The Reason for Combat Tricks

Magic's combat system is rather unusual. The rules seem to favor the defending player far more than the attacking player. For example, summoning sick creatures can block, but not attack. The defending player makes all the choices about which creatures 'fight' each other in combat, and can even multi-block or decline to block altogether. And damage wears off from creatures that survive to the end of the turn. Some implications of these rules are:

A single 2/3 can hold off many 2/2s from attacking; you would actually need five 2/2s before repeatedly alpha striking would kill the defending player before all your creatures die, assuming the board state didn't change in that time.

If the attacking player has 3 creatures and the defending player has 2, the attacking player has 8 possible choices of how to make their attacks, including not attacking at all. If they attack with all three creatures, the defending player would have 16 possible choices of how to block.

Of course, there are reasons why these combat rules are the way they are. If damage didn't wear off at end of turn, then you would need lots of little counters to track how much damage was marked on each creature. If summoning sickness didn't exist, games would be even slower as people would always keep a creature back to defend against whatever the other player could summon. If the defending player didn't have total choice about how to block, whoever summoned the biggest creature would be unstoppable.

That doesn't change the fact that these rules can lead to board stalls, though. Board stalls are not only when neither player wants to attack, but also that adding more creatures to the board doesn't improve the situation. This most often happens in the sealed deck format. Board stalls are un-fun because the complexity goes way up – if either player even wants to think about attacking they have to consider hundreds of possible blocking combinations, as well as possible counter-attacks when your creatures are tapped – and because it makes the game drag out longer than the players would like. In a tournament situation with timed matches, this can lead to draws.

Fortunately, there are several tools at a card designer's disposal to prevent board stalls from happening.

Creature size: If you make more creatures with high power and low toughness rather than the other way around, creatures are more likely to trade in combat and aggressive strategies will be better. Similarly, avoid abilities that make creatures better at blocking (like first strike) but use more that encourage attacking (vigilance, can't block, attacks each turn if able, etc). But of course, you don't want every set to be like this.

Evasion: The most common form of evasion is flying. If a player wants to attack their opponent with flying creatures, the board state is simplified considerably since they only have to worry about which of the opponent's creatures can block flyers (and if they are able to remove those creatures, it's even easier). Or you can even make creatures with the “can't be blocked” ability.

Note that forms of evasion that allow the opponent to block and trade, like menace or trample, don't do as much to prevent board stalls.
The only downside of using this approach to solve a board stall is that it makes the rest of the creatures feel irrelevant.

Removal: If players have lots of removal spells, they can keep the board free of creatures, which simplifies their decisions. Or if they have access to mass removal, they can break out of a board stall they already got into. But mass removal is too swingy and can't be printed at common. And constantly having creatures die might not be any more fun than a board stall.

Big effects: Cards like Overrun and Sleep are much more effective when there are a lot of creatures on the board, to the point where whoever casts a spell like this can win the game on the spot. Which is also a bit of an unsatisfying way to resolve a board stall, since the game just turns into “whoever draws Sleep first wins.”

Which brings us to the point of this article: combat tricks (that is, instants that change the outcome of combat, like Giant Growth). I don't know when Richard Garfield first came up with the idea of instants, but I suspect it might have been when he noticed the problem of a single 2/3 being able to hold off several 2/2s. A sorcery that could increase the power of one of those creatures would let it get through for one attack, but it wouldn't help the other creatures the attacking player controls. What if the attacking player could wait until seeing how the defending player blocks, then increase the power of the blocked creature? This solves the problem perfectly, since it also works even if the defending player chooses not to block, then the attacking player can choose not to cast their pump spell, and have it to threaten the exact same play next turn. As an added benefit, it adds an aspect of bluffing to the game.

But can't combat tricks also be used by the defending player? That depends: you could always create a mechanic like bloodrush that says these cards can only be used to attacking creatures, to make sure your cards serve the purpose of ending the game rather than prolonging it. But you don't really need to, because combat tricks are naturally more useful for aggressive than defensive players (this fact is part of Magic's steep learning curve – I was playing for years before I realised it).

The reason why is because attacking players know in advance when they will want to use combat tricks so they will always have their mana open, and if the defending player doesn't force them to use the trick, then they can spend their mana on something else. Whereas a defending player might keep their mana open, but if the attacker plays around the trick, the mana is wasted.

You can encourage combat tricks to be used in this way by making them more expensive, so that it's harder for defending players to keep mana open to use them, and by adding in aggressive abilities to help newer players to realise how the card is supposed to be used. For example, instead of making Giant Growth, make Awaken the Bear – trample is only relevant when attacking. You can still make some defensive combat tricks, since that can lead to exciting sequences of play and counterplay, but keep the as-fan down (that is, make fewer of them or put them at higher rarities so they won't be in players' decks as often) so you don't discourage attacking.

A few single-target combat tricks aren't going to break a really big board stall, but they can help prevent the game from getting to that state in the first place since attacking players can use them as removal spells to keep blockers from piling up; and if the defending player is low on life, they can allow an attacking player to finish off the game by increasing the power of an unblocked creature rather than a blocked creature.

Oh, and don't forget that defensive decks need a way to win too!

Mana Unscrewing

The mana system, you might have heard, is a big part of what makes Magic unique and even successful, since it adds just the right amount of randomness to a game so that an unskilled player can sometimes beat a skilled player and so that a deck can sometimes win a match against a deck that's a poor matchup for it. Without the randomness introduced by the lands being in random places in the deck, bringing the wrong deck to a matchup could mean that you might as well just call the outcome before playing, and no one wants that. But this knowledge is of little consolation to you as you throw your deck across the room for drawing ten lands in a row (or ten nonlands when you really need a land).

Actually, the twin problems of mana screw and mana flood don't give quite the same experience – mana screw is more annoying since no one wants the game to be over before it has even begun. Whereas, if you play as good as you can and just lose to bad luck at the end from drawing too many lands, you can say “that's just the nature of the game”. But so far it seems like you can't have one without the other (though the new mulligan-scry rule is a step in the right direction).

The reason the two problems are related is, if the player has a card that helps them out when they get mana flooded, then they can just play more lands in their deck to avoid getting mana screwed, increasing the risk of flood but not caring as much about it. Whereas if the player has a card that helps them when mana screwed, they can run fewer lands in their deck (provided that the card that helps them doesn't require them to have mana to use it – you can't use a basic landcycling ability if you don't have even two mana.)

If the randomness introduced by the mana system is such a great thing, why would you want to give the player tools to fight against it? One reason is because of player personality. There is a player psychographic out there called 'Adrenalin Timmy' who wants more randomness in the game – who will play coin-flip cards like Stitch in Time or who will sacrifice all their lands to unsuspend a Greater Gargadon and hope the opponent isn't holding Slaughter Pact. Similarly, there are players who would rather sacrifice power for consistency. Giving them that choice means that Magic can appeal to all different kinds of people.

The key is, though, to try and make the most fun strategies the most powerful overall in terms of average win percentages. You neither want to make the extremely consistent strategies the most powerful or Spikes will get bored, nor make the extremely inconsistent strategies the most powerful or Spikes will complain that player skill is less of a factor than they would like. So spells like Stitch in Time are slightly overcosted on purpose, but so are spells like Chartooth Cougar.

Two other reasons why you want to make cards that let players manage mana screw are, firstly, because it means the player can't get too angry at the game – it's good to have a scapegoat for your losses, but sometimes you want to take responsibility for your loss instead and say “maybe I'll play more cycling lands in my deck from now on.” Secondly, because it's unavoidable. Even a simple card draw spell like Divination essentially smooths out mana issues as well as giving card advantage, and similarly a player's choice of mana curve affects how often they will experience mana issues.

So what kinds of tools exist to help players deal with mana issues?

Optional increased costs: These are abilities like kicker, flashback, and activated abilities on lands and creatures, where you can choose to pay extra mana to get something extra on top of the basic effect of the card, but you don't have to. The extra effect is usually much less powerful than you would expect to get for the mana you are spending on it, but if you are mana flooded and would have wasted the mana anyway, then it's like you are getting the extra effect for free, making the mana flood less painful. The basic effect of the card is usually a little bit inefficient to make up for the versatility, but still playable.

Consider the card Valakut Invoker. 3 mana for a vanilla 2/3 is a decent card if a bit on the boring and underpowered side. 8 mana for a Lightning Bolt is terribly inefficient, even considering that it doesn't cost you a card (it's worse than Ember Shot). So a player usually would not plan on using its ability if they put it into their deck, but it just works as an emergency backup plan to make sure they still have something to do if they draw too many lands or the game goes long.

Optional reduced costs: This is the same thing except that the default way to cast the card is the more expensive way, but players can choose a cheaper way if they don't have enough mana. Evoke is one example of this. It's not as popular a tool because players sometimes forget they have the option to cast something cheaper when they are just looking at the cost in the top right of the card and not the other cost in the text box.

Scaling costs: This is when a card has no fixed cost and the player can choose to spend as much mana as they want on it to get a bigger effect. Mainly for X spells like Blaze, but even artifacts that just have an activated ability like Jayemdae Tome would qualify, since if you draw lots of lands you will have more turns where you'll activate the ability, whereas if you draw less lands then you'll spend your mana casting your spells and activate the tome less often. Also the level up creatures from Rise of the Eldrazi could be considered to be in this category.

Library manipulation: Effects like scry can be used when a player is mana screwed to try and put nonlands aside and get closer to drawing their next land, or when a player is mana flooded to avoid drawing lands. Cycling and landcycling are a special kind of effect that combine optional reduced costs and library manipulation effects, since you're spending less mana to get a lesser effect than the normal effect of the card, but you're also changing a nonland into a land or getting closer to drawing your next land.

This is not an exhaustive list. There are many more ways to give players control over their own destiny, if you want to. As the designer, you are in control of how much control they can have.