Due to my extensive testing on Magic Online for the Australian Nationals, I have a decent topic to discuss that has been bugging me from the way people test online.
As some of you may be aware, there are several ways to play constructed when you start out with a new deck, be it one of your own creation or something based on a deck already in the format. Online, there are the casual rooms and the constructed 8 mans as well as Premier/Daily Events. Each room has its merits and fills a different role in the beginning, middle and end phases of deckbuilding/testing.
This article isn’t going to be for the people who have access to the cards/program online, it’s just that’s where I notice all the flaws I am discussing (I live in the middle of nowhere remember? MTGO is my escape) and I’m using Magic Online as some medium for explanation.
Prelude:
First, you should really identify why you are testing. Is it for Nationals or an FNM? The answer to that basic question should give you an indication of both how in depth your testing process should be and how seriously you should take it. FNM is a testing ground itself, so you only have to do a little bit of work before attending, just to make sure your deck doesn’t completely fall on its face. If you bomb out, who cares? If you bomb out at nationals or a PTQ however, you are out a plane ticket (one way or another…).
The Beginning:
Ok so you have your deck idea, you’ve assembled some of the cards and you are keen to crush. What now?
Build a Base
Put all the cards you want to be playing in a list and add some lands. I’m not going to get technical with curves or ratios because that obviously depends on the deck in question, and is part of the process itself. There are a couple of things to consider though.
Firstly, in certain cases you might not be able to play all the 3 drops you want. Take here for example Knight of the Reliquary and Dauntless Escort. It’s possible you thought both these guys could find a home, but you only have four slots. What to do! There’s an obvious choice, which is just to play one of them (whichever suits the deck’s strategy best), but also a possibility is to play two of each and test them out in each matchup that way. It’s a lot easier to test a card when it’s in your deck. Each of these cards in going to be better in different matchups – one of them is better against aggressive strategies, one better against control – but you’ll find out a lot about how each card plays with your deck and the interesting things they can do in the opposite matchups, say, for example, are your escorts being particularly helpful against the Earthquakes of Red Deck Wins?
Make sure you are playing enough land. It’s better to start high and work down, because you will at least then be interacting with your opponents. It’s also the best way to not get skewed results. If you begin with two few lands and keep ‘getting lucky’ (without realising), you will think your deck is a lot better than it is actually going to be at higher levels. Yes, obviously the reverse can be said to be true, but if I’m going to sabotage myself in an event, I’d rather sabotage myself with an ability to cast spells. Note that I’m not suggesting you should over estimate here with your land counts, but just make sure you are playing enough. If you need to cut something to get to 60 cards, a land is probably not the right choice.
Make a random sideboard. Imagine some weaknesses your opponents are going to exploit (your mana, ability to run out of cards fast, colours of your creatures, etc) and defend yourself against them. Unless you are some kind of deckbuilding genius, your original sideboard is likely not to look anything like the one you end up playing. Matchups that you thought were bad you might realise are either closer to even or you simply don’t need that many cards. Matchups you thought you were ahead in might get far worse after boards and you need something to regain the edge. Whatever the case, accept that you are probably going to have to change it and don’t be afraid to try something crazy. Maybe this craziness is putting Mark of Mutiny and Eldrazi Conscription in you red deck sideboard. Maybe you’ll beat a name pro with such tech.
Play with it
Run your deck against everything. Don’t just play against the decks you expect to/built to beat. Unless you are trying to run some kind of Owling Mine swiss gambit, it’s still important to know how to play your bad matchups. It also helps a lot with your sideboarding choices going into a big event, because theoretically your maindeck is fine enough against some proportion of the field, and you really just need to make the worse matchups better.
Online, the Tournament practice room in Casual is ok. You will still play against idiots sporting Marrow Chompers and Hellkite Hatchlings. It can’t be avoided, but you should still play out the matches. Don’t concede to them. Play against their jank in between ‘serious’ games. Random decks crop up all the time in early PTQ rounds and regionals, so it’s important to know how your deck fares against janky homebrews as well as the top of the meta. For example, at GP Oakland my Zoo deck was well positioned against the decks I expected to see doing well day one… but it couldn’t beat GRW scapeshift (a poor choice and a worse build), restore balance (tier 3 deck), BW lifegaining dudes.dec (what?) and I played all these guys early and bombed out. This isn’t to say that you need to refine your decks to beat these guys, but when you get thrown some wacky curve balls, you better be willing to refocus and consider your deck’s best possible chance of beating them through all the bizarre cards they could have in their grip.
You see a lot of matches advertised in the room as Standard: Single game. This is another fatal flaw of the testing process, and if you play against one of these guys, you’ll see why they advertise this way. Normally, these guys are sporting some kind of deck that’s a little janky, though quite often reasonable, but they always have some kind of ridiculous weakness. For example, it might be a deck based on graveyard interactions or one key artifact. Usually these players advertise single games because they are good game one, but just can’t deal with splash damage from commonly played sideboard cards (Jund Charm, Pithing Needle, etc) and can’t be bothered going 1-2 every match. The next PTQ you play at isn’t going to be best of 1. If you can’t beat 70% of sideboards in the format, make your deck stronger, experiment with a transformational sideboard, or, more likely, ditch the deck altogether.
In this initial phase, it is important to always take note of the gameplan you had in mind when you built this deck, and ensure that all the cards you are playing with are still fundamental to this strategy. Basically what I mean is, if you are trying to attack to win, and there’s a creature that always seems to be blocking or doing some mediocre utility stuff, maybe it isn’t right for your deck. Even if it is ludicrously cool.
Blame every single loss on yourself. Every last one. Once you start blaming your lucksack opponents, or your repeated double mulligans, or colour screw, you stop noticing why exactly it is you are losing, which in this stage could very realistically come down to the flaws in your build. Maybe your colour ratio is wrong, your curve is too high, or you have too many cards that aren’t enforcing your game plan. If the fault is on you, you will notice the problems, and change things for the future.
In this phase, you should also be noting what your good and bad matchups are. This helps with the next phase. No need to make gross generalisations (70-30, cannot possibly lose, etc), but consider whether you are a favourite or a dog for the sake of your sideboard in the first update.
The Middle:
The middle phase of testing shouldn’t be happening a day or two after your deck has come together. That isn’t to say that you aren’t allowed to adjust individual cards within your maindeck, but you should just be testing different cards and numbers, rather than adjusting the deck’s gameplan. Obviously if the gameplan sucks, you may have to address that, but that means you are going back to the beginning. Don’t make the mistake of starting in the middle.
First and foremost, you should probably decide whether or not this is the deck you are going to play. Changing your mind now is going to be a lot better than after this or even worse the end phase. For some people, the decision at this stage will be locked in by the fact they have already bought the key cards for the deck. That’s fine, but you shouldn’t let the money side force you to play all the cards you bought “just because” if there are better ways to go.
Next you are probably going to want to take a long look at your sideboard and improve it for the future. Maybe you kept losing a lot of matches after winning the first game because a certain card either wasn’t pulling its weight or wasn’t there. It’s also reasonable to be testing with a larger sideboard for a while (so long as you aren’t bringing in more than 6-8 cards at a time), so you can try to understand which cards are going to end up being more useful, even if one looks drastically better on paper. This is actually a pretty decent method for your ‘opponents’ deck as well, as when you are playing sideboarded games, your opponent will be too, and you can’t guarantee that everyone will come to the same conclusion about attacking you on point A rather than point B.
So, lay out all your options. Think about what cards are 4 ofs and what can be singletons. Some sideboard cards are 4 or nothings, in my opinion (like Leyline of the Void, for example, in old extended). This will help you work out whether or not you have room to cater for all the matchups you want. I have no problem with ignoring a bad matchup that is only getting worse after boards to allow my sideboard more relevance. Some people will disagree with this, and that’s fine. It’s not a hard and fast rule, but I find it agonizing to spend all this work in games 2 and 3 to lose a matchup 1-2 when I could have won some other round if I’d included card X and never had to face this guy. Just keep it in mind, don’t necessarily live by it. By that I mean if it comes down to improving a marginal matchup or improving a disastrous one (due to available slots), it’s more likely best to let your bad matchup slide. Better to lose one matchup than two.
Adjust your mana base to eliminate small problems in your maindeck. Too many comes-into-play-tapped lands, not enough white mana, or too much emphasis on manlands. Also note that your mana should be updated here to cater for your sideboard update as well – a lot of naya decks pre M11 that I faced online shoved Kor Firewalker in their boards because they were ‘white decks’ and that card beats red (apparently). However, their mana dorks never live against you, so they actually need to mise two white sources, generally by turn two or three, to make that guy relevant. It basically never happened. Dudes complained a lot about that, but the error was in most cases their fault.
Take a good look at your mana curve as well. Are you too reliant on mana dudes to hit your three drops turn two? Are you just cold to red decks in this way? Do you have enough early pressure to beat counter-control decks? Do your mana base and curve speak to each other or fight? Often this is something that is a little hard to adjust in the main if your deck is trying to do ‘thing A’, because you have to include certain cards and omit others. But if there is a curve concern, maybe just chucking a 2 mana beater in your sideboard could be worth considering.
Get rid of the cute cards in your deck that aren’t actually helping your matchups. Your niche combo, random colourless land or piece of miser’s equipment that is way too expensive to use. A lot of people get too attached to certain cards that probably shouldn’t have even made it through the beginning phase – I know I have. Now is the time to cut your babies loose and leave them on the bench for the next deck you make.
Keep playing. Do not stop here. Keep playing at least 5 or 6 games (preferably more like 10 in a sitting) against a certain deck pre board, then break out the sideboards for both decks and keep bashing. Still be willing to talk about the effectiveness of each player’s board cards and consider or test other options if time arises.
Repeat every aspect of this phase until you would be happy playing this list at a PTQ or better.
The End:
Win something. FNM, MTGO 8 man, whatever. Get some confidence and some results to back up your testing. If you like, keep taking notes on how the matches went so you don’t get caught up in the success and presume that you have mastered your build. Even in victory, you can still learn something.
If necessary, jump back to the middle phase and keep refining. It’s possible that you realise card X and land Y need to be changed, but the deck is otherwise fine. It’s also legitimate to say to yourself ‘this should probably be this other card, but…’
I’m expecting more of Deck A – Meaning you are playing a card that is better in your predicted metagame. This is fine, but if you are wrong, your deck will be inferior and you have to shrug it off and use your alternate cards to the best of their ability.
This one suits my playstyle better – Still kind of fine. As a broad example, if you aren’t good at determining the threat level of various spells with your one maindeck counterspell, playing a draw spell in that slot might work better – obviously this example is a little vague, but it’s hard to identify a particular circumstance that might apply to you as it will be based on your preferences and style. It’s easier to identify deck types in this way – ie agro/midrange/control/combo – and making your deck slightly less like one type to get into familiar ground is basically the point I am making.
I’m playing for a slight margin of luck – I agree with this to a certain extent. If you cut three lands and just expect to get there, good luck, but I would hazard a guess that you are very, very wrong. However, I cut a Bird from my nationals deck, figuring there were only 6 standard rounds and I didn’t want to cascade into a bunch of mana guys against control decks, just needing to get lucky maybe once when I wasn’t playing against control or something to compensate. This is fine, as long as you admit to yourself that you are taking a risk and it is for a specific reason. As long as you are open about whatever risk you are taking and accept the blame for when it doesn’t pay off, you can pretty much change whatever you want. Just measure your risk vs reward accurately and don’t sabotage yourself out of a tournament. For what it’s worth, I would have played the extra bird in a ten standard round Pro Tour, though I maintained the same list for the PTQ at nats.
Mostly, just remember that the end phase isn’t the end phase for your deck! You can continue to play with it, and tweak it as the metagame changes at your local store or whatever, but generally if you have reached this stage with the strategy you initially started with intact, you were aiming for a decent result at some event in particular. The whole process has led you to that result in that tournament (or series of tournaments, I guess, if you are PTQ hunting).
Don’t get disheartened if you didn’t tear it up like you wanted. Remember the process, note the mistakes you made in the process (and the successes – just as important to note which bits went really well so you can repeat those efforts in your next cycle) and learn as much as you can. It’s also worthwhile to talk to other people about the process and choices they went through, and see if there were options you hadn’t considered, whether you were using the same basic decks or not. Yes, yes, you got unlucky at times (now it’s actually acceptable to claim that as well, I believe), but don’t let that override anything you can get out of the results for the next time.
To conclude – this is basically my process from the nothing stages to the end of development. Maybe it’s not all right for you, but like anything, tweaking can get you the results you want. Just make sure you identify the reasons why you are deviating from parts of my method if you are using a similar process in your next testing regime.
Until next time,
Cameron Veigel
