City of Brass – When Is a Card Not a Card?

Tuesday, October 19th – Ancestral Recall is in contention for the distinction of “Best Card in Magic the Gathering.” It’s not a card one lightly says “is terrible.” But sometimes it is.

If you’re reading this, you’re paying for SCG Premium. Or it’s months from now, and you bothered to check back through the archives. Or you’ve convinced some ne’er-do-well friend to email you the article. In any of these cases, you’re probably a Good Little Magic Player; and as a Good Little Magic Player you probably know something about card advantage. You know that cards are A Good Thing and more cards are A Better Thing and sometimes you make a two-for-one trade, and we all get to give each other high fives and go out for milkshakes after the game.

Card advantage is a great shortcut. When we’re first figuring out how to win matches, card advantage is a way to take a complex situation and quickly figure out “what’s a good play?” At some point in your growth as a player, though, this ceases to be a relevant question. At some point, the question changes to “what’s the right play?”

Zvi wrote an amazing article on the subject of value where he states the rule of reflection. “Everything in Magic has value equal to the chance it will win you the game.” This sounds obvious, but there are important implications. By contrast, this means something does not have value equal to the number of cards it draws, or the amount of mana it generates, or the amount of damage it deals.

Just like your chance of winning the game, this value is dynamic. It changes as the game changes, and the game changes any time anything happens. In a tournament situation the game changes even when nothing happens — because some strategies become less viable or more critical with less time on the clock.

When is a card not a card?

When it sucks. When testing or discussing the game with friends, I’ll use the phrase “that’s not a real card.” The expression is used in the context of what I’ve been writing about. When a card has extremely low value, because the cost isn’t worth what you get for it, it might as well not be a card. Obviously you’re not going around putting terrible cards in your deck on purpose, but value is contextual. Mana can be useless in the late game, and many cards are great in one matchup but terrible in another. Spells that use cards as a resource, like Force of Will and Brainstorm, can be used to turn anything “into a real card,” and the decisions you and your opponent make can quickly change the value of anything and everything.

At five life with a Rack and Ruin in hand, you won’t hesitate to kill one of your own artifacts to kill your opponent’s lone attacking Juggernaut. Despite the fact that this is card disadvantage, you’ll make this play every time and would be hard-pressed to find someone who disagreed. This is because Mox Pearl and Juggernaut have vastly different values here. Whatever benefit you gain from keeping a Mox Pearl in play is dwarfed by the fact that you’re not dead. You’re simply trading your resources for his, and you’re trading up.

It’d be easy to look at this situation and think, “Well that sucks; I’m down a card now,” but that’s not entirely accurate. Without that Rack and Ruin, that Mox in play has little value. If you were short on lands it would have more, and if you were flooded it would have less. With a Juggernaut on the board turned sideways at five life, the Mox has value equal to the chance it will win you the game: zero. Now when Rack and Ruin comes along, suddenly that Mox skyrockets. Now you can trade it to not lose, making it infinitely more valuable, literally. When you cast that Rack and Ruin, you’re down cards, but you’re up “real cards” — you’ve increased the total value of your resources.

Dark Confidant in Counterbalance, and focusing on what matters.

Around the time Patrick Chapin was championing Next Level Blue in Extended, I was getting more serious about Legacy. Pulling from Extended, I played a Counterbalance list with Chrome Mox and Dark Confidant, and had some success playing variants of that. I was pretty satisfied with the list, until I sat down against Kevin Binswanger for a few games in between the rounds of a local tournament. He was also playing a Counterbalance deck with white cards where I had black. I hadn’t tested the match much at all before, as Counterbalance decks weren’t very popular at this point in time. I assumed that my Confidants would give me the edge here, as drawing cards is supposed to be good in control mirrors — but I was surprised when I lost game after game with them.

Kevin was already running Swords to Plowshares and Threads of Disloyalty, as a way of winning the fight over Tarmogoyf — this meant my Confidants weren’t often netting too many cards. Some games though, they did stick around. I was able to drop one early, or stop his Threads, or Threads them back myself. I played a game where I drew seven to ten more cards than him, and still didn’t win. The reason was, if Kevin resolved Counterbalance and Sensei’s Divining Top, the cards I was drawing did nothing for me. I could draw fifteen cards, and if they all cost zero, one, or two, they were irrelevant. With that Counterbalance on the other side of the table, my cards dropped drastically in value. While I was drawing two or three cards a turn and Kevin was drawing one, each turn he was getting more value out of his draw step because his one playable card was worth more than all three of mine.

This was a revelation for me. The best card advantage engine in Legacy wasn’t Dark Confidant or Fact or Fiction or Standstill — it was Counterbalance. There’s certainly nothing inherently wrong with drawing cards — but you can often gain more advantage by changing the cards in your deck, increasing the average value. So I took another look at the matchup. Adding even one card that’s relevant under a Counterbalance changes the game quite a bit. Just one Krosan Grip for instance, may not seem like a lot. Consider though, that it changes the value of all of your Brainstorms and Tops. Even after a Counterbalance lands, if you resolve a Sensei’s Divining Top first, now your fetchlands are much more valuable as well.

Trinket Mage was a perfect fit also. Not only is a Trinket Mage more likely to dodge a Counterbalance than a Dark Confidant, it finds you exactly want you want, an Explosives that can come down through the Counterbalance lock, kill the enchantment, and leave Trinket Mage around as a clock.

Now Counterbalance wasn’t a guaranteed win. In many games I played, one player would drop a Tarmogoyf before the other player established Counterbalance control, and the Counterbalance player would have to dig for an answer. Vedalken Shackles and Sower of Temptation proved incredibly strong in these situations, and unlike more popular removal spells like Swords to Plowshares, they often completely circumnavigated Counterbalance.

Of course, being so impressed with the card, I wanted to resolve my own Counterbalances too. Immediately this changed my decklist from three Counterbalances and three Tops to four and four, but trying to maximize the chances of sticking the lock was the nail in the coffin for Bob. Resolving Counterbalance first not only means you have your own, but it also makes it much harder for them to resolve theirs. Ponder just did a better job of finding a Counterbalance or Tarmogoyf to play on turn 2. Even though it doesn’t generate any card advantage, it was better at creating game states where the total cards I had were worth more than the total cards my opponent had.

I took that list at GP Chicago, where, ironically, I lost in the finals to a Counterbalance deck running Dark Confidant. This story would be a lot better if I had won, but ultimately I’m sure I was on the right list. To get to the finals I had to defeat six other Counterbalance decks, including David Ochoa and then-highest-global-rating Louis Scott Vargis, both running the exact maindeck Nassif had.

Ancestral Recall, and what’s in a card.

Ancestral Recall is in contention for the distinction of “Best Card in Magic the Gathering.” It costs thousands of dollars, is run in almost any deck that can cast it, has gotten other cards restricted purely because they can find you your one copy, and people have run Misdirection against decks with no other targets. It’s not a card one lightly says “is terrible.” But sometimes it is.

When Gifts Ungiven was legal in Vintage, Ben Kowal and I were looking for an edge in the mirror match, and we sat down to test. As the evening progressed, we noticed something peculiar, the player who resolved Ancestral Recall would consistently lose. This was obviously confusing. Ancestral generates a big advantage for a really small investment of resources; people at the time were warping their decks around finding and resolving it quickly — but we just weren’t getting those results. Ben came to the conclusion that Ancestral Recall was terrible, and when I played it, he stopped caring, and he started winning.

We quickly realized that it wasn’t the Ancestrals that were causing problems; it was fighting over them. We were burning Force of Wills and Mana Drains on a spell that ultimately didn’t matter. It wasn’t that Ancestral Recall isn’t extremely good, but conditionally, other cards are better. With some minimum amount of resources, either lands or specific cards in hand, Gifts Ungiven would just win — and that wasn’t terribly hard to come by. When one player was about to win, Force of Will was just more important than three random cards.

Later, I was testing one of the first Repeal Gifts lists against Rich Shay, and neither of us had had much experience with the card. I broke a Black Lotus to cast a Yawgmoth’s Will; Rich was able to counter it, but the ensuing war left both our hands empty. On my next turn, I drew Recoup. This let me flashback the Yawgmoth’s Will, but with no extra mana, unlike the five I had the turn before. I played a Polluted Delta and the Black Lotus from the graveyard, and went into the tank.

“Just play the Ancestral,” Rich said, which seemed like the obvious play. I wasn’t quite sure how, but I thought I could do better.

I must have taken a while to think, as quickly there was a chorus of spectators trying to push the game forward. “What are you doing? Just play the Ancestral!”

“Just see what you’re going to draw already!”

After an embarrassing amount of time, I fetched out an Underground Sea and broke the Lotus for black mana. I cast Vampiric Tutor, Repealed a Mana Crypt, and replayed it to generate an extra mana, and cast the lethal Tendrils of Agony I had found with Vampiric (Rich had dealt a few damage to himself already this game.) This kind of Repeal play isn’t terribly uncommon now, but at the time, neither of us was used to using the card like that. Casting Ancestral would have turned a win into a “pretty decent situation.”

This isn’t to say “Ancestral Recall is bad” or “never use a Force of Will to defend Ancestral Recall.” Just the opposite, this illustrates why hard rules about card advantage don’t always cut it.

Ancestral is worth three random cards, but what are three random cards worth? The value of a random, unknown card is just the average value of all cards left in your deck. This is why it’s important to understand how the values of all those cards are dynamic, and constantly changing. When you’re making a decision, you need to evaluate what those random cards are worth, and compare them to what you’re giving up to get them.

In the case of the Gifts mirror, Force of Will is just more valuable than three random cards. If you can derive through Duress or otherwise that this isn’t the case, that neither player is in a position to win off of a single spell, those three random cards have more value. For instance, if one player is bottlenecked on mana, the value of their mana sources shoot up — this greatly increases the average value of their unknown, random cards. In these situations, countering Ancestral Recall becomes a lot more appealing.

In the case of Repeal into Tendrils of Agony, the cost of Ancestral Recall is only the mana spent to cast it. While usually negligible, in this scenario that one mana was necessary for a win, which means leaving it open is far more valuable than the three random cards Ancestral trades it for. Even the slightest change warps those values though. If Rich were at one more life, that Tendrils wouldn’t kill him. Consider that this Repeal Gifts deck had only two ways to win, Tendrils and Darksteel Colossus — one life isn’t much worse for him than 11. While bringing Rich to one would shut off fetchlands and Force of Wills, those three random cards start to be worth that one mana. In fact, you could use the same Vamp-into-Repeal trick to cast both Ancestral and Tinker that turn, which is clearly better than a nonlethal Tendrils.


Cards are great, but getting them or denying them has a cost. For paying that cost to be correct, it has to be less than the value of those cards, which means you have to know how much those cards are worth. Most of the time, calculating the exact odds of winning a game is an impossible superhuman task. As the game progresses it gets easier, more information becomes public, and life totals shrink. Many games converge to a “got one turn to draw my out” point, which is fairly easy to figure out, but for the bulk of the game and the bulk of the difficult decisions you make, things are not so straightforward.

This is the primary value of testing. By paying attention to the game states that lead to a win or a loss, you can more accurately value cards in a particular matchup by how likely they are to create those game states.

When you draw a card, consider not only what it does, but how it changes the value of cards that have already been played. When your opponent plays something unexpected, consider not only how that spell affects the game, but how that changes the value of unknown cards in his deck — one unexpected card often infers the presence of more.

While this isn’t new information for everyone, it’s essential and fundamental. Until next time, may your cards be more valuable than your opponent’s!

Andy Probasco