The Magic Workstation


Growing up, I spent a lot of time around my grandfather, who would be working on things out in his garage. His workbench spanned the length of the garage, and it was neatly organized with tools and things he had set aside for the future. He wasn’t so much using this space as an escape; this is where he got things done. He’d often be cutting wood for any number of projects, repairing something mechanical, or fixing something around the house, and when there was nothing more to do, he’d spend time with his family, sometimes with his hearing aid turned off, because . . . you know, it’s family. He was always apt to correct a problem, and his mentality helped set the gender role for me that men are supposed to work with their hands.

In the twenty-first century, few of us have the space, time, materials, or inclination to tackle problems of that nature, and so we find other things to engage ourselves in. As it turns out, whether you want to engage in incredibly complex interactions, simply do something better than someone else, or make the biggest splash possible, Magic has something to offer you. In Magic, we have the same need to create, shape, and refine our creations, but we have a very different (and at times limited) set of tools to meet this end. Today, I want to talk about one of those tools in specific: Magic Workstation.

The Big Idea

Magic Workstation (MWS) is a pivotal tool for Legacy players, as many Legacy decks are tested and primed using MWS as their primary source for nearly all data. What I wanted to do was to run my own pet deck against fifty decks on Workstation and record the results to see what we can learn about the merits of testing with the program with that information. Are we seeing a good saturation of competitive decks? Does the ability to play any card skew the metagame toward the more fringe decks?

Before I share those results with you, I want to talk a little bit first about my experiences with Magic Workstation over the last few years, and discuss some of the pros and cons that I see with the program as opposed to alternatives such as Cockatrice, Lackey, and OCTGN.

Overall, with Workstation, there is a much bigger inclination for people to play new, fringe, or rogue decks. This happens not only because it gets tedious to play against the same decks over and over with your specific pet deck, but because with limitless access to cards, you can quickly and easily build whatever catches your eye. You can almost instantly play it against something without inviting friends over, sorting through a collection of any kind, or having any impact on a rating. Because it’s so easy to build decks, and there is so little reprisal for doing poorly with a specific deck, you’ll tend to see the overall player skill be incredibly low. With no real incentive to learn a deck, people are content to “netdeck” something and misplay with it to their heart’s content. This does mean you’ll get more wins, but that the experience you get from the games is somewhere between negligible and minimal.

The other side of the coin is that people will want to play the decks they really own or intend to own against the field to see how it pans out—this results in the two most common decks on Workstation being Pox and burn. Anyone who plays on Workstation regularly can tell you that it’s certainly able to teach you how to play against Pox—as helpful as that is. Someday I’ll spend time explaining why that deck is so bad, but people have done that time and again for years and it’s made no impact, so maybe not.

Without a rules engine like MTGO, people are able to do two things very well. First, people can shortcut very easily, something that MTGO doesn’t allow for. Second, to the detriment of all these programs, not having a rules engine means that people are incredibly skilled at accidentally misunderstanding or sometimes intentionally misrepresenting the rules. Sadly, the normal reaction when someone thinks the rules of the game work differently than you do is to call you a “n00b,” tell you to learn the rules, and rage-quit. It makes me wonder exactly what kind of people are on the other side of the screen, but when I think about the kind of people I know who would react like that, I try not to think about it anymore.

It is worth stating that because there is no rules engine, new cards don’t need to be implemented in the painstaking fashion in which they are with MTGO. This makes it not only possible to play with the new cards before you would be able to on MTGO, which will be after the Grand Prix, but within minutes of a specific card being spoiled, let alone the entire set. This is one of the major advantages of these kinds of programs, because as long as the card exists, you’re going to be able to get a feel for it before you commit any resources to acquiring the card.

In summary, essentially, the problem with all of these programs is the people—but you really can’t have Magic without them, can you? Not really, but it’s an interesting pastime, which is something that I’d like to address now. It’s actually the conclusion of all of my experience with Workstation: It’s a pastime; it seldom passes for testing because you’ll almost never play against competent people unless you actively go out of your way to organize something with someone—in which case, the program is great because it allows you to supplement testing sessions from a distance.

The Deck

For a long time, I’ve been a fan of Angel Stompy, but there has always been a fundamental problem with the creatures. In Angel Stompy’s inception and early stages, you had to use creatures with {W}{W} in their mana costs—such as Silver Knight and Soltari Priest—because these were the best available creatures, not to mention that Silver Knight would often give wins against about half the field. However, this clashed with the fact that you’re trying to use Ancient Tomb to speed out cards like Parallax Wave and Exalted Angel. What the deck wanted for a very long time was aggressive creatures at the {2}{W} mark so you could easily play them off Ancient Tomb or City of Traitors and a Plains. New Phyrexia was exciting for me because it didn’t just provide one of these creatures, but two, and this was really going to enable the archetype to change gears from where it had traditionally been. Let’s take a look at the list; I call it Porcelain Stompy.

To download a Magic Workstation file for this deck list, click here.

Originally, the deck was named Angel Stompy, taking its name from the powerhouse of Exalted Angel. Since I play no angels in the main of this deck, I wanted a name that still said this was a White deck, but identified that it played a lot more artifacts than other decks in this vein. Porcelain Legionnaire encapsulates that, being an aggressive White artifact that can be cast off an Ancient Tomb. Remember, kids, naming your deck is the single most important thing you can do, so make sure it counts and is catchy so that other people are obligated to use the name when discussing the deck.

I’ll go through a quick rundown of the cards that aren’t obvious.

All of that equipment—I love playing with toolboxes, and Stoneforge Mystic really enables me to do that with these cards. I always wanted to have the right tool for the job, and the breakdown of these five pieces of equipment gave me what I feel is the best assortment to choose from. If you’re looking to play a deck somewhat along these lines, I want to strongly encourage you to stay away from Sword of Body and Mind. The card’s ability is pretty good; getting a Wolf can be among the biggest upsides a Sword has, but the drawback of milling your opponent for ten is huge. There are far too many decks in Legacy that use the graveyard as a resource, and you really don’t want to risk playing a card that enables or assists those plans of attack, especially considering how easily some of those can blend in with more mainstream decks.

Aven MindcensorThis deck really wanted an evasion creature to help get equipment hits through, and Aven Mindcensor is among my favorites. The disruption it presents is highly relevant against most combo decks and is highly effective against things like Natural Order and Green Sun's Zenith. Of course, it’s a nice bonus that bird is the word, and this bird also has a chance of taking care of fetch lands.

BatterskullThis has actually been the card I get with Stoneforge Mystic the most, followed closely by Jitte. The ability for Stoneforge Mystic to get a huge creature onto the battlefield on turn three is a big deal, and many decks can’t handle the 4/4, let alone equipping it to something like a Lodestone Golem.

Blade SplicerThis card is so great for me because there is finally an aggressive creature that costs {2}{W} and it even is a little better than a 4/4 as far as targeted removal is concerned. There were several games in which my opponents were overwhelmed by Sword of Light and Shadow returning this guy. Additionally, because most people on MWS are functionally illiterate, people seldom saw that the Golem had First Strike—something I expect to carry over into the real world.

Lodestone GolemJust before New Phyrexia came out, I’d wanted to fit this guy into this type of deck, but the artifact count was way too low to justify it. However, I’ve really enjoyed him in this shell, and there is an unintentional synergy between Blade Splicer and Lodestone Golem that is easy for the opponent to overlook.

Mirran CrusaderThis was added late into the first builds of the deck. I was initially really skeptical of the {W}{W} in the mana cost, as I’ve identified that as a big weakness of the older builds of this sort of deck. The power of this card is incredibly high when it’s standing alone, and doubly so when it gets to wield a piece of equipment. The fact that B/G/x decks are so popular right now, and that in the short term some people will be moving away from Swords to Plowshares for things like Go for the Throat is really attractive.

Mox DiamondWhere other decks in this vein use Chrome Mox, the artifact count in this deck is rather high, and even if the artifact is colored, you can’t imprint it on Chrome Mox. I was really forced to toy around a lot with the number of Mox Diamonds I was playing. It may be that playing something like the Junk package of three Diamonds and twenty-four lands is where this wants to be, but it’s something I’m still having trouble coming to terms with.

Porcelain LegionnaireNot just here for the naming convention, Legionnaire is rather aggressive as far as White cards go, and is relevant in combat for a remarkably long time. He holds equipment incredibly well and does defense pretty well against any tribal deck. The one toughness is particularly saddening because it opens him up to Grim Lavamancer, but I’ve been pretty happy with the card overall.

This deck, with most Chalice/aggro decks, really isn’t designed to be taken into a blind tournament and meet great expectations. It’s a deck you take when you have a feel that the metagame is ripe for it. In the case of this deck, I was pretty sure I was going to do quite well against most of the B/G/x decks such as Team America and The Rock, against Bant aggro and to some degree NO Bant, and against most control decks in the format. Furthermore, I expected great things against Merfolk, as I’m playing a lot of guys who hit hard and fast. Combo was up in the air, but I wanted to get a feel for what the deck was really going to be performing like.

The Results

You can find the raw data notes from my test results here. If you’re a true soldier and read through it, you’ll get to see me make some commentary on a few specific decks (note, only forty-six games are recorded there for some reason).

I entered this session of testing with some specific rules that I tried to follow. First, I would never SPL (stands for System: Player Lost, which is when a player quits from a game) from a game; not only is this poor etiquette, but I wanted to give every deck I played against a fair chance. Second, I would host all of my own games with a game description that clearly communicated my desires, and I would only join a game when someone had hosted one that closely reflected what I wanted. Third, I would keep accurate results of my experiences with the games and program during the course of the testing. Last, I would never insult, criticize, or say anything that would influence the plays that my opponent would make. Trust me, there is a lot of flaming on MWS, and I am not really one to take part of it. I didn’t want to say anything to upset anyone or make any comments that were going to impact the plays my opponents would make, unless they were asking me a rules question. I held to all of these rules, even the last one, despite having my patience put to the test, and didn’t have much to show for it.

I ended the tests with the deck at thirty-five wins and fifteen losses in matches, which isn’t bad as far as a metagame deck is concerned. I’m a lot more interested at looking at the decks I played against over this time rather than really examining how the deck did. That is to be saved for real testing against opponents of known competency. You may scoff at the idea of only playing fifty matches, as even last week, I said that you should probably be testing a single match in ten pre-board games and twenty to thirty (possibly even more) post-board games. However, when you’re doing this against fifty different opponents, this time really adds up. It took the better part of my free time over two days to get this in; a lot of it was spent waiting for someone to join, as most people weren’t interested in playing a competitive game. People incessantly telling me how lucky I am got tedious almost immediately, but the feeling of crushing people who are prone to complain so much remained savory. I’m positive that if many of you ran this same experiment, you’d come up with at best very similar results. What I’d really like to do is to show you the breakdown of what I tested against.

I decided to go with a pie chart to show you just how much of a mess the experience was—fifty matches and thirty-six different decks. Most of the competitive archetypes I played against are labeled. Had I been playing a combo deck and tried to do this testing, it would have been much more difficult to get an accurate read on all of the decks that I was playing against, so playing a more fair deck worked out in that regard. For nearly every game, I created a game with the description “Competitive Legacy with NPH Phases 2/3SB,” which is a bit longer than the window will actually display, but it gets the message across. I didn’t ask for tier decks because that generally doesn’t work well, but also because I wasn’t playing an established deck; I just wanted to play against decks that the pilot thought competent enough to win.

In asking for people to play competitively with New Phyrexia, I skewed these results in two ways. First, anyone I was going to play against would likely need to know how to import sets and have an idea of the ramifications of the new set on Legacy—meaning that I was skewing it a bit toward competitively minded people who wanted to test out a tier deck or a new brew. Second, I created a barrier where I clearly defined that I was not interested in playing against casual decks and wanted to play a full match to test against. Ideally, this would remove any of the players who disconnect when losing, or the Dredge players who like to get the rush of winning Game 1 and disconnecting.

To condense the above results into something a bit easier to read, I had this experience:

In setting the bar as high as reasonably possible, I played against nearly 60% real decks, and about 40% decks that likely were not competitive or at least not established, and thus granted little in the way of experience. If you’re only looking at the decks I played against, this means that for testing purposes, assuming all opponents were competent pilots, I wasted a full third of my time trying to play against irrelevant decks. Of course, these numbers don’t take into account player skill, which is sometimes difficult to gauge, but when people play spells directly into your Chalice of the Void or miss a lethal attack, you can pretty much get the gist of where they stand. If I had to estimate how many players out of the fifty I played were actually competent, it would probably be around ten, and of those ten, probably four were playing nonestablished decks. As I’ve said, this is after making the wasters as hospitable as possible to play against good players with good decks who have a workable understanding of the rules.

Had I done this test asking for just competitive Legacy decks (once again, trust me—asking for tier decks means nothing), the results of this experiment would have been much worse. In asking for anyone to play, I would have opened myself up to playing against an endless stream of Pox decks, which I was actually very fortunate not to play against, and some of the most subpar builds of burn you can imagine. You’re always going to run into people with no experience playing decks they have no concept of how to pilot correctly, so that remains the same regardless. I’d estimate that when you factor the competency issue in with the fact that people tend to play non-tier decks, my time spent playing meaningful games where I could have learned about my deck against the metagame was probably something a lot closer to a tenth of those matches. Five matches over the better part of two days isn’t stellar.


I feel that Workstation can be a great tool for testing with known partners, or even getting your bearings straight with a new deck. It’s especially good if you’re just looking to spend some time. However, it’s really important to understand the point at which you stop spending time in a valuable sense and start wasting it. The broad landscape of Workstation decks and players can only improve your ability to a degree, and coming from someone who’s used Apprentice since 1998, the learning curve you experience with blind opponents comes to a plateau that stretches out into the horizon rather quickly. I hope you don’t get lost there.

The Bazaar of Moxen V results are up, but sadly, there is no word just yet on when the lists are going to surface; I was hoping to talk a bit about them in closing today. While I spent Sunday waiting in hope of the BoM V lists being posted, I watched SCGFL’s coverage, and someone posted in the Twitter feed that the event really showcased the strength of a two-round bye in an eight-round event. The difference between needing to go 6–1–1 to Top 8 and 4–1–1 is pretty significant; it’s something I’ve touched upon before, but with so many eyes looking to the SCG series as an indicator of the state of the metagame, it’s disconcerting. At the Bazaar of Moxen, there were over 600 players and nine rounds of Swiss with a cut to the Top 16, which demands a record that’s slightly stronger than 4–1–1 to make. European tournaments are normally shunned by Americans for having such alien shifts in the metagame, because decks like Storm and Aggro-Loam tend to be very popular over there; notably, neither of those archetypes made Top 16. Regardless of the metagame breakdown, I’m quite interested to see what the result is, since there are so many people on both sides of the Atlantic who are wondering what to do now. Back stateside, I’m not surprised to see so many copies of Mental Misstep in the opening weekend. I don’t think anyone was, but with the smallest turnout of a Legacy Open yet this year, I’m left wondering what the Florida Legacy scene is really like. I think it’ll be prudent to wait a bit longer before we form too many conclusions on the state of the new metagame; until then, I’m going to keep checking French message boards hoping to find a link of the Top 16.

~ Christopher Walton in the real world
im00pi at gmail dot com
Master Shake on The Source
@EmperorTopDeck on Twitter