A lot of discussion about the game by avid players revolves around a single spell, paralysis, which is considered by many, including myself, to be unbalanced, and possibly even broken. Its abusability has even led to the formation of a guild, the Paramancers, who specialize in this one spell. One of the interesting questions to arise from this discussion, however, is what exactly it means for a spell to be "broken."
When we talk about game balance, we generally assume that what we have in mind is a series of equally viable options. If one strategy is disproportionately represented or effective, then it is unbalanced. This, at first, seems like a fair criterion for a balanced game - if, in Starcraft, almost no one ever played Zerg, but 90% of players played Protoss, that would be a clera sign that something was amuck in the balance of the game. However, if we push the idea a little further, it starts to become murkier. Isn't the very idea of "strategy" supposed to be that some moves ARE better than others, and that finding the good moves is what makes the game fun? Rock, paper, scissors is a perfectly balanced game, because no move is better than any other, but for that reason it's impossible to strategize about anything other than player psychology, so the game is shallow.
In the case of Warlocks, some spells are clearly more popular than others, but this creates a self-balancing factor: the more commonly a strategy is used, the more predictable it is, and in a game that relies on predicting your opponent's moves as much as Warlocks does, that can be a fatal weakness. This means that if you use less popular spells, you can take your opponent by surprise, and in doing so may be able to make up for the features of the spell that make it less popular.
So in the case of paralysis, the fact that it is used much more often than most other spells is, in some sense, unbalanced, but that doesn't necessarily make the game worse. This is where we can draw a distinction between an unbalanced and broken spell, because a broken spell will interfere with the overall playability of the game. The problem, in this case, is that paralysis DOES interfere with the overall playability of the game. The criteria that can be used for a broken strategy might include:
1. Whether or not predicting the strategy makes it possible to "punish" the player who is predictable.
2. Whether or not there exists effective counters to the strategy.
3. Whether or not there is a good motivation for a player to use a different strategy.
As I mentioned before, there exists a guild called the Paramancers, who specialize in paralysis. In other words, just by being in this guild, they are effectively announcing to their opponents before the duel begins what strategy they will use, and yet they are still successful players. This simply wouldn't be the case with any other strategy - if I announced that every game I play, whenever possible, my right hand will constantly be casting antispell, I would lose every game I played (even though, in a given game, my right hand might end up making those gestures anyway).
Discussing counters will require a more detailed discussion of the spell itself, which requires some background knowledge of Warlocks. So if you've been reading up to now simply because you liked the idea of distinguishing "unbalanced" and "broken" mechanics, here's your chance to escape.
Paralysis - FFF
In Warlocks, you submit gestures on your left and right hands that add up, over several turns, to spells, that take effect when the gestures of that spell are completed. The gestures needed to cast the spell paralysis are FFF, and it causes one of your opponent's hands to be "paralyzed" into the same gesture on the next turn (well, except that W is paralyzed into P and S into D). The important thing is that since the gestures of paralysis are so symmetrical, gesturing another "F" on the next turn allows you to cast it again immediately (because now your last three gestures were, once again, "FFF"), resulting in "parachains" where one hand gestures "FFFFFFF..." ad infinitum. There is one restriction on parachains built into the basic rules, however - on consecutive turns, you can only paralyze the hand you already were paralyzing.
At first, this doesn't seem terribly abusive. If I use an endless parachain, I can keep one of my opponent's hands tied up, and effectively make us both play one-handed. This makes the spell useful on its own, in case you have some other advantage you want to hold onto, or force your opponent to respond to you and a summoned monster with only one hand available. It also grants a small initiative advantage, because as soon as the paralyzer decides to end his parachain, he can immediately begin casting a new spell, whereas the paralyzed player must suffer the effect of the final turn of paralysis before he can move his hand freely again, leaving him one turn behind on one hand. This is a significant advantage by itself.
However, the real problem with paralysis occurs when you start changing targets. Since it counts as a mind-affecting enchantment, it cancels with other mind-affecting enchantments, which means that you can cast paralysis on yourself to counter an opponent's interrupt, and then go back to paralyzing them on the next turn. What's more, when you stop paralyzing your opponent for one turn, and then start again, you can switch which hand you're affecting, allowing you to alternate and restrict your opponent's use of both his hands. Finally, paralysis cast on a monster stops the monster from attacking that turn, which means that a parachain can be used, on any given turn, to disrupt either of your opponent's hands, hold his monster at bay, or counter one of his interrupts on your (including his own paralysis). This versatility in a spell that can be cast every single turn is incredibly powerful.
There already exists, however, a standard variant that helps to make paralysis easier to disrupt, called "parafc". It means that when paralysis is used on an F gestures, the F is paralyzed into a C gesture. The significance of this is that it makes paralysis targeted at yourself (to counter another enchantment) into a risky move, because if your opponent was bluffing, and doesn't complete their own enchantment, you've paralyzed yourself, and either cannot continue your parachain (because now you've gestured FFFc), or must disrupt your other hand. The effect of this rule is that many situations in which a parachain could not be countered are now situations in which your opponent can generate "50/50" opportunities to disrupt you.
The problem with this solution seems, to me, fairly straightforward. While it's better to have a 50% chance to disrupt a parachain than a 0% chance, if I really know, before my opponent has made a single move, what he's going to do, I should have a 100% chance of countering him. Paralysis can still be cast every turn, can restrict both of your opponent's hands, and can be used defensively - now it's simply that you have run a 50% risk of disrupting yourself when using it defensively (which is the case with every other mind-affecting enchantment, anyway). The fact remains that even playing parafc, Paramancers can play effectively even when their opponents know what strategy they are using, and this simply shouldn't be the case.
I think there are two ways to balance out paralysis as it stands, and they each essentially involve removing one of the spell's advantages - either it shouldn't be cast every turn, or it shouldn't be able to target either of an opponent's hands. I, in fact, already play with the former restriction self-imposed: I do not allow myself to cast paralysis more than three times consecutively, and this is a restriction that is public, and which my opponents know about and take advantage of. The reason I play with this restriction is simple - because playing without extended parachains makes the game more dynamic and interesting, and if I didn't have an explicit rule, I would end up using them simply because they are so advantageous. However, a fiat "do not cast paralysis more than three times" rule is not a very elegant solution, so this restriction, were it generally enforced, would best take the form of a chance in the gestures of paralysis such that it was impossible to cast every turn.
And restricting which hand can be paralyzed would also balance out the spell effectively. My preferred way of handling this would be to say that whichever hand I cast paralysis with is the hand that gets paralyzed (if I gesture "FFF" with my right hand, I can paralyze your right hand). This is significant for a reason that goes beyond the ability of a parachain to alternate hands every other turn - even though only one hand can be paralyzed, as long as the caster gets to choose which hand it is that is affected, and the choice is made after the spell is successfully cast, the target of paralysis has to restrict which gestures he makes on BOTH of his hands, in order to avoid having a gesture on either hand that will become particularly unusable once paralysis takes effect. In fact, it might be a downside of the parafc variant that any spell featuring an F can be so disastrously disrupted by paralysis (given how few spells use the C gesture). The result of this is that on a turn in which I expect to be targeted by paralysis, I will try to gesture either a W (paralyzed into WPP - counterspell) or PS (paralyzed into PSDD - charm monster) on each hand to prevent my opponent from being able to totally disrupt my spellflow. However, since my opponent can alternate the paralyzed hand every other turn, that means that in theory I have to be prepared to make one of those restricted gestures on each hand every other turn. If I knew ahead of time which hand would be paralyzed, this would not be an issue.
There's plenty more to say on the topic, but this is plenty for one post. I like to hope that some of my idealistic self-imposed restrictions gain traction in the community at large, or that at some point a further evolved version of the game will address some of these issues, but in the meantime paralysis is a legitimately broken aspect of an otherwise unbelievably well-crafted game that we must live with.