Aggro Decks Appear Artificially Over-Represented in Ladder Games

Everyone knows that aggro decks are more common in ladder games (such as best-of-one or best-of-three queue on Magic Arena). There are two obvious reasons for this. The first is that, for the most part, the quantity of games matters more than the quality of games—for the majority of outcomes, be it climbing the ranks or acquiring prizes, it’s better to win 60% of your games in 10 minutes than to win 65% of your games in 30 minutes. The second obvious reason is that aggro decks are cheaper than control decks, especially if you consider that the mana base is the biggest roadblock for a lot of the archetypes and aggro decks are often mono-colored.

Both of these factors make aggro decks over-represented in ladder games compared to real life—there are simply more people playing aggro in online queues than there are in a tournament at any given time. But there’s also another factor most people don’t realize that makes aggro decks appear more numerous even when they are not, and that’s the fact that they play more games in the queue.

Imagine that a game of mono-red takes, on average, 10 minutes to complete, whereas a game of Jeskai takes 30 minutes. If both players play for one hour, that’s six games of mono-red that will exist on the ladder, but only two games of Jeskai. For all intents and purposes, mono-red will be three times more common, even though it’s equally popular.

Let’s assume that there are four people in the queue: two mono-red players and two Jeskai players. They all start at the same time, and the mono-red decks are paired against each other, as are the Jeskai decks. Everyone queues for another game as soon as they are done.

The mono-red match takes 10 minutes. Since the control decks haven’t finished yet, the mono-red players queue into each other again (which in reality might not happen, but will happen if we assume there are more than two players in each archetype). They play another match, and then another. By the time half an hour has passed, they’ve played against three mirror matches, whereas the Jeskai players have played versus one mirror.

When the Jeskai players are done, they get back in the queue. They’re now paired red vs. Jeskai twice. This match takes 15 minutes from both sides. They queue again, and are paired versus the reverse matchup. After 1 hour is done, this is the result:

  • Each red player has played against three red decks and two Jeskai decks
  • Each Jeskai player has played against two red decks and one Jeskai deck

In this scenario, everyone leaves the client thinking that there’s more mono-red than Jeskai, when in fact there are exactly as many mono-red decks as Jeskai decks. Once we add that to the fact that there probably are more mono-red decks than Jeskai decks to begin with (since speed and price are factors), this makes the ladder look like a sea of mono-red.

So, what do you do with this knowledge? Two things. First, you tune your deck to beat the aggro decks. Unless you’re aiming for a top Mythic spot, in which case win percentage might mean more than number of wins, you’re better off beating aggro and losing to other decks since you’ll face aggro more than anything else. This doesn’t mean that you should start maindecking CoP: Red (though, really, you could), but it does mean that you should lean more toward having an early game plan and cheap cards. For example, I’d err on the side of maindecking a Cast Down over maindecking a Negate.

The second thing is to acknowledge that this illusory representation exists and that the ladder metagame is not a reflection of the tournament metagame. If you rarely see control in your online matches (be that Magic Online, Arena, Bo1 or Bo3), this does not mean control will not be present in a tournament—it could just mean that they are taking longer to finish their matches. This means you cannot make absolute conclusions based on the numbers on the ladder, and that the deck you should be playing on ladder is not necessarily the same one you should be playing in person. You could, for example, main deck a Fungal Infection or a Moment of Craving in an online metagame and it wouldn’t be the craziest thing in the world, but you might find that it’s a lot less useful if you play a real life event.

Win Percentage Is Win Percentage

Sometimes, people have difficulty remembering that win percentage is just win percentage, no matter how you get there or how it’s spread between the possible outcomes. This manifests commonly in two forms:

When you’re in a situation where you are looking for a particular outcome, and you get that feeling of “go big or go home.”

Imagine you’re playing a tournament and it’s round 1. You open a certain hand with one land, and you think “well, that’s risky, but I believe this hand is a keep for XYZ reasons,” and then you keep. So far so good.

Now, imagine you are 3-2 in a tournament, and you have to win the next four rounds to make Day 2. You’re down a game. You see your opening hand—the same one as round 1. You think, “it’s risky, and I’m not in a position where I can afford to take this risk,” and then you mulligan. This doesn’t make any sense.

In reality, all that matters is how likely the hand is to win versus how likely a mulligan is to win. If you think this hand gives you the best chance of winning the game, then you should keep it, even if you risk not drawing a land and going home without having any meaningful agency over your last game. If you think this hand does not give you the best chance to win, then you should mulligan it in both situations (though of course whether you mulligan or not in the first scenario shouldn’t impact your later decision—maybe you learned that it was right/wrong after that). Sometimes, players also do the opposite. They think “I can risk this hand because I’m up a game,” for example, and you should not do that.

The same can be true about drafting. A lot of the time people think, “well I need to 3-0, so I’m going to make this risky pick,” and it’s rarely correct. The overwhelming majority of the time, the strategy you use when you have to 3-0 is the exact same you use when you don’t—the one with the higher win percentage. There are very few picks that have a binary “3-0 or 0-3” feel to them because Magic, especially Limited Magic, is very uncertain. You can 3-0 with a bad deck just like you can 0-3 with a good deck, and the most likely result for anyone is a 2-1 or a 1-2 regardless of anything else, so you certainly don’t need to “go big or go home” during the Draft portion for it to work.

There are picks that are more polarizing. For example, taking a pick 1 gold card over a single-colored card is going to either be a higher ceiling (you end up playing that card) or a lower floor (you end up not playing that card), versus the medium floor/ceiling of the single-colored card (you play it more often but it’s not as good). In this spot, taking the single-colored card is seen as “hedging,” and you might not want to hedge if you need to 3-0. But it’s usually the case that it’s either going to result in a better deck on average to take the gold card, or in a better deck on average to take the single-colored card, and that’s not going to change based on the record you need, and the best way to have a 3-0 deck (or to not have an 0-3 deck if that’s what you need) is to go for the strategy that is the one most likely to result in a good deck. Basically, you do not need to pick the most busted card at all points and hope it makes into your deck if you want to 3-0, and you don’t need to go for this risky but potentially great strategy—you can just 3-0 with a normally good deck.

Because of this, I recommend ignoring records, desires and past decisions in your current decisions. You should do what you believe gives you the most chance to achieve what you want, and that’s usually constant if what you want is to win a game of Magic.

When we tend to value a balanced win rate versus the field rather than a polarized one that is better.

As a professional player, I always want the feeling of being in control. I prefer matchups that are 50/50 over a couple of 100/0s and 0/100s, because I think I can play better in the 50/50 and not in the 0/100. This is true, but only to some extent. If my win rate is still higher after play skill with a polarized deck, then I should play the deck that is overall better, even if it has some free wins and some free losses.

For example, imagine all the decks are equally represented in the field. Deck 1 has the following win rates when I play it versus the expected competition:

55% vs. deck A
53% vs. deck B
55% vs. deck C
57% vs. deck D

Now, deck 2:

80% vs. deck A
30% vs. deck B
80% vs. deck C
35% vs. deck D

Most professional players I know would pick deck 1 in this example to play in a PT (myself included, truth be told), but this is incorrect. Deck 1 is 55% vs. the field, and Deck 2 is 56.25% against the field. Deck 2 is just a better deck, but it doesn’t feel like it because it has bad matchups and deck 1 doesn’t. So basically, the moral of the story here is that it’s OK to have horrendous matchups, as long as you have reason to believe your good matchups are good enough (and prevalent enough) to make up for that.

This is also true when it comes down to sideboarding, or even tuning your deck. It’s common to read the sentence, “I’m already OK in this matchup, so I’m not going to make my deck better against it or pack sideboard cards for it—I’d rather shore up my weaknesses.” While there are some situations where this train of thought makes sense, there are many in which it doesn’t, because win percentage is win percentage, whether it’s against your good or bad matchups. Basically, as long as you expect the right field, polarization in itself isn’t bad.

Imagine you’re playing a WW deck. You feel your matchup against Burn is good (you’re 60%) and your matchup against U/W is bad (you’re 30%). You have four slots in your sideboard. You can play four Kor Firewalker, which would make the Burn matchup an 85% overall, or you can play four Unbreakable Formation, which you’ve determined is the best possible card versus U/W since you keep losing to Supreme Verdict (and for the sake of argument let’s say it is—this is just an example. But as an aside, it was really hard to come up with a card in WW that would be even functional as a sideboard card against U/W for my hypothetical example). Unbreakable Formation does wonders for you—it swings the matchup overall from 30% to 52%!

Now, should you play Kor Firewalker, which improves an already good matchup, or should you play Unbreakable Formation, which turns your bad matchup into a good one? Well, assuming you expect the same number of Burn decks and U/W decks, you should just play Kor Firewalker, since it increases your overall win rate by more. If you play Kor Firewalker, your win rate against the field is 57.5%. If you play Unbreakable Formation, it’s 56%.

Sideboarding Has Diminishing Returns Against Good Matchups

Here is the counterpoint to the previous statement. As a general rule, the better you are against a deck, the harder it is to gain percentage against them. A 20% increase against a good matchup is worth the same as a 20% increase against a bad matchup, but it’s much harder to gain a 20% increase when your matchup is already good. In my WW example, this can be true, because Kor Firewalker is such a good card and there’s nothing that’s remotely good vs. U/W, but it’s not often true.

Imagine the WW vs. Burn matchup is already extremely lopsided, and you only lose games where you mulligan to oblivion and where you get stuck on one land. In this spot, clearly having sideboard cards strictly for this matchup is useless because if you have a functional draw you’re going to win anyway, and if you don’t you’re going to lose anyway. The only subset of games you lose are the ones where something goes disastrously wrong, and having Kor Firewalker isn’t going to help with them specifically. Some of the games are always going to be “unwinnable,” and changing your deck composition and sideboarding strategy will not do anything to change those, so the marginal utility of Kor Firewalker is lower.

Besides that, your deck is already great versus Burn. You don’t have any bad cards to take out. This means you’ll end up swapping a good card for an even better card, which is good, but not as good as swapping a bad card for a great card, which you could do versus U/W. Simply put, gaining percentage points in an already good matchup is harder, but gaining percentage points in a bad matchup is usually easy (like slamming Rest in Peace against a graveyard related deck).

So, how do you reconcile these two statements in practice? The best thing to do is to understand that the two concepts exist and to keep both in mind so you can evaluate each situation as it comes. It is harder to gain percentage in your good matchups, but it’s also worth the same as the percentage in your good matchups if you do manage to get it. Most of the time, you should err on the side of trying to make your bad matchups better, as that’s the place where it’s easier to gain more percentage points, but you should keep in mind that the reverse can be true as well, especially when the cards versus your good matchups are much stronger than the cards versus your bad matchups (e.g., it’s much easier to sideboard a ridiculously strong card against a Nexus of Fate deck than it is to sideboard against a deck like Temur Energy because there’s simply no card that’s good enough versus Temur Energy).

Cards Lower in the Curve Can Occupy Higher Spots in the Curve

When we’re deck building (particularly in Limited), it’s very common to lay a deck out in a mana curve. I do it and you probably do it too. It’s also very common for someone to say, “I need to play this to fill my curve” and then play a bad 4-drop to fill a curve that’s full of 2s, 3s, and 5s, which is weird because there’s nothing stopping you from playing a 2- or 3-drop on turn 4.

Cards that cost more mana tend to be more powerful, but they are not more powerful because they cost more mana (it’s the opposite). Sometimes, a 4-mana card will not be noticeably stronger than a 3-mana card, at which point it makes no sense to play it in your deck if you have a lot of 3s already. The 4-mana card has to be meaningfully better than the 3-mana card because you need 4-mana to play it, and two 3-drops are a much better curve than two 4-drops. If I have an Azorius deck that is full of 3-drops and Lawmage’s Bindings, I don’t feel like I need 4-drops, because Lawmage’s Binding is a perfectly fine card to play on turn 4 (and in fact might be better on turn 4 than on turn 3), even if its converted mana cost is 3.

When people say you need a “curve,” they mean you want to be able to play powerful cards as the game continues. You don’t want to be playing “1-mana-power-level cards” on turn 10 because then that won’t be enough. Having cards that cost more mana is a good way of having cards that are more powerful, but it’s not necessary. There have been way too many times where I saw people who were unhappy with their curves and where I moved half their 4-drops to the 5 slot and suddenly their decks look OK, so don’t be completely fooled by a mana curve.

If a Card is “Good in the Main Deck for Limited in Best-of-1,” Main Deck It in Best-of-3

I’ve mentioned it in passing before, but it comes up often enough on my stream that I think it deserves its own section. Whenever I’m playing best-of-1, people say something to the level of, “Disenchant/Plummet effects are better in best-of-1.” This has never made much sense to me. All you care about in best-of-1 is to win the first game of the round, which is exactly what you care about when you’re playing best-of-3. If you thought that adding a Crushing Canopy to your main deck was a good way to increase your game one win percentage, then why didn’t you just do that? You can even side it out in best-of-3 if it’s bad!

There is one exception to this, and it’s very rare. The exception is when you believe your sideboard versus a deck is so good that you’ll always win post-board even if you lose game 1. For example, imagine you cannot ever beat a red deck in game 1, but you have four Circle of Protection: Red in your sideboard (of your Limited deck). In this spot, your sideboard is so strong that you’ll very likely win game 2s and game 3s versus red decks, which means you are okay with always losing game 1. In a best-of-1, you don’t have this leeway, so you might main deck a CoP: Red to give yourself a chance. But this is really a very rare scenario and, honestly, sort of unrealistic, so I’d say “don’t main deck cards you wouldn’t main deck in a Bo3” is a good heuristic (unless best-of-1 dictates how decks are built. If there are a lot more flyers in a best-of-1 for some reason, then you can main deck Plummet, and so on).

In Constructed, it makes sense to change your deck for best-of-1, but that’s mostly because the metagame is just different, which doesn’t happen in Constructed.

Well, that’s what I have for today.