by Cameron Veigel
In this scenario, the beast shall be played by ‘Testing’, and the hero by myself. Roll film.
“I shall obliterate your matchup percentages!” yelled the beast.
Not so fast,” our charming and attractive hero replies, “you have forgotten something.”
“What is that?” The beast asks curiously, its appearance reminding the hero of why many players avoided seeking it out before an event.
Calmly, whilst putting his Elder Dragon Highlander deck back in its giant case, the hero replied, “I am aware of your weakness. And I know how to use you to my advantage.”
CUT. Aside from the plot of this so called film being rather far fetched – surely the hero doesn’t really know how the beast ticks – the hero will invariably die in the ensuing conflict that takes place in Yokohama, Japan, due to believing things that aren’t true. Presumably he will end up with 2-4 inch cuts over his dreams, in shades of Black/Red and a slight splash of Blue. If you catch my drift.
My point is thus: I have not determined the be all and end all of testing methods, but I think I may have done some things right in the past. With a similar event to test for coming up in the future, albeit with a capital Q tacked on the end, it seems relevant for me to throw some of my ideas out there and see if I can be of any assistance. I did well to get to the PT. Surely I did something good once, right?
I’m not going to talk about the PTQ win, as I believe the process for that came down to two things; choosing a deck that would beat 60% or more of the expected field, and knowing every inch of my deck backwards and forwards. I could play that tournament 50 times and I wouldn’t change a card in my list. What I want to talk about is the prelusive process for Yokohama. I’ve spoken about it in brief before, but that was slanted more toward my performance at the PT and retrospectively what I did, rather than a helpful analysis for aspiring players.
When it all began, I didn’t really have a clue. I have never really been one to invent ‘best decks’, though I am capable of tuning them. I am comfortable designing rogue decks to take ‘risks’ on matchup percentages, but on a high level circuit, that’s not always the correct gambit. Certain Owling Mine players may, however, disagree. I was relying a lot on the other strong players in my area to help me by giving up their time and ideas for my free trip. Dave Crewe was by far my biggest asset in this time. Aside from the fact that he put together a bunch of decks with proxies, he had some degree of connection with pros to feed him ideas and pointed me to articles or relevant MODO tourneys from which I could potentially learn. I owe him a great deal for not only how he helped me in this process but the lessons I learnt about the process itself to apply to future testing regimes.
The first step in testing a new format is to identify the most played decks. You may not necessarily want to play them (in fact, I almost never do initially), but regardless of your decision, a large percentage of people are going to play the strongest decks. It makes sense. They are the strongest decks. If you have thus not tested your deck against and developed ways to beat these decks, than in my opinion, you won’t perform well. In this example, there were two decks locked in battle at the top; UB teachings and White Weenie. I felt it would be better to play UB Teachings if I were to touch one of these, but decided that playing UB mirrors many times at the PT against people who were at least as good as I was would be a bad idea. It would put a lot of stress on continually playing well and making perfect decisions or losing. That’s just the animal of control on control. White Weenie, on the other hand, seemed like a bad choice itself for my style of play. I like being able to react (and I’m not in any way talking about Mana Tithe here), and would feel better with a deck that wasn’t just crossing it’s fingers and trying to play fair by attacking. Also, White Weenie had a high expectation of being played, as it was incredibly easy to build. Insert red t-shirt here.
Dave and I had already been testing with a mono-black Gauntlet deck that he had designed. Obviously, this deck crushed White Weenie, but was unfavourable against Teachings. Therefore, if we could tune it to have some game there, it would be a valid choice. After these two initial threats were identified, mono-black control started coming out in the top 8s of premier events. It seemed the winners were on rotation, UB to beat the last MBC decks, WW to beat those, then MBC to beat those again. This created a third ‘top end’ deck to beat. Admittedly, the MBC deck was different to ours, and was more likely better, but ours had Null Profusion and Gauntlets of Power and cast a billion spells a turn. So we kept testing with it instead, hoping it would perform better.
Next we had to create copies of some of the lesser decks, to which end I will explain in a moment. Slivers, both aggressive and Wild Pair combo, were something I messed around with for a while, as well as various forms of mono red agro. The problem I hear with a lot of people’s testing stems from this stage. Person A tests his UR counterburn deck against Person B’s modified UB Faeries build. Ok, A is getting an arguably legitimate testing session, Faeries is dominant. But what is B getting out of it? What does testing against an unknown deck gain you in prep for a tournament? Aside from learning how to play your deck a little better, maybe, it doesn’t really help you at all. This is fine. Testing can’t be about everyone at once, but you have to be aware of who is getting relevant testing.
On the other side, A thinks he is preparing against Faeries, but he really isn’t. Clearly, a lot of people are going to play different cards in their lists, due to availability, personal preference, whatever. The key here is to build the most standard version (therefore sharing the most cards in common with everyone else as possible) and play against that. Too often people are just playing the decks they have been playing for months against one another and claim it as testing. You need to develop a Gauntlet. That’s a list of decks like I was talking about, both ‘tier 1’ and ‘tier 2’ is preferable, and run your deck through it. If you haven’t decided on your deck already then you should be running all the decks against everything else in a round robin fashion. For both Worlds and Yokohama I developed a matrix on which I could determine at a glance what matchup percentages should be like across the board for a deck, which could aid me in my decision of the ‘correct choice’.
The other problem people seem to have with testing is showing up to every ‘testing’ session with their Faeries build, and just playing with that. For a start, and this is just my opinion, I don’t think you should choose a deck before you start testing. If you want to win an event, you can’t be swayed by your own emotions. This was the problem with our MBC deck. It was really fun to play and we stuck with it for way too long before we decided it couldn’t compete, making excuses for why it had lost and boosting its confidence. If you pick your deck then start testing, don’t be surprised or angry if you don’t win. To be honest, if you’ve been playing that deck for as long as everyone says you have, why did you even bother testing? You didn’t learn anything new, and you stubbornly didn’t change any cards as you went along. Of course, in any format, there will be decks you like and decks you don’t, so you will want to start with the decks you like and test from there. This is alright, and is the same way I will begin. You may as well get into this methodical process having a little fun and maybe finding a deck that you like rather than making it a chore.
The lesser decks all seemed just that in this format, and Teachings was the clear strongest deck. It was then we finally admitted to ourselves we had to update our MBC deck to stand a chance against UB. We were already playing disintegrate as a finisher, and added more red cards in the form of Void and Hellkite. I had made a list of the ten strongest cards in this format, and this deck had five of them. We reran it through the gauntlet and worked some stuff out. Never assume your changes will work right – test them. It thrashed all the agro decks, as it was almost running 7 wraths then a few wraths that flew as well. The chances were better against teachings with a card like Hellkite to pick an end of turn fight with them, but still not dominating.
On our last day of testing, Dave rocked up and said ‘hey I saw a version of this deck online splashing blue for Aeon Chronicler’. We gave it a try. It was sick and also improved the teachings game, being able to keep up to their answers with our threats. We made the sideboard then, on the basis that agro matchups were already awesome so only needed a few cards, and the rest went to Teachings. This was an error. Making a sideboard at the last minute is really a dumb choice. Aside from the fact that at least half your games will use your extra fifteen cards, you still need to test with and against sideboards to make sure the cards you have put in are fulfilling their roles. This could obviously seem a painstaking process, but at this point you should have limited your options to one or two decks instead of a huge range, and testing only those is much faster.
Two rounds into the PT the guy next to me cast Mystical Teachings and suddenly it clicked, ‘That card should be in my sideboard.’ Would’ve meant I could have just played one copy of Extirpate and Haunting Hymn, and would’ve been good with my Hellkites as well as Sudden Deaths for opposing Teferis. Oops. If only I a) hadn’t been stuck on the initial version of this deck for so long and ‘wasted’ so much time and b) had more thoroughly looked at sideboarding options. It could have been a much different story. At the end of the day, Frank Karsten used a deck with a lot of similarities to mine, so turns out we didn’t end up doing too badly. But only because we had a method and I was eventually willing to change.
There is no defining way to test, but there are certainly wrong ways. My method is just this:
1. Determine the main threat decks – Don’t lose to these more than you beat them.
2. Determine the other decks – Don’t lose to too many of these decks.
3. Build your Gauntlet and run things through it. Usually we do 8-10 game sets initially, alternating play and draw.
4. Develop a list of decks with the best percentages and test these on a greater scale, refining your list to decks you might play.
5. Decide on a deck(s) and build a sideboard to shore up your weaknesses. Having more than one option at first is ok.
6. Test your sideboard and develop a SIDEBOARDING STRATEGY. Know what your worst cards are in each of the important matchups before you register your deck. Stop taking out one copy of four different cards for your playset of hate. You are only making your deck less consistent.
7. REPEAT THESE LAST TWO STEPS AS OFTEN AS NECESSARY UNTIL YOUR DECK IS WINNING MORE THAN IT IS LOSING. Don’t put in all this effort just to say ‘oh I’ll just play really well and earn some wins there’. If you were that sold on your deck to begin with you’ve just wasted everyone’s time and have probably learnt nothing.
8. Win a PTQ.
And DON’T…
1. Choose a deck before you test.
2. Assume your changes are correct without testing them.
3. Have zero copies of decks to beat in any duel.
4. Forget to record your results.
5. Expect to do well if you haven’t put in the time beforehand. That’s just how it works.
Hopefully some of this is useful to you with block season looming. It’s a lot of words, I know, but there are a lot of relevant points to put forth. I could’ve just directed you to the end for a quick overview, but that wouldn’t have been as enjoyable, would it?
Until next time,
Cameron Veigel.
