'Incoherent' metagame

[Dante]

I think that in a game of incomplete information, like Magic or Poker, it's much harder to come to a determination of things like "best play" vs. things like Chess.  In a game like Chess, there is no "topdecking" to bring in a surprise Queen (Yag Will) out of nowhere.  You can plan a whole strategy line and play out every outcome.  This is much harder in a game like Magic where your resources are drawn and used semi-randomly.

My post from a similar thread in the open forum

However, combo_dude, I don't see why a perfect mathematical model of Magic would be impossible. It would be based on statistics and probability, and it would require awe-inspiring amounts of computational power, but it should be just as theoretically possible as a mathematical model of Poker, for example. Though, again, it would require vastly, vastly more computational power because the "deck" wouldn't be defined in advance--though the possible contents of the deck (the cards legal in the format, and in what quantities) would be.

The problem is that Chess is a game of total knowledge, not partial knowledge like Magic and Poker. What I mean by "total knowledge" is that all the information about all future moves are available to both opponents (and observers). Thus a reasonable powerful computer can compute, based on all possible outcomes, the top strategy and particular move.

Magic and Poker have incomplete information, where even if you not only knew your own hand, but your opponent's hand, any forumla or strategy could/would be susceptible to totally change after each card is drawn, as opposed to Chess, where every specific move can be calculated EXACLTY since there is no hidden information.

This is why Chess computer programs can play to a Master-level and beyond, whereas Poker programs are mediocre (at best) that can't really hold up to live players (and I've tried a lot of them).

Bill

[Saucemaster]

The problem is that Chess is a game of total knowledge, not partial knowledge like Magic and Poker. What I mean by "total knowledge" is that all the information about all future moves are available to both opponents (and observers). Thus a reasonable powerful computer can compute, based on all possible outcomes, the top strategy and particular move.

Magic and Poker have incomplete information, where even if you not only knew your own hand, but your opponent's hand, any forumla or strategy could/would be susceptible to totally change after each card is drawn, as opposed to Chess, where every specific move can be calculated EXACLTY since there is no hidden information.

This is why Chess computer programs can play to a Master-level and beyond, whereas Poker programs are mediocre (at best) that can't really hold up to live players (and I've tried a lot of them).

Bill

Hmm, but there is a type of strictly mathematical "optimal" play in Poker.  If you bet solely on money-making hands--where your pot odds, or I suppose your implied pot odds, are favorable--then you'd make money.  The PROBLEM with this is that, of course, actual people learn and will purposely deceive you in order to make you fold or underplay your strong hands and miscalculate the odds, etc.  Interestingly, this is just like the metagame adjusting to the best strategies/decks; what your opponents are doing is actually just changing the odds.  They're making it too expensive for you to play certain hands, too cheap to play others, etc., and their play has actually changed the odds you're receiving in much the same way as the reaction of other players to the dominant decks in a format changes the nature of the game the dominant decks are playing.  The most fascinating thing about Sklansky's "The Theory of Poker", to me at least, the first time I read it, was how much of the game actually IS quantifiable and the degree to which it behaves in entirely well-understood ways.  A good part of the attraction in Poker, and in Magic, comes from what happens when humans, who are totally capable of irrational action (e.g. who bluff more often than is "theoretically correct", or who always play too tight, etc.), have to face each other and find a way to act rationally in response to their opponents' potentially irrational actions, so as not to lose to decisions that "shouldn't" beat them.  To this degree, I agree with Milton.  Alot of what makes Magic fun is that it ISN'T actually a game that is played perfectly and about which we can all easily agree.  That doesn't change the fact that it's capable of being modeled and dissected and analyzed and deciphered, and since those with more understanding of the game will always do better than those wth less, it honestly surprises me that there's not more of an "arms race" to understand the game better.

Anyway, it's worth noting that the biggest distinction between "correct" play in a game of chance and partial information vs. "correct" play in a determinate game of total information is that "correct play" in the first will just lead you to win more often than not, whereas "correct play" in the second will ALWAYS lead you to win.

[Dante]

Hmm, but there is a type of strictly mathematical "optimal" play in Poker.  If you bet solely on money-making hands--where your pot odds, or I suppose your implied pot odds, are favorable--then you'd make money.  The PROBLEM with this is that, of course, actual people learn and will purposely deceive you in order to make you fold or underplay your strong hands and miscalculate the odds, etc.  Interestingly, this is just like the metagame adjusting to the best strategies/decks; what your opponents are doing is actually just changing the odds.  They're making it too expensive for you to play certain hands, too cheap to play others, etc., and their play has actually changed the odds you're receiving in much the same way as the reaction of other players to the dominant decks in a format changes the nature of the game the dominant decks are playing.  The most fascinating thing about Sklansky's "The Theory of Poker", to me at least, the first time I read it, was how much of the game actually IS quantifiable and the degree to which it behaves in entirely well-understood ways.  A good part of the attraction in Poker, and in Magic, comes from what happens when humans, who are totally capable of irrational action (e.g. who bluff more often than is "theoretically correct", or who always play too tight, etc.), have to face each other and find a way to act rationally in response to their opponents' potentially irrational actions, so as not to lose to decisions that "shouldn't" beat them.  To this degree, I agree with Milton.  Alot of what makes Magic fun is that it ISN'T actually a game that is played perfectly and about which we can all easily agree.  That doesn't change the fact that it's capable of being modeled and dissected and analyzed and deciphered, and since those with more understanding of the game will always do better than those wth less, it honestly surprises me that there's not more of an "arms race" to understand the game better.

Yes, a lot of poker and Magic is quantifiable, but it's the rest of it that makes it interesting to a broad audience.  Playing ABC poker will only get you so far in poker, same with Magic.

Also, things like "bluff more than is correct" or "play too tight" are all based on assumptions of your opponents' play, which right there is a skill in-and-of itself, not to mention the adjustments that need to be done based on those reads.

[Smmenen]

Here is what I wrote in an article recently:

In fact, it goes further than that. While some decks are better than others, far too many decks are viable simply because most of the best decks share with each other a key set of the most powerful cards in the format. Oshawa Stompy is half U/G Madness, which is 1/3 Fish, which is half Landstill, which is half Keeper, which is half EBA, which is half Tog, which is half Drain Slaver, which is half Workshop Slavery, which is half Draw7, which is half Belcher... okay, you got the idea.

The point I'm making is that there is that there is a small set of synergies and card alliances that basically make up Type One at the moment.  

Incoherence?  Two things happening.  First, people are mixing and matching these cards bits.  People combined Dragon and Madness.  They put Fish and Madness (UGmadness).  

What is my point?  This metagame is NOT! Incoherent.  It appears that way because of the multiplicity of decks.  But when you examine the core components, there are basically only half a dozen or so decks out there.  

As for the supercomputer issue:  If a level 20 intelligence sat down and analyzed Type One it would be so broken your head would be spinning.  Same with Extended.  Pro Tour teams can only go so far because they are human and miss things.  (as was evident last year).  

I'm not certain, but I think I beleive that there is at any one time a truly optimal strategy for Type One (regardless of metagame variation).  This can lead to some self-contradictoins which are resolved when you realize that there are degrees of dominant strategies.  

Steve

[DavidHernandez]

It seems to me that over the years we have seen that certain card combinations produce decks that "can't be beat".

For example (in their 'original' incarnations):
Academy
Long
GAT
Necro

...and decks that hold their own no matter what anyone else plays:
The Deck/Keeper

In between those decks are the "random innovations" and the "hate" builds.

Regarding "Decks that can't be beat", just about anyone could play one of the listed decks and win, regardless of their own (or their opponents) play skill, and regardless of the Random and Hate decks.

This shows that the card pool can be unlocked to yield a best/defining deck, and that such a build will define the Metagame.  This in turn will force tweaks to the Random and Hate builds, and will require Keeper to modify it's sideboard.

So, to answer G.I.'s question:

What is more important to success, the cards you play (best deck), or the cards your opponent plays (metagame)?

I say that the answer is "the cards you play".  By piloting the best, strategically synergistic pile of cards, you will control the direction of the Metagame.

Currently, there is no "broken" deck.  There is a semblance of "balance", and I believe that this is why the Meta appears "incoherent".  

However, at some point, a broken deck will make it's appearance, and another set of restrictions will be required.  If history is an indication (and I think it is), this is not a case of "if", but of "when".

--Dave.

[kirdape3]

There are three utterly degenerate decks in the environment.  This is actually a good thing - it's when there's only one is when the BANHAMMAR has to show up.  Those decks are Tog (when a control deck kills turn 4, it's degenerate), either Slaver build, and Draw7 combo.  Hey look, that's a control deck, a Prison deck, and a combo deck.  Seems like a fine metagame to me!

But wait you ask?  Where's aggro?  It does exist, and some of it is quite good.  But the replacement for Food Chain Goblins in the metagame is really Draw7 (it's just faster and is pretty resilient to hate by just winning).  If anybody played the deck (they should, since it wins a whole lot), then you'd see FCG and probably U/G Madness drop right out.