PV’s Playhouse – Professional Playtesting


Today’s article will be about one of the most important aspects of tournament Magic — playtesting. Most of the time, when people say they’re going to “playtest”, they just jam a bunch of games with no purpose and, though that’s not necessarily a bad thing, there is a lot more to it than just playing games mindlessly. Today, I’ll try to elaborate on that by selecting a few short tips that I think you should follow.

Have a Reason to Play

The first thing to keep in mind is that playtesting is answering a question. When there is something you want to know, you playtest, and, depending on the answer you seek, your process and your focus are going to be a little different. For PT Barcelona, for example, we actually played a lot, but we didn’t know what questions we were asking — we were just playing — so we found no answers.

Sometimes, you come up with this great deck you want to test, and you say, “hey guys, someone play me,” your buddy shouts “sure!” and shows up with his four-color Tezzeret deck — at this point, someone has to stop this madness and tell the other person you’re not getting anywhere. Playing two rogue decks against each other is something you should simply never do, because it serves no purpose — in all but the latest of stages, one of the decks should always be something that you expect to face in the tournament.

Don’t be Afraid to Try New Ideas

The first reason to playtest is to find out if a new deck is viable. I’m a big fan of trying everything you think is good — I’ll build every idea that comes to my mind as potentially strong, even though people yell at me. For example, Owen looks at almost every deck I build and says, “this is horrible because of this, this and this.” The problem is, if I thought the deck was horrible, I wouldn’t be building it — even if I agree with his points, maybe I think it’s good in other areas to make up for that. This is not to say he shouldn’t give his opinion — obviously he should — but that should not stop me from building the deck and trying it.

This works for cards as much as it works for decks; sometimes someone will say, “I like this card better than this other one,” and I’ll say that I want to play with it a couple times first to try. If I don’t like it, then I’ll try the new card. You don’t have to try a deck against everything to change it — if you feel that something isn’t working, just change it on the spot and keep going. What you have to do is test the final version against everything — if I beat control, then play against aggro and decide I want -4 [card]Duress[/card] +4 [card]Terror[/card], then I have to play against control again to see if I still beat it.

You don’t have to test as much as you did before, though — you already know how the match goes, and a couple games will tell if it changed radically or not. One way to try out new cards without changing the deck is to simply think, every time you draw a card, if it’d be better as a different one — if the answer is yes most of the time, then change it! Some people even like playing with split cards (i.e. you can use your card as either [card]Galvanic Blast[/card] or [card]Incinerate[/card]), but I’m not a fan of that — I’d rather just imagine whether I’d prefer that Incinerate to be a Galvanic Blast or not.

Play Against the Best Deck First

Your first challenge should always be the most popular deck — right now, in Standard, that would be Delver. When you do that, you’ll usually be able to see very quickly if it has a chance or if it doesn’t. If you’re getting destroyed, then your goal is to identify why. If you can pinpoint a card that kills you every time, then change your deck and try to beat that card. If you can’t point at one single culprit, then I recommend not wasting any more time, and to just ditch the deck. This may seem a bit harsh, but realistically, you don’t need more than five games to see if you have a chance or not.

“How can you figure out anything with only five games?!”

Well, sometimes it takes only one game. What you need to analyze is the scenario: if everything went in your favor and you still didn’t win, then you’re never going to win. If everything that could possibly happen for a card to be good happened, and it still wasn’t good, then it’s never going to be good. Imagine you decide to try [card]Stone Rain[/card] for a certain matchup, and your opponent stalls on three lands. You destroy his third land, and he never plays another one — he spends the entire game with two lands. Now that’s the dream scenario for [card]Stone Rain[/card] — it’s never going to perform better than this. And if it still doesn’t help you win, if your opponent still beats you — clearly, [card]Stone Rain[/card] is not for this matchup. You don’t need any more games to figure that out.

Don’t be Afraid to Dismiss New Ideas

When it comes to testing, I hate wasting time. The reason I can afford to build every idea that crosses my mind is that I have no qualms about dismissing them immediately. Sometimes it takes me three games to realize the deck is not good, and I’ll just throw it away with no hurt feelings.

You shouldn’t make excuses for why you’re losing, you shouldn’t complain that your opponent’s deck is built just to beat you, you must see that you’re losing due to a fundamental problem that can’t be fixed (i.e. your deck sucks) and that’s it, you move on — you don’t have to spend your days thinking, “oh but what if that deck was actually good,” and you won’t waste much time. Someone trying a deck for a couple games is fine, but insisting on a deck that clearly cannot win a game to save its life is very irritating to me — even if my time is not being wasted, it’s the principle!

Yes, yes, I know what you’re thinking — “what if I lose horribly to the best deck but smash everything else? Shouldn’t I play against other decks before I dismiss it?” — No, I don’t think so. Let’s be honest, you’re not going to smash everything else — no deck does that. You might have a favorable matchup against most other things, but if there is a most popular deck and you get destroyed by it, then I don’t recommend playing the deck anyway — so there is no point in trying it.

Sure, once in a lifetime you’ll actually come up with a wacky combo or turbo-fog type of deck that can’t beat 30% of the format, but can’t lose to the other 70%; but people have limited time, brainpower and will to playtest — on average, decks are not like those, and you’re much better off spending your time trying other stuff.

If you are actually trying one of those combo decks, then by all means go for it — but to think your UR control deck is going to beat everything else by such a significant margin as to make up for its resounding losses against Delver, you’re just fooling yourself.

Have a Generic Aggro Deck in Your Gauntlet

There might be a reason why you can’t test against the most popular deck — either there isn’t one or, most likely, someone is already using all the copies you have. Alternatively, you passed the test (which doesn’t require beating the best deck — just not flat out getting destroyed).

If any of those happen, then you should play against a generic aggro deck — anything works, really, as long as it’s creature-based. Playing against an aggro deck will expose most weaknesses of any deck, since it’ll put it under pressure — it will have to work against actual opposition.

Again, losing by a little bit is not a deal breaker, since fixing an aggro matchup is often easy; and your deck naturally won’t be very tuned, whereas the aggro deck is either established or won’t change much. If you get absolutely destroyed, then you might consider abandoning your idea — it doesn’t matter what tournament you’re playing, aggro is always present in big numbers.

Don’t Get Personal

People are, as a whole, very attached to their ideas — you see that every time someone builds a deck and another person criticizes it. Even if they are obviously right, they get an avalanche of defensive comments — it sounds like you’re offending the person and not their material.

This is not only true with Magic — you see it everywhere. In college, for example, some people will get mad at me if I tell them I dislike what they’ve done in the group assignment. I understand the behavior, but I think it is irrational — if I tell you something you wrote is not good, it’s not because I hate you or because I think you’re incapable, it’s because I think we can do better.

If people get too attached to their ideas, you have to walk on eggshells with everything you say about their decks, and, frankly, this is a terrible environment for playtesting. I want to be able to be brutally honest with someone, and if I can’t do that, then I don’t want to test with them. Owen says my ideas are terrible because he knows I won’t make a big deal out of it, I’ll just listen to what he has to say. Conley, for all the hard time I give him, never takes it personally when I criticizes his decks (if he did I’d probably be dead by now).

The point of being in a group is to find the best deck, and you’re not going to do that if you have to hold back because you’re afraid to hurt someone’s feelings. In Barcelona, for example, we were all pretty much set on a deck, but Kibler kept throwing obstacles in our way, such as, “you’ve only tested against this version, you have no idea how this does against anything else.” Part of me hated him for it, because we had something we liked, and why did he have to go and ruin it for everyone — but it was obviously good that he brought those issues up. He was right, and if he had only said it earlier then perhaps the PT fiasco wouldn’t have happened.

Some think that you can only criticize if you have a better idea — this is one of the worst misconceptions people have about dialog. No, I don’t have a solution, but what am I going to do, pretend I don’t see the flaw? In the end, we can still go back to the flawed version if we find nothing better — I’m not eliminating it simply by criticizing it — but it’s important that we understand that it is the flawed version, so that we may look for something better — even if it turns out that a better version doesn’t exist. If you see flaws in a deck or process, point them out, even if you don’t have a solution — if your friends hate you for it, find new friends (or at least new playtesting partners).

Play the Deck You Want to See Win

This might conflict a little with what I said before about not getting too attached, but they are different things — if you have someone who wants a deck to be good, then you have them play it, because they will try harder. The worst case occurs when the two people playing want the same deck to win — the win % increases by a lot when that happens.

I remember we once left Brad and Kibler to play a matchup, and then it ended up being overwhelmingly in Kibler’s favor — but we didn’t really believe any of it. We knew Brad wanted to play Kibler’s deck, so he’d be prone to making worse plays, not mulliganing worse hands (or mulliganing better hands), and scooping too soon, because he also wanted that deck to win.

Then we got Web to play in Brad’s place, and we trusted the result a lot more. Not because we thought Web was a better player, but because we knew Web wanted to play the deck Brad was playing, so he’d give it his best.

Playtest the Decks, Not the People

I’m all for playtesting to learn how to play a deck or a matchup, but you’ll usually do that once you’ve already chosen a deck, or at least narrowed it down by a lot. When you’re in the process of testing a deck, it is better to make sure both decks are being played correctly — if you realize an obvious mistake, take it back; but more importantly, if you realize your opponent made an obvious mistake, then point it out!

Let’s say your opponent didn’t notice he could attack with a guy that you weren’t able to block and missed two damage, what purpose does it serve to let that stand? Just tell him and take the two damage, you want a correct assessment of whether a deck is going to beat the other. If you’re watching, and one of the players makes a glaring misplay (not a misjudgment, just lack of attention — like a typo), then intervene!

Playtesting Is Not About Getting Data, It’s About Understanding

This is the most important thing I can tell you regarding playtesting. Often, when people come with, “we played 10 games and I won 6-4,” they understand that it means very little. The problem is, when they say, “we played 100 games and I won 60-40,” that also means very little!

Now, don’t get me wrong, I’m not saying results are not important — in fact, I believe the only thing you should keep track of are results (basically, write down or mark with cards how many games each deck won — that’s all I ever write during playtesting). I’m just saying that, without other things, results are almost meaningless. You can look at a couple dailies and get a bunch of results — they don’t mean much.

It’s important to know why things are happening, why am I losing/winning? If you don’t know why, then you can’t fix it. More importantly, you need to know how the games felt — to the people playing, were those results accurate? This is the reason I insist so much that you “be detached.” If you aren’t, then you can’t trust feelings and you have to go by result, which I think is the inferior method. Here, you have to be able to say, “I think I only lost because you got very lucky,” without fear of sounding like a sore loser — the only way you’ll be able to do that is if you only say that when they actually got lucky.

Sometimes, you’ll get a record, but then you’ll feel that it wasn’t representative — maybe the other person mulliganed to five every game, maybe they lost their second land drop every game, or maybe you did. If this is the case, and if people can be truly unbiased with their feelings, then feelings are more important than results.

I remember, for PT Honolulu, we were testing RG Ramp versus Delver. I played the Delver side a bit, and I was winning a little more than losing. I felt like this was how the match should go — a slight advantage to Delver.

Then Ben said he didn’t think the match was bad, and he wanted to try his new sideboard cards — [card]Batterskull[/card]s, [card]Garruk, Primal Hunter[/card]s, among others. I didn’t think those cards were good. We played and he won the first three games, and he immediately said something like “see, the match is good, you guys lost those games because you weren’t playing correctly. You were probably walking right into his [card]Mana Leak[/card]s or something.”

However, I had played those matches, and I knew that was not the reason they had lost — I knew the other guys were not walking into my [card]Mana Leak[/card]s. The reason he had won those three matches was that he had gotten luckier than the other people. That left me somewhat angry, and the fact that he had just beat me with [card]Garruk, Primal Hunter[/card], a card that I knew was horrible in the matchup, didn’t help. I took the, “oh come on he can’t get lucky four games in a row!” approach, and played recklessly, walking into things I didn’t have to — basically just threw my cards in the order I drew them, and hoped “he didn’t have it,” so of course he ended up winning even more — to a point where the overall matchup record was positive in RG’s favor.

In the end, though, that didn’t change my opinion of the match, or of the sideboarding strategy. During the play of the games, I felt that the match was slightly better for Delver, and the result we had gotten to was just an anomaly. I made my choice between Delver and RG fully aware of the fact that the match against Delver was even at best, if they played well.

Have Cards with You

This is only half a piece of advice, because sometimes there isn’t much you can do — you can’t make cards appear out of thin air, and I’m not saying you should buy [card]Moat[/card]s so you can test Legacy, even if you might end up not playing them. Living in the South of Brazil, I know that it is often hard to get cards.

Still, if you can make an effort to have real cards, do it. If you can’t, then maybe print the cards for better proxies — not having real cards discourages people from building decks, because it’s an enormous effort to proxy everything, and some decks are just impossible to play with proxies. I’d rather not play a Storm deck if I have to read every card as I’m comboing, for example.

Whatever you do, remember that other people might want to play those decks too, so proxy in cards that make sense — I often forget that, and people grab my decks to play and have to ask me what everything is. The worst is when someone proxies on a basic land that the deck plays; so you miss the writing, keep the hand — then on turn two find out that this land you were going to play is in fact a [card]Primeval Titan[/card] (Matt Nass does this a lot, you’ve been warned).

So, to sum up, here is how a playtesting session usually goes for me:

1 – I have a new deck I want to play.

2 – I grab someone to play the best deck against me (a couple games is fine — 5-10). If I lose horribly, I ditch the deck very quickly. If not:

3 – I play against a random aggro deck, again 5-10 games.

4 – I ask my opponents how they felt about the deck.

5 – If I like the outcome, I keep testing against everything else, then go back to the best deck if things changed radically.


1 – There is a specific matchup I want to try between two established decks — perhaps I want to try a new card, or I’m not sure how it goes.

2 – I grab someone to play that matchup against me.

3 – We play 10-game sets, and note the results. Then, we talk about how the games felt, why we thought that was the result, and if we think it’s representative. If we want to change a card, we can do it in the middle of the set, but we start a new count (i.e., “we went 3-2 with Galvanic Blast and 1-4 once those became Bonfires”).

4 – Ideally, other people play the matchup too — it’s important to know if everyone gets similar results and feelings, because people might be playing the matchup in radically different ways. I always try to get people to play the same matchup, and then I watch — you look at things with a different perspective when you aren’t the one playing. It’s also much easier to see if someone is getting lucky or not when you’re not one of the players.

Above all, always do three things 1) be honest with yourself and with your friends; 2) don’t get personally offended when your friends are honest with you, that’s the reason they’re there; 3) look at other things than just results.

Well, this is it. I hope you’ve enjoyed it and I hope I’ve helped you improve your playtesting sessions! See you next week,



Scroll to Top