Building A Better Gauntlet

Valeriy tells you how to build a better testing gauntlet in the wake of changes to Magic Online & shares his current ones just in time for #SCGPROV & #SCGVB.

Magic Online was probably the most important Magic-related topic of the last week. Many words have been said about it, and I’m not going to add more. Instead, I’ll focus on an indirect consequence of the decision to take down large events: decklists. Everyone likes decklists, and Magic Online was one of the major sources of them. Larger tournaments tend to set the direction of metagame evolution, while Magic Online Daily Events were there to prove ideas and set up an equilibrium for next week. Now that Daily Events are gone, the evolution of decks will change. Are you ready for it?

Everyone knows the concept of a local metagame, especially if we’re speaking about a FNM-level tournament where most players have one deck for a long period of time and don’t change often. Everyone knows that the metagame of the StarCityGames.com Invitational has many more blue decks than any other comparable tournament. And by this point, everyone knows that deck price is a very significant factor in Magic Online deck selection since cheap decks always overpopulated Daily Events. These are just a few things that influence metagames and deck selection.

Let’s look at an example from the current Standard format. There are many viable decks, with a complicated equilibrium among them. This equilibrium can be set in very different points, which makes playing the same 75 cards a very good choice for one week and very bad for the next one. The cheapest good deck is Mono-Red Aggro (note that there are many versions of it but they all line up quite similarly in price), which is significantly overrepresented on Magic Online; even if it’s mostly chosen by newcomers, you must be well prepared for it and should probably avoid decks that have a bad matchup against it.

Esper Control, for example, is a very good deck ([author name="Reid Duke"]Reid Duke[/author] and Huey Jensen showed that last week in their Magic Online battles), but it’s specifically bad against Mono-Red Aggro.  Therefore, you’re forced to either ignore the deck entirely or have a strong sideboard against it, which lowers your chances against the rest of the field. This lose-lose situation forces players to skip over Esper and opens the door for Mono-Blue Devotion, which occupied nearly a third of 3-1 and 4-0 decklists in Daily Events.

The next step is the popularity of Mono-Blue Devotion, making it more popular at offline tournaments. The deck is powerful, well-tuned, and has proven successful for many good pilots, so you can expect to face many Mono-Blue decks at the SCG Standard Open in Providence. The difference between Magic Online and real life is that Esper Control will not have to worry about Mono-Red Aggro as much, so I expect to see Esper at the top of standings since it has a favorable matchup against Mono-Blue.

However, my article is not about this situation but rather about changes in it.

Wizards’ announcement about canceling large events didn’t include a word about decklists, so many people expect them to go away as well. Do you remember the buzz created when it was announced that instead of posting all of the Daily Event results they were only going to post one per day?  As it turns out, this decision was fine because 70 decklists aren’t worse than 300 if you want to check what happens and still leaves some room for innovation. The complete lack of decklists would’ve changed the metagame drastically both on Magic Online and in real life. Do you remember how people played Slaughter Games before Pro Tour Dragon’s Maze just to screenshot the opponent’s deck? The lack of online decklists would increase the value of SCG Opens as a weekly source of technology.

Some people have discussed the possibility of posting winning lists from eight-man queues (which is very reasonable), but Wizards made what I feel is the strangest decision possible. I was unable to find the announcement (which says a lot about their public relations and the quality of their website—but I digress), but it seems that they now will choose two eight-man queues at random and post four decklists from each one, i.e. one 3-0 deck, one 2-1 deck, and two 1-1 decks. So we’ll get eight decklists per day, but four of them will be ones 1-1 records. This just has to be worse for the long term.

Okay, let’s stop whining and quickly summarize what we have right now.

  • Magic Online decklists are very random, statistically insignificant, and aren’t better than random decklists from forums.
  • The value of decklists from well-known writers has dramatically increased.
  • The value of decklists from the StarCityGames.com Open Series just skyrocketed.

These changes mean that you should tweak your principles of choosing decklists for your testing gauntlet. To help, I will also voice some basics of building a gauntlet. I think that it’s useful to go back to basics in this case. It’s also important to note that I’m speaking about gauntlets for fairly established formats; building a gauntlet for a tournament like a Pro Tour is completely different.

The first question I get asked a lot is “why should I have a testing gauntlet if I can simply go on Magic Online and test against the field?” This is very valid point and I agree that online testing is very important, but it can’t fully replace other ways of testing. As stated earlier, Magic Online is infested with cheap decks that as a result will significantly affect your testing experience. Another thing to remember is that it’s not that easy to have the same amount of decks in real life as you have online (you can’t make proxies online). And if you’re testing for a real-life tournament, it’s probably smarter to spend your money on paper cards instead of digital ones.

Magic Online is very useful to grind for long intervals and improve your playskill regardless of deck and/or format, but if you’re going to test out a matchup at length, dedicated testing with teammates is a lot more useful. Personally, I’m a fan of group testing, as it allows in-depth analysis and the reconstruction of certain game states. It’s highly likely that the people around you catch something you missed during a game, which leads to everyone learning together.

Let’s go to the real principles of building a gauntlet. The first one is to have a gauntlet that takes your free time into account. If you have a week to spend for preparation, that’s one story, but if you have a full-time job and only two weeks to prepare, that’s a completely different one entirely. In the first case, you can try nearly any deck and then test the best ones against a wide opposition. In the second case, you’re forced to make preliminary decisions without testing by looking at decklists and reading articles.

My solution for this problem is to have two testing gauntlets: a small one consisting of a very limited number of the best decks and a larger one with a much wider variety. The point is to estimate the potential of a certain brew in a very limited amount of time and save the rest for more precise testing. A small gauntlet should consist of very popular decks, and having a bad matchup against each of those is an easy reason to reject a deck. However, you should remember that there usually isn’t a deck that has a good matchup against everything, so you should analyze your experience appropriately.

My small gauntlet for our current Standard format is four decks: Mono-Blue Devotion, Mono-Black Devotion, Mono-Red Aggro, and Esper Control. There are more good decks, but these four are especially popular and also maintain another important condition—all basic strategies are represented (save for combo, which simply doesn’t have a representative in Standard).

For example, if Mono-Black Devotion didn’t exist, I’d still put some midrange decks into the small gauntlet simply to be sure that I’m not overlooking something. This is by the way a reason to include a green deck in a small gauntlet; none of them are particularly popular, but many of them exist and they have a fair share of the metagame if combined together. Would you include one of them in your small gauntlet? If you so, which one? Kibler’s G/B Aggro? Mihara’s G/R Devotion? G/R Monsters? G/W Aggro?

I don’t include decks like W/R Aggro and U/W Control in my small gauntlet since even if they’re good the gauntlet already contains similar strategies; it’s not very useful to have two kinds of fast aggro. However, when you’ve chosen your deck or maybe narrowed down to a pair of them, it’s good to determine deck-dependent features like W/R having Boros Charm to protect them from Supreme Verdict or G/B strictly outclassing G/R in terms of removal for Master of Waves. The size of a large gauntlet strictly depends on the amount of time you have (and sometimes it’s even better to skip it and simply play online), but when building it I generally try to include second representatives of each basic strategy and cover specific strategies like G/R Devotion with its combo-like outs.

The second very important principle is that you’re testing against archetypes, not against certain decklists. Aside from situations where your opponent plays the exact 75 from last week’s tournament (which is almost always a mistake), you never know your opponent’s exact decklist and can only guess what they have. A good example is a situation where you play some kind of combo with various ways to be hated out and should consider the popularity of them and the ability of a certain deck to play various kinds of hate.

I used to have two decklists for each archetype: one from last week’s large tournament (or from a recent article of a respected author) and another one randomly picked from Magic Online and having some significant differences with the first one. Now this in not the case anymore, so I’m probably going to build some kind of generic “other” decklist with a reasonable variety of maindeck and sideboard singletons to simulate all the cards I could face.

It’s also worth noting that if there is a principal difference between two versions of a deck it’s better to have two decklists in your gauntlet. The different versions of Red Devotion, one featuring Domri Rade and one with Assemble the Legion, are a good example. These decks are similar but use different splashes to solve similar problems, and you must consider both Mistcutter Hydra and Chained to the Rocks when testing your Mono-Blue Devotion deck against Red Devotion.

There are many ways to simulate small changes (say, Vaporkin and Omenspeaker in Mono-Blue Devotion). You may have two different decklists (which is questionable in general but may be the case if you compare Frostburn Weird and Rakdos Shred-Freak in Mono-Red Devotion); you may have a 2/2 split for the slot under investigation, or you may have a sort of split card turning into what you want right now (and locking upcoming copies in this state or not). Each method can be good—the exact choice probably depends on the deck—but keep in mind that this is a fine way of making your testing gauntlet more representative.

The third principle of building a gauntlet is a direct consequence of a sentence in every article dedicated to testing. These articles always say “play more post-board games.” I say “always have a good and wide sideboard for your gauntlet decks.” There is nothing strange in the existence of different post-board strategies for certain matchups. The exact one may depend on how many slots you can dedicate for a matchup, or on personal preferences, or on the interchangeability and versatility of cards. I normally keep 75s from large tournaments as is, but it’s totally fine to have a 25-card sideboard featuring a ton of different strategies.

You can put as many cards against a certain matchup as you like, but it’s important to understand that you can’t do the same at a real tournament. Clearly determine how many cards you want to side in and then choose them among your options or adopt some known sideboarding strategy without touching cards from another one. The first approach is also helpful when you’re trying to put your own fifteen-card sideboard together—you state how many cards you can use and then accurately choose them from all the available options.

Say the extended sideboard of Mono-Blue Devotion contains four copies of Gainsay; three copies of Domesticate; a copy of Cyclonic Rift; a copy of Nykthos, Shine to Nyx, etc. You simply can’t put them all into your post-board deck against the mirror match, so you should choose what you’re going to do first. For example, if you sit on four Gainsays, it may be fine to leave Nykthos, Shine to Nyx (and not put it into your actual sideboard), but if you’re going to increase your Cyclonic Rift count, Nykthos is vital.

Testing against different sideboarding strategies and asking your playtesting partner to choose one of them at random for each game to simulate the uncertainty you experience during a real tournament are both fine strategies.

The fourth and last major principle is never use your actual deck as part of a test gauntlet. There are some exceptions (like if you’re testing for the Pro Tour and you’re a part of Team StarCityGames; testing the mirror of your actual list may be a very good idea in this case), but you should avoid it in most cases. This principle is part of a previous one, but I wanted to emphasize it specifically since many players feel that it’s fine to test against teammate’s list as a representative of a deck. This situation is in fact a worse trap than testing against last week’s winning list. Moreover, it’s very helpful to test a stock list of a deck you’re going to play to notice what opponents expect from you and how can you stymie their hopes.

It’s good to have each teammate responsible for some gauntlet decks (piloting them in testing, updating lists, etc.), but even if a person responsible for a certain deck plans playing the same archetype at the tournament, it’s vital to keep gauntlet lists and tournament lists separate. Overcommitting to specific deckbuilding decisions may significantly worsen your decks against the rest of the field. By the way, I recommend having someone else also playing your archetype as a gauntlet deck just to have more opinions about the deck and cards.

Another related point concerned with playing a tier 2 deck (not in terms of quality but in terms of popularity): don’t let your teammates overestimate significance of your deck in the field. Yes, you will be very positive about your deck and your chances and some your teammates may be too, but it doesn’t mean that the rest of the world understands it even if you actually broke the format.

Breaking the format may become a more frequent thing without Magic Online decklists, so just remember the good old times and improve your gauntlet and testing process for upcoming tournaments like the SCG Standard Open in Providence. The last thing I’m going to talk about before leaving is my current testing gauntlet.

Small Gauntlet

Mono-Blue Devotion: Sam Black list from Grand Prix Louisville (yes, the old one) and a generic list from Magic Online with a 30-card sideboard.

Mono-Red Aggro: Two lists I built for myself, a very fast one and another one with eight one-mana creatures and a playset of Fanatic of Mogis.

Mono-Black Devotion: Brian Braun-Duin GP-winning list and a Magic Online list with four Pack Rats maindeck.

Esper Control: The list from Reid Duke last video and a random list from Magic Online with nearly 40 cards in the sideboard.

Large Gauntlet

Green Aggro: Brian Kibler G/B Aggro, a generic G/W Aggro list, and G/R Monsters from the Top 8 of GP Santiago.

Red Devotion: Ben Stark’s R/G version and an amalgam of various R/W lists (as I consider the deck as my own choice, I gathered a lot of data).

Green Devotion: A generic G/R list from Magic Online and a G/U variant splashing for Prime Speaker Zegana.

White Aggro: Ben Lundquist Open-winning W/R list and a W/B list with Thoughtseize and Orzhov Charm (I built this list for myself and consider it a variant).

Junk Midrange: I don’t consider this a good deck, but Blood Baron of Vizkopa should be respected in a green shell as well as the Whip of Erebos plus Obzedat, Ghost Council combo.

Versions of Tier 1 Decks: Blue Devotion splashing black for Thoughtseize, Black Devotion splashing green for Abrupt Decay and Gaze of Granite, B/R Aggro from GP Santiago (which I consider a Mono-Red Aggro variant), and a Red Burn deck splashing white for Boros Charm and Chained to the Rocks.

The large gauntlet is very large, which indicates how healthy and interesting the current Standard format is. Don’t waste your chance to prepare for and play it at SCG Standard Open: Providence or anywhere else.

Valeriy Shunkov