear, I’ll describe examples from my own games. The first illustrates how replacing a ban with a cost creates speciousness, and the others show how ban-lifting can apply in other circumstances.

Case Study #1 – Catchup

A game of Catchup, in early mid-game

Catchup, my most popular game, is a neat case, because it deploys the advice above twice over.

First, the game’s central mechanic is to impose a cost on what would otherwise be the clear best move-type.

It’s a simple abstract stone-placement game where the goal is to build the largest connected group of stones on the board. The obvious thing to do is to connect your stones in a clump in the center of board, because it both grows your own largest group and cuts your opponent off from stretching her largest group across the board. The game would be broken if there were nothing more to it.

Early on, I tried various conditional bans to prevent clumping – that is, you could only enlarge your groups under certain conditions – for example you could only make a large group by connecting together small groups of a certain size, etc. These schemes all failed.

Then I hit on an idea: you can place your stones wherever, whenever they want, but if you increase the size of the largest group on the board, your opponent gets an extra stone on her next turn. That’s the price you pay for making a strong move. It created a temptation: you really want to enlarge your biggest group because it’s the goal of the game, it’s right there for the taking and it just feels right, but if you do, the extra stone you give your opponent will hurt. Speciousness.

Even after that rule was in place, there was another ban in those early versions: to prevent frequent ties, players were banned from placing stones such that their largest groups were the same size. Here I’d banned a move not because it was too strong, but because it created a structural problem. Like most bans, it complicated the rules and forced players to repeatedly check whether their intended moves were legal.

Eventually, I realized I could lift the ban by using tie breaks: specifically, if the players’ groups are the same size when the game ends, they compare their second-largest groups, and so on until they come to a pair which aren’t the same size. Whoever owns the larger wins.

This had a profound effect: it made ties impossible, simplified the rules, made play smoother, and to my surprise, dramatically deepened the game (for reasons too complex to discuss here). It turned Catchup into one of my proudest achievements. It also reinforced the value of lifting bans.

Case Study#2 – Stinker


I like this example because Stinker isn’t even a strategy game – it’s a party game and the design goal is laughter, yet the advice above applies – an indication of how general it is.

Each player has a jumble of letter tiles (like Scrabble tiles). Each round, a prompt is read aloud and each player races to form a response to that prompt with her letter tiles. A judge decides which response is best each round, and that response gets points. It’s like a cross between Bananagrams and Apples to Apples.

Early versions had a ban: players weren’t allowed to make responses using fewer than 20 tiles. I’d made this rule because players form their responses under time pressure, which encouraged them to make short responses with few letters. If players only make short responses, the game is less funny.

But the ban also made the game harder by forcing players to count their tiles while responding, and to fit answers within strict length constraints. It only worked for word-lovers and crossword enthusiasts, but I wanted a game for a broad audience.

Eventually I took my own advice and allowed players to use any number of letter tiles, but I changed the scoring rules to compensate. Each round’s winner no longer gets a fixed number of points; instead she gets a point for each letter in her response, which encourages long responses and neutralizes the time-pressure incentive to make short ones.

Players now make responses of a variety of lengths, depending on their proclivities, skill, and available letters, and the game is a schmillion times better for it. Stinker’s now one of my favorites among my designs, and players of many stripes routinely go wild for it.

Exception to the Rule

There may be rare occasions where banning is the right choice. Usually only in simple designs without other sources of cognitive overhead, so players have enough mental space to grapple with a ban.

A good example is Yavalath, a game built almost entirely around a single banned move: it’s a simple abstract n-in-a-row game where you win if you make a row of 4 but lose if you make a row of 3.

A 3-player game of Yavalath
A 3-player game of Yavalath

Because the entire game is built around the banned 3-in-a-row, the banned move has the attentional spotlight to itself and the cognitive overhead it introduces feels less like a distraction and more like the game itself.

But that’s about as far as that goes. Unless you’re designing an austere abstract game like Yavalath and a ban is its centerpiece, you’ll improve your design by lifting bans. If you’re designing a Euro-type game, for example, you should almost certainly ban the ban.

Nick Bentley

top photo courtesy Viewminder

24 thoughts on “Ban the ban: essential game design advice (with examples)

  1. Good points; I tend to agree. If nothing else, rules written with lots of “You can’t do X” statements lead to arguments about “Well, I think I can do Z because they would have said I can’t do Z if they meant that I can’t do Z…”

    One nit-pick: Making 3 in a row is not banned in Yavalath – it’s certainly allowed, but you lose if you do it. The distinction matters particularly in the 3-player version, where you are obliged to block the following player from making 4 in a row.

    If that seems like too pedantic hair-splitting, perhaps it shows that the concept of “banned” is a bit more nebulous than we’d like to think. 🙂

    1. Agree Re: Yavalath. I included it as an example because the psychological dynamic feels the same for me. I have a little brain subroutine running when I play to check whether I can take an intended move.

    1. Apologies. I have a weakness for that gif. I guess I should try to resist sticking it every possible place on the weakest of pretenses.

  2. Hey Nick, really great article. I like how making actions cost something, instead of explicitly banning, opens up the game to more interesting decisions. There’s lots of ways game designers can encourage players to play games in a particular way (or to encourage them to enjoy the game). With human and animal behaviour, I believe positive feedback generally presents better, more consistent results than negative feedback. Imposing the cost (negative) in Catch Up seems to discourage players from simply pursuing a less interesting strategy of just adding to their biggest group, and open the possibility for strategising to join groups later in the game for the win (positive). Stinker’s a more obvious example of positive feedback. Instead of discouraging uninteresting gameplay, you encouraged interesting gameplay by rewarding longer lettered plays. Thanks for the insight!

    1. Thanks kotzuryang. w/r/t Catchup, bear in mind that without the negative feedback I’ve included in that game, positive feedback is so ridiculously strong that the first mover wins every time, even if it’s a first time player. Even with the negative feedback I’ve included, there’s no shortage of positive feedback for a variety of actions. The negative feedback turns what would otherwise be a completely broken game into something subtle. I agree positive feedback is critically important in games. But I also think a *small* dose of negative feedback in the right situation can work wonders. But it must be titrated very carefully.

      1. I guess compared to outright banning moves, negative feedback could be considered positive feedback. I think negative and positive feedback are both very important in creating a balanced game, but positive feedback does more to reward the player, making them feel in control and creating an enjoyable experience.

  3. Wonderful post Nick! It can be challenging to explain why a designer should avoid taking a “short cut” by using rules exceptions to define competitive constraints but you did an incredible job doing so.

  4. I’m glad you added a qualifier in your last paragraph: “Unless you’re designing an austere abstract game like Yavalath and a ban is its centerpiece, …” Or I would have asked you, what about Crossway. The ban on two specific patterns is IMO an elegant solution to the ambiguity problem for connection games on rectilinear grids.

    I agree that, in general, a ban can make the task of learning to play more difficult, and my game Lazo provides an example to support that. I could have used spheres, such as marbles, instead of the complicated tile shapes I use. But that would have required a rule banning placement of marbles on many apparently perfectly playable spots. So I created the expensive shapes to avoid that problem.

    There are always exceptions. To quote some French guy, less is more. Sometimes added restrictions are what a game needs, to create the dynamic tension that makes it interesting. I agree that it is usually best to present these restrictions clearly enough to avoid confusion.

    1. Thanks for pointing out Lazo as an example. It’s great when the rules are physically embodied in the components. It’s a way to add rules/constraints with zero extra mental overhead. That’s maybe the very best way to constrain gameplay. But yes, there are always exceptions, and no shortage in game design.

  5. Great articles, some really interesting points. I think any action should be trade off, “you can do this, but it will cost you this” and I that’s important. When I was younger, I would design games where you could only make move based upon the [For example] cards in your hand. But then what if you were dealt a bad hand. This was an error in my part. So I think giving things a cost really covers a few separate challenges.

    Oh btw, I was distracted by the guy flipping the table. HAha I just kept watching it over and over and over! 🙂 That was pretty funny.

    1. Thanks John! I agree regarding tradeoffs. Also, you may be the only person besides me who isn’t annoyed to death by that gif.

  6. It’s a well-taken point, and also makes me think of the difficulties facing an MMO moderator/designer (persuading group behavior/keeping the play environment balanced) or of course an economist or regulator. As abstract game designers we have it quite easy in the finite realm of our small-ish tabletop dynamics, but I suppose you’re right that even then it might “seem” painless to impose a rule but that dissuasion with an imposed cost is the better route.

  7. And one other point– the temptation to make a rule banning something often comes up when playtesting your game and let’s say one of your best testers finds a vulnerability that he exploits and he won’t relent, argh– he’s gonna abuse that weakness even if it wrecks the fun of the game. We’ve also all experienced this with published games too. I think the Catan Card Game (not Catan itself, don’t flame me peeps!) breaks when someone relentlessly takes knights as an early strategy, and in my family we’ve just “banned” that as our own quick fix.

    But back to the point of designing an original game. When the iterative weaknesses show, there’s other subtle ways to refine your game that can function as a ban, like changing the design of the board, simply changing the problematic rule (instead of adding a second rule that limits the first), or changing how pieces/units are managed. It’s especially satisfying when the problem is then fixed and invisible to everyone except that game-tester who can no longer take advantage that “hole.” It’s his turn to say arghh.

    This is more of a sidebar, though, ’cause I don’t challenge your point, Nick, that choosing to ban as a game designer is generally déclassé. And yet it’s always there tempting us because easy fixes are, well, easy.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s