Tibalt's Bag of Tricks

Tibalt's Trickery caught the eye of the Magic player base as soon as it was revealed. It was a counterspell in red, something that hadn't been heard of before; not only that, the rest of the text was exceptionally bizarre, involving milling, random decisions, and Chaos Warp effects. While it looks like it was added at random, this text is Wizards' best attempt at preventing abuse while maintaining a smooth pace of gameplay. But is it the optimal decision, and what else could have been used instead? Let's find out together.

The concept behind Tibalt's Trickery is essentially a version of Chaos Warp for spells on the stack. In this sense, the accusations of it being a color pie break are not as well-founded as it seems, as we've previously seen colors use out-of-color effects as part of a larger effect that's in-color. (Polymorph is the ur-example, though Reality Shift is perhaps more relevant to the conversation.) In order to ensure that the opponent gets something, it keeps exiling cards until it turns over a nonland permanent, compared to the original Chaos Warp which could be a single-target Terminus if your opponent got unlucky.

Of course, Magic players being as enterprising as they are, this would then enable you to cheat out an absolute monster on turn 2 by casting a 0-mana spell, countering it with Tibalt's Trickery, then cascading into some nine-mana behemoth. In order to prevent this, safeguards have to be put on the effect to stop players from being able to do this. The design that Wizards arrived at, clunky as it is, makes it almost impossible to choose what you get from Tibalt's Trickery without putting three lands on the top of your library. (I imagine the "1, 2, or 3" text also has an origin in being easy to decide with a common d6.)

However, that is not the only way to prevent players from casting an Emrakul on turn 2, and perhaps not even the best way in this situation. Let's look at some of the alternate possibilities.

Why not use a shuffle effect?

The original Chaos Warp makes the owner of the targeted permanent shuffle it into their library. This is essentially the "nuclear option" for top-of-library randomization; it's the best way to prevent cheesing the card's effect, but designers justifiably dislike shuffling because you're interrupting your game for 30 to 60 seconds so you can do an action that requires good manual dexterity (or a shuffling machine and a deck you don't care about). 

So is an effect like Tibalt's Trickery so dangerous that we need to make the target shuffle their library? This edges a bit into the subjective, but I would argue that as long as we have access to subtler options that sufficiently randomize the top of the target's library, we should explore those first and see if they're appropriate. For the Commander-minded player, Trickery does slap onto Isochron Scepter, and there's likely other ways to copy it repeatedly, so do we want to take the chance that someone will make other people shuffle their $800 libraries every turn?

Shuffle effects would also make the "differently named card" clause of Tibalt's Trickery harder to do, because you no longer have the original card as a reference. I doubt this is much of an issue unless your opponent is actively trying to cheat, but I do like the new clause as a way of making the results more chaotic and less frequently underwhelming for the caster, so I'd like to try to preserve it as much as possible.

Why not target opponents' spells only?

An easy way to make Tibalt's Trickery less abusable is to simply have it target your opponents' spells only. You can jettison all the text between "counter target spell" and "exiles cards..." because you can't use it to cheat out an Emrakul, and if your opponent scries something to the top of their library to clutch out your removal spell, that's a feature, not a bug. 

This wouldn't be a problem except for the inherent nature of the card. Tibalt's Trickery was designed from the beginning to be a force of chaos, and being able to self-target helps in two ways: One, chaos effects are often more fun if they target the caster, and two, part of a chaotic effect is a wide range of play patterns with a wide range of effects. It's OK if you cheat out a behemoth on turn 2, you just have to do that essentially through luck instead of through manipulating the top of your library.

The kind of player Tibalt's Trickery appeals to is a hardcore "randomness" Tammy who plays Magic primarily because of the wild situations that it can create. Restricting Tibalt's Trickery to opponents only indicates to this player that not only was there a cool strategy that they could have used the card for, Wizards put a car boot on it so they can't actually use it themself. It would fix the card, but would perhaps also "fix" a fair amount of its charm.

Why not dump a high number of cards instead of random ones?

Instead of determining a random number to mill off the top, why not get rid of a large number of cards, like 5 or 6? You can put them on the bottom of that player's library so you don't break the color pie, and it's just as difficult to manipulate as getting three lands on the top. Frankly, if you're able to stack the top 5 or 6 cards of your deck in order, you're probably playing a format where cards like Scroll Rack are legal, in which case Tibalt's Trickery is not the most broken thing you can be doing.

This technique has less downside than the other ones I've covered in this article, and it would be what I would pick if I was in charge of finalizing how Tibalt's Trickery works. However, putting cards on the bottom of your library, like shuffling, requires a fair amount of dexterity that can be irritating if it occurs too often. In particular, if you're placing more than two or so cards on the bottom, you have to maneuver your library into the air and neatly place another stack of cards under it with one hand without accidentally showing any cards to your opponent. What's more, you'll then have to do it a second time when you start cascading. It's a relatively innocuous-sounding task, but it's difficult to do repeatedly or for people with motor issues.

Additionally, although this method removes the headache of randomly determining a certain number of cards to mill, that randomness might actually be an upside. The title and art (though of course it's likely the card was done before the art) are relatively generic and could apply to many red effects, so having a "chaotic" effect that generally requires the use of something – dice – that are usually restricted to silver border gives the card a flavor it wouldn't have if it just fatesealed the player for 5.

Conclusion

Tibalt's Trickery is a great example of how a tiny change to a card can have far-ranging impacts on how it plays and feels. It's also a good example of the subjectivity of card design – many possible iterations of Trickery have their upsides and downsides, and in the end the final version has to be selected more by designer fiat than by any "correct" measurement. Was it better to have the random effect for flavor, or is the headache of having multiple weird processes on a single card not worth it? Was the "different name" clause a good idea or one more thing that could be easily overlooked? While we can all agree there were many different ways to take Tibalt's Trickery, and that the one that was chosen might not have been the optimum, I doubt we can agree on what the "optimum" really would have been.