Zac Hill’s Loss is Your Gain

So based on a couple of reader requests (I’m looking at you, CMH2003 and thewachman), plus an absolutely glowing testimonial from Rich Hagon “the best thing that happened to Magic writing in the last year was Zac falling off the train and going to Malaysia” I went back and looked at Interaction Advantage (or Card Advantage Is Wrong, Kinda) and Applied Interaction Advantage, its sequel from this week.

My general impression is that there isn’t really anything wrong with Zac’s theory… only that it isn’t new.

A lot of you probably didn’t have Brainburst Premium when I was writing for that site (which was around the heyday of Who’s the Beatdown II: Multitasking and some other fine articles from Zvi), but to me, this is fairly settled stuff.

Hill attempts in his Interaction Advantage to “reconcile the competing existing concepts of Card Advantage and Card Quality with one another in a coherent fashion, while incorporating the reality that certain effects have an impact on the board that far exceeds the expected interaction value of your average ‘card,’ but aren’t easily measured by the other ‘cards’ they destroy, negate, or generate.” This is in large part a reaction to Stephen Menendian’s claim that “In fact, within the game of Magic, there are only interactions. There is no such thing as a ‘Magic card’. This is a difficult ontological reality for many Magic players to accept.”

I suppose in order to properly address Zac’s article I will have to get Stephen’s statement out of the way first.

It’s asinine.

Okay, that’s probably a little harsh.

There are obviouslycards. In fact, the cards say “cards” all over them (look at any “discard” card). There have to be cards. It’s silly to pretend that there aren’t cards, because even if you agree with Stephen’s ultimate claim, the physical cards themselves have tremendous value “within [a] game of Magic.” If nothing else, cards give you something to bluff with. They give you something to look at when assessing if your opponent “has it” or not. They give you something to draw extra of if you are a cheater, or to watch that your opponent isn’t drawing extra of if you suspect him of being a cheater. They give you something to draw extra of if you are not a cheater but don’t know what else to do with the five Blue mana staring at you. The physicality, the reality, of the cards is essential to interactive Magic and to pretend that they don’t exist is to remove from the game that single solitary thing that differentiates the very good players from the absolute masters of the game. I can tell you that Jon Finkel walked by a playtest session once and ordered us to stop. We were playing poker deck with a Survival of the Fittest list. Jon insisted that what we were doing was actually hurtingus, and that there was no reason to test a Survival deck with proxies. “You just don’t have the right feel for the cards, and never will.”

Jon at the time had numerous Grand Prix wins and Pro Tour Top 8s on the back of his unparalleled skill with Survival of the Fittest, and to this day claims the absolute mastery of Survival of the Fittest among his proudest capabilities as the game’s greatest technician. If there was any one person’s opinion that absolutely mattered on this topic, it was his, irrevocably.

And he believed in cards.

But that probably wasn’t what Stephen was about in his statement.

He was really just being over dramatic about card counting, and I suppose that is forgivable insofar as what he was “getting at” (I assume) is essentially true from a pure counting perspective. It was true the last time I said it, and when edt told it to me, too. Just not new. I stress that this is accurate only from a counting perspective, that is, the recording of exchanges, and not true in any other sense.

In the early to mid 2000s I was obsessed with counting and argued that all card advantage is ultimately virtual. From the perspective of instantaneous utility there is no difference between playing Keiga, the Tide Star scaring off three attackers and playing Wrath of God to kill three attackers. You might be shaking your head at this because in one case there are still potential attackers on the board and in the latter case there aren’t. It is ultimately futile to argue the point because I can just pretend they will never attack and therefore never have utility (what if I play Windborn Muse and the opponent chooses never to attack with them due to mana constraints?) … Or what if the opponent plays Living Death and gets them all back? We can really only look at things at a given instant because the idea of long-term card counting really comes down to where you stick the words THE END … It’s the difference between comedy and tragedy and all the difference in a game of Magic.

Have you ever played ninety-nine one hundredths of a perfect game of Magic? Your back to the wall the whole time, mana flooded, desperately chump blocking, holding on until you weasel into a topdeck, then still have to juke it perfectly for three turns, pray he doesn’t realize you’re bluffing (there are those cards again) until you pull again… Only to mis-click on the last turn (real-life mis-click counts too)? Congratulations. Ninety-nine out of one hundred. Wasn’t good enough.

Okay, back to Zac…

Not a bad set of articles. I’m sure many readers benefitted from reading these, and that is all we can ask of ourselves as writers. Not bad; simply not new.

In my mind Card Advantage and this nebulous idea of “Card Quality” don’t have to be reconciled.

Correctly counted, all card advantage is virtual, cardboard or no.

The easiest way to describe it is this:

Turn one of a Standard game, summer of 1999.

Your opponent plays Swamp, Blood Pet. The table popped up before he even played his Swamp, and he can’t possibly contain his excitement or disguise his grin.

Okay, you think.

Seven cards, not one of them a Mogg Fanatic. For the sake of this example, you have four Mogg Fanatics in your deck, but anything else, you’re dead.

You have 53 cards in the stack in front of you, or put more simply, a 75% chance of losing on the spot.

You pull “a card” … that is, a piece of cardboard (it exists). It is not, however, a Mogg Fanatic.

For sake of counting — real honest to goodness counting — you pulled nothing. You pulled a blank. It’s as if you didn’t draw at all.

If your are reading my blog, I am fairly sure you understand this example. Now I am going to make you a better Magic player. For reals.


All the counting in Magic is instantaneous. There is no difference between being too scared to set off a Veiled Serpent and simply not having any hand at all from being Mind Sludged beyond the option of someday NOT being too scared of setting off a Veiled Serpent and manning up (or bluffing that you might someday similarly man up). Player behavior dictates everything. It always has.


You understood the summer of 1999 example because of the virtual boner the opponent showed you, your knowledge of your four Mogg Fanatic deck (probably Counter-Phoenix), and the fact that you realized that you were staring at a textbook Hatred kill. Now even with these parameters filled, it would have been up to player behavior. Maybe you could bluff a Shock or a Force Spike. Maybe the opponent doesn’t know that you have no out there. But you, the guy reading this blog, understood from the example that the in-game utility you got from a non-Mogg Fanatic pull was zero.

The difference between you before you read this post and you after reading the next paragraph is that you will understand that the lesson of Bob Maher is that you should always assess your cards as if you are under the exact same kind of pressure.

The value of card drawing is tied directly to probability. More cards give you more chances at relevant pulls. Playing “as if you are under pressure” leads to crafting a strategy that will necessarily conform to the threats and interactions that your opponent can present within the appropriate time frame. This is actually quite simple if you think about it.

We often go back to the timeless Finkel message “Focus only on what matters.” Certain pundits have complained on occasion that they don’t know what matters. It may be simpler to think of the relevant interactions. You all understood the relevance of the Mogg Fanatic versus Blood Pet scenario. Every Magic interaction making up every Magic game can be broken down into similar buckets of relevance and probability, which in turn fit into larger buckets that make up the Stage superstructure that describes every Magic game.

For example take MWC v. Zoo in Extended:

MWC will win essentially every game that is allowed to go to Stage Three. How MWC wins is more-or-less irrelevant because there is almost no play that is bad enough that Zoo would be able to come back and win if it gets that far (I like making the opponent die to his stupid Dark Confidant, personally).

MWC has a suprising Stage One. Not good, but surprising (you are surprised when you lose to Mana Tithe, and also surprised when you lose to Lightning Helix).

Zoo crosses Stage One superbly, and acheives Stage Two within two turns most games. Zoo has to try to win in Stage Two. This statement is the entire framework of the strategy that guides both decks.

When players say they don’t know what is relevant — at least if they were playing one of the two decks described here — it is possible they are failing in the basic identification of strategy.

Zoo can only win one way: Kill the opponent before he acheives Stage Three, likely with a combination of beatdown and burn spells. One of my favorite things when I was actively playing the MWC deck was when my Zoo opponent would play Umezawa’s Jitte, typically in an attempt to not play into a big old Wrath of God. I liked this because I never lost a game when the opponent went Jitte. He would cede time by not killing me. I would react with Lightning Helix or Unmake (or sometimes Condemn) and buy a precious turn that brought me closer to Stage Three.

There are two cards that are sometimes played in Zoo that could complicate issues for MWC. Both Tidehollow Sculler and Gaddock Teeg could remove MWC’s ability to interact in Stage Two (typically pre-empting or removing Wrath of God); without such ability to interact, MWC would die before Stage Three (obviously losing in the process).

Naya Burn and the Lightning Bolt Deck have different ways of interacting with MWC than Zoo.

Naya Burn can play Molten Rain — especially on a Mistveil Plains or Temple of the False God — which restricts MWC in Stage Two (or even shanks MWC back to a manascrewed Stage One in some cases); either deck can play Sulfurous Vortex, which can not only kill MWC when MWC should be trumping in Stage Three (and certainly Stage Two), but can undermine the active dictation of Stage Three (whereas Zoo would always lose in Stage Three).

You will notice that the evaluation statement that describes both decks’ strategies differs when MWC is playing against Naya Burn (which looks like Zoo). If the MWC player assumes “[this deck] has to try to win in Stage Two” MWC may fail. MWC may in fact inevitably fail. In fact, this may be tantamount to a mis-assignment of role! What does it mean if Naya Burn can violate Stage Three airspace? We are in a very different world than “MWC will win essentially every game that is allowed to go to Stage Three,” aren’t we? It sounds almost like racing (God forbid) could be a relevant if not necessary option in our bundle of sticks, an arrow in our quiver we just might have to string up.

Think about what is necessary to identify the necessity of racing in the MWC v. Naya Burn fight. At the micro level, you have to figure out which if any of your cards is appropriate for racing and how and when to play them. At a slightly wider level you have to identify how to either find cards to race or how to prevent Naya Burn from racing you (or perhaps if you are very spoiled) how to make the game no longer about racing. But at an even wider level, you have to have correctly assessed that the paradigm of the game is a little bit different than playing against Zoo. When you are very good, the appearance of a Sulfurous Vortex will begin a domino cascade in your mind that will inform your next five turns’ of tactics. A lesser player will simply lose and not realize how he lost, having drawn all his best cards “against Zoo.”

Those of you who followed me on MTGO when I was actively playing MWC know that I was sideboarding as many as four copies of Kataki, War’s Wage, not for Affinity, but primarily for the Lightning Bolt Deck. Kataki did several things in the Lightning Bolt matchup, over and over, even when it didn’t seem to make a lot of sense. Kataki was always a lightning rod. Or when he wasn’t, the opponent didn’t realize what was going on. Often, he would buy Shrapnel Blast! You see, the Lightning Bolt deck is full of Darksteel Citadels and Great Furnaces and really doesn’t have enough operating mana to let you run all over them with Kataki forever. Secondly, Kataki almost guaranteed that Mana Tithe was good. Especially once I had removed Oblivion Ring from my deck (Unmake was basically always better), I actually had to gamble on Mana Tithe being able to stop Sulfuric Vortex. Finally, Kataki gave me a little racing game. You’d be surprised! He’d get in for six or eight. He never killed the other guy, but the other guy was often so cavalier about his life total he didn’t realize he was in a race. Therefore he would have fewer turns than expected to draw his 20th point of burn and I could shave a turn or two off of my normally glacial Decree kill.

Last thing. This is super important!

Zac — we have to get back to Zac, Zac inspired this post — says in the second of his two referenced articles that he is not trying to describe a kind of “option advantage” with his theory. That is fine; what he is describing is his to describe. However remember that good Magic play is about the preservation of options. When you are presented with two similar plays, typically the one that leaves you more options is superior. This is the root of our theories on mana efficient play, card advantage, life total preservation… everything we think of as “clean” technical play is actually about generating and preserving our options.

And because bluffing is an option… You already knew there was such a thing as a card so I won’t bother repeating that.


P.S. I realize there are infinite difficult concepts in this post, and I will be happy to follow up more specifically on any topic any of you want me to in the comments.


Example Zoo
Example Lightning Bolt Deck
Example Naya Burn
Example MWC (I’ll write a separate update on this before the 21st… Might play it!)

facebook comments:


#1 KZipple on 02.13.09 at 11:42 am

Thanks for writing this. I had actually been thinking recently how all card advantage is virtual, so interaction theory is not really new. I just didn’t have a good way to codify that. The one thing that I didn’t like about this post was that, “EVERY SINGLE PULL YOU MAKES OPERATES THE EXACT SAME WAY.” I think the point is that every card you draw only matters to the extent that it actually does something [kills Blood Pet, removes Sulfuric Vortex, whatever], but that could use more explanation I think.
Anyway, very useful. Looking forward to a new MWC list.

#2 Aten on 02.13.09 at 1:14 pm

This is the best thing you have written in a long time.

#3 wrongwaygoback on 02.14.09 at 4:28 am

Suggesting there’s no such thing as a card is only slightly worse than the suggestion that everything has haste (which I had a think about here). If there’s no such thing as cards, then I’m assuming I’m gonna kick ass with my empty deck box at my next FNM (“look, you don’t understand… I win! I do! I didn’t even need a card to do it!).

Ok. I get it’s a metaphor. Of sorts. But Zac also uses the term “literally” in the worst possible way. So, there’s also that.

#4 thewachman on 02.14.09 at 9:21 pm

Thanks for your take on this.
I’ll digest all of these things and if I need follow up you’ll know.
Eric from Saint Louis

Listen to My Podcast at

#5 CMH2003 on 02.16.09 at 10:14 pm

Does this mean Im famous like Mike now?

You must log in to post a comment.