Lately, I’ve seen a number of posts and conversations about the decision to draw in certain matchups. While there are spots in which it’s close, I believe that the great majority of those situations result in a lower win percentage for the person choosing to draw.

Basically everyone understands that the default in Magic is to play. Therefore, whenever someone chooses to draw, they have a reason—they believe they’ve found the exception. My point isn’t that those exceptions don’t exist—they do, it’s just that the great majority of the time you think you’ve found an exception, you’re wrong. An exception should be, “I’m playing Manaless Dredge,” and not “both our decks are slow.”

Choosing to draw is, in my opinion, a relic of a time long gone. It used to be that in a lot of attrition matchups (for example, the Mono-Red mirror), it would be right to choose to draw, as everyone would just remove all the creatures and the last threat standing would win. In fact, choosing to draw in those matchups would often be the mark of a good player who understood the matchup they were playing.

That made sense back then, when threats were weak and removal was good. Nowadays, threats are very powerful, and if you fall too far behind, you can’t catch up. Missing your curve on the draw used to mean you took 3 or 4 extra damage, and then the game continued normally—nowadays, if you miss your curve on the draw, you can end up facing 7 or 8 power with no way to control the game ever again. By choosing to draw, you give up on some free wins that you could have gotten because your opponent didn’t have the right answer on curve.

If you play a card like Grim Flayer, and they don’t remove it immediately, they don’t just lose 2 life like they did before—they lose 2 life, you fix your draws, and it might become a 4/4! The punishment for not having an answer on curve is huge, whereas if you were on the draw, then there’d be several more answers they could have.

Not only are threats more powerful these days, they also often come with abilities attached, which makes trading 1-for-1 a losing proposition. If you look at cards like Whirler Virtuoso, Rogue Refiner, Glorybringer, Verdurous Gearhulk, Earthshaker Khenra, Hazoret, and Torrential Gearhulk,  those are all cards that will give their owners an advantage even if they are removed the following turn. You can’t just react to those cards being played every turn, because the reaction doesn’t give you anything other than simply answering the threat, so every time that they spend a turn playing a threat and you spend a turn removing it, they pull further ahead.

The biggest culprit, other than the general power creep of threats, are planeswalkers. Planeswalkers are naturally hard to remove when you don’t have a board, and they give you a big advantage the longer they stay in play. If you do have a board, on the other hand, they’re often trivial to remove, and that’s a big reason to choose to play every time a planeswalker is involved on either side. If I can play my Liliana into your board of one creature rather than two, then that’s already worth way more than the extra card you’re getting for being on the draw. If I can be the first to play my Gideon and you are then unable to play yours, that’s worth a lot more too.

Then there’s Sealed. Every time I do a prerelease article, people ask me, “is this a play or draw format?” I always answer that “it seems to be a play format to me,” and after I’ve had the chance to play with the set, I’ve always stood by my initial assessment. There aren’t that many planeswalkers in Sealed, but cards are still more powerful than they were before, and catching up is still hard. I can’t remember the last time I thought it was correct to choose to draw blindly in Sealed, though I will choose to draw sometimes if both decks are slow and full of removal.

Some people think that if you have a “draw” deck that you should choose to draw. In this context, a “draw” deck can be defined as a deck that doesn’t care about applying pressure and that is full of removal, preferably cheap (like Magma Spray). I disagree with this assessment—I think that for it to be right to choose to draw, you have to have a “draw” deck and also be playing against another “draw” deck. If you don’t know what they’re playing, you should choose to play. The reasoning for this is that it can never be great to be on the draw, but it can most certainly be awful.

Choosing to play is likely to be correct most of the time, but when it isn’t correct, it’s just a small mistake. You could be, for example, 51% on the draw, and 49% on the play. There’s never going to be a matchup where whomever is on the draw is heavily favored. There are matchups in which being on the play makes you heavily favored, though—there are many games that you will never win on the draw but would win easily on the play, even with what is theoretically a “draw” deck. For this reason, I think drawing blindly is bad. In games 2 and 3, if you both have “draw” decks, then you can choose to draw.

So to sum it up, I think it can be right to choose to draw, but only in very, very, very rare exceptions that most likely aren’t the ones you have in mind. If you’re unsure whether you’re right or not, choose to play, because that can never be too wrong.