SQUARE PEG THEORY
OR
THE PROBLEM OF DIMINISHING RETURNS


TO: Frank [webmaster@classicdojo.org]
FROM: Andrew Wills [awills@pathsys.com]
DATE: 7/15/98
RE: Some thoughts - Magic Theory

What determines duels? Ok, now what determines duels besides mana screw? Is it Engineering? Is it play skills? For the purposes of this exercise, I am going to assume it's not cheating.

I can identify a handful of distinct activities that I find myself doing in connection with every constructed deck I play:

Please observe that I don't consider these to be stages, but 'activities.' I do think these activities happen roughly in the order outlined above. But the choices of one activity have consequences that affect the possibilities of another. And sometimes I find myself reevaluating the earlier choices to accommodate the goals of a later activity.

Let me give you an example of what I mean. You're choosing between Suicide Black and WW. You know that, if you choose WW, you can support 3-4 extra artifact/enchantment removal in the sideboard. If you go with the Black, you can board 4 Dread of Night.

The decision between white and black is Archetype Selection. But building a sideboard is something that takes place later.

Archetype Selection: For the most part, I do not come up with central deck concepts myself. And I really don't consider this a problem. There are enough other areas where I have to do my own work, I think, to keep MTG fun and challenging to me indefinitely.

There is, of course, a species of metagaming that takes place here. But it's a different activity from the metagaming that goes on within a chosen archetype (intra-archetype).

Engineering: Well, I'm not even sure I can tell you what this concept means for certain. I think of it mostly as optimizing a chosen archetype without reference to a single or even a short list of opposing archetypes. Engineering is when you figure out how much land you want to see in your opening hand and how much you need to include in your deck in consequence. How much manna of my off color should I include? Which is better, death stroke or terror? These are the kinds of issues you come to grips with when you're engineering.

Engineering is important, but the truth is that now everyone can do it, or get it done for them. Some do it better than others, for sure. But my feeling is that most competitive MTG players soon reach a point of diminishing returns. Engineering is something they have to do lest they loose. Not something that, if they do it well, they stand to win.

Play skills are even less relevant in a constructed environment. Honestly, you and all of your opponents are on automatic pilot. It's not that folks play badly in constructed environments. To the contrary, the archetypes are both simple and familiar to the point that most competitors are in little danger of playing poorly. At least not poorly enough to make a difference.

I can't tell you how many times I've heard:

Or sometimes I hear:

Likewise, I read tournament reports where the author blasts himself for a mistake in the eighth round when he's already 5-2.

Don't get me wrong, I don't mean to suggest that sloppy play is they key to success. But what are you going to say when you stop making mistakes and you still loose? My point is that you can't ask MTG to let you win if only you don't make play related errors. Improving your play is another exercise in diminishing returns. If you ask your deck to win for you as long as you play the cards optimally, you're asking too much. You're just looking for results in the wrong place.

Those of you who are still reasonably awake will have concluded by this point that I mean to discuss that activity which I referred to as 'Intra-archetype Metagaming' above. What do I mean by 'intra-archetype?'

You see, most of the times I hear someone talk about how he is playing the metagame, it is clear that he is speaking about the activity of Archetype Selection. "Sligh beats Necro, Necro beats control," etc.

Quoth Mr. Pikula at the Type II portion of Dallas:

"This is the biggest game of Rock-paper-scissors ever!"

Rock-paper-scissors. Pairings. Metagaming. This is the vocabulary we have for talking about the fact that some combinations of (competitive) cards just seem to dominate others. But I almost always hear this phenomenon discussed exclusively at the level of entire deck archetypes. From where I'm standing, this is a mistake.

My point is that the activity of Intra-archetype Metagaming will yield the largest improvement in results per amount of effort. Basically, I think it is the modern game.

There was a time when many tournament players played very poorly. They made horrible decks. "At the end of your turn, I'll Tim my Spiny Starfish and regenerate it to make a token." "I'll put Psychic Venom on your Plains and cast an Icy Manipulator. Go!" Winning was easier then. Now we have the so called Dojo Effect. But honestly, my feeling is that a much larger portion of the MTG community knows how to evaluate the worth of a card or concept accurately, or at least more so.

Let me see if I can come up with more concrete examples.

Here's the first deck I qualified with. I think I can remember it to the card.

4 Llanowar
4 Fyndhorn
4 Birds

4 Er'nham
4 Dervish

1 Sylvan

4 StP
4 Disenchant
3 Land Tax
4 Armageddon

1 Zuran Orb
1 Ivory Tower

4 Strip Mine
4 Brushland
9 Forest
5 Plains

At least I think that was it. It was Black Summer in Ohio. I think I played 4 Necro decks in a row.

Ok, now here's the Underwood deck:

4 Priest
2 Chicken

4 Monk
4 W.Knight

4 W.En-core
2 Sky Spirit

4 Tithe
2 Earthquake
3 Disenchant
1 E.Tutor

4 Mox Diamond
4 Scroll

4 Wasteland
2 City
2 Undiscovered
2 Gemstone
12 Plains

There are two decks I've experienced some success with. Now I'll tell you why I think they were successful.

Think of it this way. Your opponent comes to a tournament with a piece of plywood having several holes drilled into it. You need to be prepared to offer your opponent that square peg that won't fit into any of his holes. Don't try to out engineer him, don't try to out play him. You may succeed in doing so, but nevertheless loose the match. Try very hard to present him with something he can't handle very well.

Protection creatures are fabulous examples of these 'square pegs.' In the early stages of a duel, a priest and an untapped en-core can dominate a Sligh deck. And 2nd turn Dervishes gave the old monochrome Necro decks fits.

Here's another example. Remember the extended qualifiers? Remember PT Jank? Very smooth deck. Made a big impression. It made such an impression that, by the end of the PTQ cycle, some 40% of the (competitive) field was playing it. All of a sudden, the Jank decks that included Wildfire Emissary started dominating. Other decks with Wildfire Emissary started doing better too.

And there are other cards, besides protection creatures, that have this dominating potential. Most by now realize Armageddon is devastating to control. Also, a 2nd turn Miner can wreck U5C donais. Firestorm is at its best against suicide black.

The trick to winning, as I see it, is to understand the dynamics of how cards like these get under the skin of otherwise sound archetypes. Once you have some expectation concerning what the other competitors at a tournament are likely to run (play), you need to reconsider the product of your engineering. You need to maximize the potential your deck has to dominate the other decks in the field.

Obviously, no part of your deck requires more consideration of this nature than the sideboard. I have observed some players include cards like:

  • Extra counterspells
  • Extra creatures
  • Primal Order

This is not right. Put in Gloom. Gloom is an excellent example of the calibre of card you want to include in your sideboard. Gloom wins matches. Gloom has the potential to destroy someone's game. If it doesn't do something amazing, it shouldn't be in your sideboard.

But don't limit this kind of analysis to building the sideboard. You need to maximize the potential for domination, both in the sideboard and in the main deck.

The conventional wisdom provides that the card Gloom belongs in the sideboard. I feel fairly confidant that, except in rare cases, this is correct. The same thing is true for Pyroblast, Dread of Night, and Warmth. These cards are fabulously thorough, but against an unknown deck they have a great probability to end up dead cards.

For the most part main deck Intra-archetype metagaming means creatures. Take for example the Underwood deck. From an Engineering standpoint, what the deck wants is 20 or so two casting cost, two power creatures. These creatures are efficient both in terms of cards and casting cost. And there were certainly enough to choose from:

  • Priest
  • Monk
  • Warrior en-Kor
  • W.Knight
  • Pump Knight
  • Longbow

6 * 4 = 24 potential creatures in this category (off the top of my head). Why, then, did the deck have 2 Chickens and 2 Sky Spirits instead of any Pump Knights or Longbows? Simple. Against the decks I expected to face, Penetration and Protection from Red were more effective than the small sacrifice in efficiency.

This is the kind of analysis, I think, that offers the greatest rewards in the contemporary environment.
[Ed. Note - This is the kind of analysis that will take your game from also-ran to contender.]

--awills