Attacks that can wipe out a turn if stacked.
"Every other player discards a card" is an example of such a card. By itself, it's fine. But if it gets played multiple times -- which, even in a kingdom without Villages or Throne Rooms, can happen easily in a 4-player game -- then the game can degenerate into a state where players discard most or all of their hands all the time and can't do anything.
Many of the official Dominion cards are great examples of how to circumvent this problem. Notice that Militia doesn't read "discard two cards" but rather "discard down to 3 cards." Torturer does say "discard two cards" but allows a player to take a Curse instead. Sea Hag has you discard the top card of your deck to prevent five Sea Hags from obliterating your next turn.
Cutpurse does hurt in multiples, but its effect is limited to how many Coppers you have in your hand. The damage multiple Bureaucrats can do is similarly constrained to how many Victory cards you have in your hand, and the "gain a Silver" clause of Bureaucrat cleverly dilutes a Bureaucrat-heavy deck, so that multiple Bureaucrats aren't likely to be played repeatedly throughout a game.
The bottom line is you don't want to have an attack so strong that it can, by itself or stacked, completely ruin the next player's turn. Otherwise you can wind up in a game state where one player is locked out of being able to make any meaningful progress.
If you know of the notorious King's Court-King's Court-Goons-Goons-Masquerade pin, you might cite that as a counterexample to this principle. This is a combination that CAN lock a player out of the rest of the game. But Donald X. has said that if he'd caught this combination in testing, he'd probably have nerfed it. In any case, it's an expensive five-card combination requiring three different unique cards from two different sets; as such, it only comes up relatively rarely and is tricky to pull off even if it does. It's not merely a case of playing a single card multiple times.(8)
Attacks that target a specific player.
By purposeful design, Dominion doesn't have attacks that target specific players. Many Dominion players appreciate the lack of politics in the game -- that is, the inevitable over-the-table negotiations and protests that result whenever a game allows a player to choose a specific opponent to target -- and thus may not appreciate a fan card that opens that door. That said, there is not necessarily anything broken about having targeted attacks in Dominion, so if that's what you want to do, go for it.
Donald X. on this subject: http://forum.dominionstrategy.com/index.php?topic=80.0(9)
Attacks that offer no benefit to the player.
Like almost everything in this list, this is not a firm rule. But before you create an attack card that ONLY harms other players, consider that of all the official cards, there are only two Attack cards that don't also provide some form of benefit to the player. These are Sea Hag, which is so strong for its cost that any additional benefit would overpower it; and Saboteur, which would be even more unpleasant than it already is if there were any additional incentive to use it. Even attacks as brutal as Mountebank, Witch, Ambassador, and Ghost Ship provide additional benefit to the player. Besides that benefit being necessary to achieve forward movement in the game, they also ensure that, if the attack portion is blocked by Moat, the attacking player hasn't utterly wasted his action.
All I'm getting at is that if you create an Attack card with no benefit, make sure the decision not to include a benefit was a conscious and considered choice and not just something you overlooked.(10)
Cards that reference limited card types.
Examples of this would include "Gain an attack card," or "If you have at least three duration cards in play," or "You may trash up to 3 Curses," or "+$1 for every dual-typed card in your hand." The problem with these cards is simply that if they show up in kingdoms without any of these types present, the card is useless.
You can still make such a card work if the card also has behavior that is sometimes worth buying anyway. This is the case with reaction cards such as Moat, Secret Chamber, and Horse Traders. In the absence of Attack cards in the kingdom, these may still be worth having for their other functions.
Another way to solve the problem is for the card itself to force the intended condition. For example, a card that specifically references Potion cards is fine if it carries a Potion-based cost, as then, whenever that card would be present, Potions would be present also. Similarly, "Choose one: Trash any number of Curses from your hand; or every other player gains a Curse" would work, as then you wouldn't need a separate cursing card to be present.(11)
Cards that require resources that frequently aren't available.
This is similar to the above. One example of this kind of card would be something that has "-1 Buy" on it. I actually think that's a really cool idea; the problem is that it's dead in a kingdom with no +Buy cards.
Another example would be Diadem, a card whose behavior is based on having unused actions. Donald X. experimented with Diadem as a regular kingdom card but found it to be a dead card too often, as in many kingdoms there are no sources of extra actions. (But it works fine as a Prize, where it's not taking up a whole kingdom pile.)
Still another example is a repeatedly proposed fan card that does only this: "+1 Card, +1 Action." Normally, this does nothing, as it merely replaces both the card slot and the action it uses up. But the justification for it is that it helps enable Conspirator, it lowers the price of Peddler, it can provide extra cards and actions with Throne Room or King's Court, it provides an extra unique card for Horn of Plenty, and so on. Yes, but the number of kingdoms with any of these cards in it is very small -- and in many of those, other cards will accomplish the same things. On the majority of boards, it's a dead pile.(12)
Cards that retrieve cards from the trash pile.
When cards are trashed, they're trashed for a reason. A fan card that does something like "Gain a card from the trash pile" is going to be useless most of the time, because even if there are trashers in the kingdom (and there may well not be), who wants Coppers, Estates, and Curses?
If there are trash-for-benefit cards in the kingdom (like Salvager, Apprentice, Bishop, and the Remodel family), then you might find something good in there, but this will only be a small minority of boards. And within that minority, there's a good chance the card will be brokenly powerful, and/or players will be dissuaded from using the trash-for-benefit cards on good cards in the first place.
Rogue and Graverobber retrieve cards from the trash, but they also put useful cards there to be retrieved at a later time. The balance is extremely delicate; any card that retrieves cards from the trash should be carefully tested.(13)
Cards that stay out permanently.
Nothing inherently wrong with this idea, but a minor point: Usually this kind of idea is proposed with a new card type of "Permanent" or something. If the card otherwise behaves like a Duration card, then just call it a Duration card instead of inventing a new type. According to the Seaside rules, Duration cards stay out until the last turn in which they have an effect. Although the specific Duration cards in Seaside all get discarded after the following turn (except for a failed Tactician, which gets cleaned up immediately), the Duration type itself allows for a card to remain out for any arbitrary number of turns.(14)
Actions that are unnecessarily non-terminal.
The presence or absence of +Actions might be the single most defining characteristic of an Action card. Two versions of the same card, differing only in one having +1 Action and the other not, will likely play wildly different from one another. It's not the differing number of +Cards, for example, that distinguishes Smithy from Laboratory.
There is no firm rule concerning what kinds of Action cards should be terminal and which non-terminal. That decision is largely subjective. However, many fan cards include +1 Action when they might be more interesting without it. Make sure you consider the gameplay ramifications both ways. The right choice will be the one that makes the game's strategy more interesting, not necessarily the one that makes the card more appealing to purchase and use.
Sometimes the lack of +Actions is best. Consider how much less interesting Courtyard would be if it provided +1 Action (assuming its price was adjusted to compensate). The strategic feature of the official version of the card -- being able to save a dead Action card for the next turn -- would be destroyed.
On the other hand, sometimes you need +Actions to make the card work. A terminal version of Minion would be cute but weak; certainly not a card that a whole strategy could be built around.
Again, though, the distinction isn't always clear, and there isn't always one right answer. The important thing is just to consider the matter and make a thoughtful decision.(15)
Cards without accountability.
An example of such a card would be, "If this is in your hand at the start of your turn, you must play it immediately." The problem is that players can't be held accountable for following this rule. You'd have to have each player reveal his hand at the start of every turn, just so the other players can confirm that there is no copy of that card in hand to play. Otherwise it would be easy to keep the card in hand and secretly discard it underneath the other cards during clean-up.
That said, if you only intend to use such a card when playing with friends you trust, by all means, try an idea like this out. Otherwise, make sure your cards allow for accountability. Note Bureaucrat and Cutpurse, which provide such a mechanism.
Interestingly, not all of the cards in the base set are fully accountable. Throne Room, Mine, and Moneylender all require you to do something with a card in your hand, but if you don't have a card of the correct type, they don't tell you to reveal your hand to prove it. (Treasure Map, from Seaside, is also like this.) Donald X. has expressed regret over Throne Room specifically; however, these cards are not really that problematic since the "fix" for them would probably have been to make their effects optional, rather than adding in accountability to the requirement. On a card whose effect needs to be mandatory (e.g., the aforementioned Bureaucrat and Cutpurse attacks), accountability is much more important, and you should make sure your own cards allow for it.(16)
Cards that look at the number of victory points or coins in your hand.
There are lots of different incarnations of this. And example might be, "The player to your left reveals the top two cards of his deck. +$ equal to the number of victory points he reveals." Another one might be, "All players reveal their hands. The player with the least total treasure gains a Curse."
Sidestepping balance issues in these specific examples that might jump out at you, these types of cards are unworkable because it's not always easy to quantify how many victory points or how much treasure you have in your hand. If only the base Victory cards are out, fine, but what if Gardens, Duke, Vineyard, or Fairgrounds is in play? You can't really calculate how much these are "worth" until the end of the game.
Treasure values are similarly nebulous. Technically, treasure isn't worth any coins at all until it's played. When a treasure card is played, only then does it yield some number of coins. In the case of the base treasures, this amount is always the same. (Well, almost. See Coppersmith.) But Bank, Philosopher's Stone, and Diadem vary, and Potion's yield isn't in coins at all.
Better approaches would be to count the number of Treasure or Victory cards, or look at the costs of those cards. In the latter case, you'll still have special cases in Philosopher's Stone and Vineyards, but many other Dominion cards (Salvager, Forge, Apprentice, etc) deal with (or ignore) Potion-based costs just fine.(17)
Cards with special-case rules.
Before using a special-case rule, be very sure you cannot achieve the effect you're after any other way. Even then, think long and hard about whether your special-case rule will cause conflicts with other rules or cards.
For example, you might want to have a powerful card limited by the clause, "You may only play one copy of this card per turn." Fine, but what happens when Golem turns up two copies of it? Which card's rules get broken?
Generally it's better for cards to work within the rules of Dominion than to override them. Even if a special-case rule poses no problem with the current set of official cards, you never know if something in a future expansion will cause a conflict.(18)
Cards that depend on the order of the discard pile.
An example of this would be, "Look at the top five cards of your Discard pile. Put two of them in your hand."
Usually opening up Dominion's strategy space is a good thing; once in a while, not so much. It's no accident that no official Dominion card cares about the order that cards appear in the discard pile. The moment you introduce one that does, suddenly every player who buys it will have to think very carefully about how they perform every single clean-up phase, just in case they happen to draw that card in the next hand. This will dramatically slow down the game, and most of the time it'll still be wasted effort.
You can still use the discard pile -- Counting House does it fine -- but if you use it at all, it's probably best to use all of it.(19)
Cards that allow unlimited accumulation of victory tokens.
Designing cards that award victory tokens is a trickier challenge than it seems. The reason is that you have to be careful not to allow a game to devolve into a game state where the optimum strategy for all players is to forego buying victory cards in favor of playing and replaying their victory-token-earning cards. Then the game never ends.
Of the three official cards that award victory tokens, two of them are tied to finite resources. Goons only awards victory tokens when you buy a card, which ensures that sooner or later the game will eventually end normally. Bishop awards tokens by trashing cards, which indirectly ensures the same thing -- if you don't buy cards, you won't have cards to trash.
Monument is the exception. In theory, if all players wind up with hands consisting of King's Court-King's Court-Monument-Monument-Monument, you could indeed wind up in an unending game state. But this doesn't really happen in practice, probably because of two things: one, Monument being a terminal makes it difficult to spam; two, it offers $, which encourages the purchase of cards.
The lesson these official cards teach us is that if you have a fan card that awards victory tokens, make sure the game can't wind up in an unending game state. If you can't prove this to yourself with theory (as is the case with Goons), then you'll need even more playtesting than usual to make sure.Part III. Myths About Card Prices
Before I discuss how to price your cards, it's worth saying how NOT to price your cards, so you can undertake that task without any misleading preconceptions.(1)
Myth: The cost scale is linear.
One of the most important things to remember about card costs is that they are not linear. Since your initial starting deck generates an average of $3.5 every turn, that means the difference between $2 and $3 -- in terms of how difficult to is to achieve that amount -- is near insignificant. By contrast, $5 takes a bit of work to achieve, so the difference between $4 and $5 is quite a leap.
In general, card costs of $6 and up are probably close enough to linear that you can treat them that way. But below that, you should be aware that the jumps between the different costs are not necessarily all equal.(2)
Myth: Card costs are proportional to their strength.
Card cost and strength are usually correlated, especially at $5 and up. But since there is so little practical difference between costs $2 to $4, costs in that range are more a function of gameplay balance with respect to the opening.
Treasure Map, for example, is not priced at $4 because it's "roughly stronger than the $3 cards." (Ambassador would take issue with that.) It's priced at $4 to prevent players from opening double Treasure Map and potentially cashing them in for four Golds on Turn 3, which would be ridiculous. Treasure Map is swingy enough as it is, but a $3 Treasure Map would reduce Dominion to the strategic equivalent of High Card Draw.
Another example of a power/cost mismatch is Chapel, whose strength would be competitive at $4 but was priced at $2 for game balance reasons.
These thoughts on pricing, from Donald X., are essential reading for anyone creating fan cards: http://forum.dominionstrategy.com/index.php?topic=84.0
A more obvious way that card costs can be out of sync with their strength is when there are special circumstances concerning buying them. Grand Market is probably worth $7.5 or so on strength alone, but its "no Copper" clause allows it to be balanced at a cheaper price. Peddler, similarly, is not as strong as an $8 price tag suggests, since its cost fluctuates depending on the circumstances.(3)
Myth: Given the above, card strength doesn't matter with respect to cost.
Well, yes it does. Card strength is not a completely distinct issue from opening gameplay considerations. It's just not, on its own, the only and overriding concern.
However, you do want to be careful about cards that are "strictly superior" or "strictly inferior" to other cards and price them accordingly. Worker's Village, being strictly superior to Village -- that is, essentially always at least providing all the benefits that Village provides -- kind of has to be priced higher. I say "kind of," because a $3 Worker's Village is more of an aesthetic problem than a gameplay problem. Worker's Village wouldn't really break the game at $3. But if Village and Worker's Village were both on the table at the same time, it would feel wrong, right? And the poor Village pile would go untouched. Pricing Worker's Village higher is better game design, because it keeps both cards useful and, therefore, leaves a larger number of strategic decisions up to the player.
So if you figure out that your custom card is strictly superior or strictly inferior to another card, you should probably price it accordingly, and then make sure the card is still balanced at the resulting price. This may mean that some cards are just too difficult to price well. Suppose your card offers "+3 Cards, +1 Buy." It has to be higher than $4, because the strictly inferior Smithy is priced at $4. But at $5, your new card looks pretty weak next to Torturer, which also offers +3 Cards. Still, it is not STRICTLY weaker than Torturer, so $5 it is.
Note, however, that it doesn't take much to break out of these "strict" constraints. Hunting Party might look funny priced at the same price as Laboratory, but there's at least one case where Hunting Party is inferior: if you have one copy of every card in your deck, playing Hunting Party will just cycle through your deck without letting you draw a second card, whereas you'd get a second card with Laboratory. This is a small case, but it's enough to allow Hunting Party to cost $5 instead of getting stuck with a less appropriate $6 cost.
If you get stuck in a pricing dilemma like this, another option is to use a Potion-based cost. Note how Laboratory's and Alchemist's costs are incomparable, because they're priced on completely separate scales.
[Continued in the next post...]