Flow of Ideas - Signals Are Overrated
Constructed theory has been plowing ahead since the turn of the millennium. The face of Constructed is wildly different from that of five years ago, and five years ago, the face of Constructed was wildly different from ten years ago. Decks are constantly being built better, card choice theory has become crisper, and the playtesting process has been refined to an art.
That's not true for Limited.
Limited's last real theory revolution was shortly after Magic Online was released. People could draft more than ever before. Suddenly, there was an information overload. Within two years, the finer points of drafting had been codified, and the fundamental maxims of Limited had been hammered into players' minds. For a time, Limited theory may have actually surpassed Constructed theory.
However, the sharp blade of Limited strategy has dulled to the edge of a butter knife.
In the past few years, Magic has changed significantly. The way most people talk about Limited hasn't.
Some great innovations have been made. There's no doubting that. For example, video draft walkthroughs allow more insight into the best players of the world than ever. If you watch enough of them, eventually you'll absorb elements of their style through osmosis.
However, watching draft videos can only take you so far. At some point, it's important to be able to explain what you've learned and put it into terms. Otherwise you just end up teaching everybody ridiculously unhelpful drafting acronyms like BREAK and telling them to go watch Anton Jonsson draft. That's nothing against Anton, but rather that trying to teach a new player how to draft by watching a master is like trying to learn how to pitch by watching Tim Lincecum. You understand that he's throwing the ball, but you can't see all of the finesse behind his choices.
Part of the problem behind evolving Limited strategy is that players want to learn how to draft what's available right now instead of learning how to draft better overall. Players want pick orders. Players want guides. What is often less clamored for is generic Limited theory.
In Constructed, people yearn for decklists in a similar fashion, but decklists in Constructed actually further Constructed theory on their own by introducing new ideas and new ways of building decks. A deck like Critical Mass (hi Flores!) has innovated new ways of thinking about deckbuilding all the way up to today.
In contrast, a pick order is very format-specific. Knowing that a 2/2 for two is a high pick in Zendikar and a low pick in Rise of the Eldrazi doesn't help when drafting Scars of Mirrodin.
There's a lot to tackle in the wave of new-age Limited theory. I hope other authors will help take up this task, and perhaps some of the contestants left in the StarCityGames.com Limited Talent Search will pick up on these issues. While I have plenty to say and the perpetual drive to tell you that eighteen lands is the new seventeen lands in every other article, I'd actually like to look at one very important feature of Limited which has changed a lot: signaling.
The premise behind signaling is simple: you try and take cards of a certain color to keep the person to the left of you from taking cards of that color. Since the people to your left weren't taking that color, in pack 2, you're rewarded with cards of that color. Then, since you cut off the person to your right from that color in pack 2, more cards of that color should come in pack 3.
When this method was first introduced, it had flaws but still worked well enough. Today, it's fraught with problems. Furthermore, new players are being taught signaling like that was a drafting doctrine, giving them all kinds of wrong ideas.
When the packs were weaker overall, signaling worked better. (Though it certainly wasn't flawless.) When there were only a handful of cards you wanted in your deck per pack, it was easier to arrange signals. Today, so many cards in your pack are playable that you can try to actively cut a strategy with every single pick and still end up feeding the person next to you!
Compare a pack from an older set with a pack from a newer set. Here's a randomly generated pack of Odyssey from Magic Workstation. I just took the first pack it generated.
Sphere of Grace
Rites of Initiation
I'm not as proficient in Odyssey draft as some, perhaps, but I think the best cards in this pack are probably Patriarch's Desire, Werebear, and Wild Mongrel. After that, there's a steep drop-off afterward, and you hit cards which, if passed, don't really signal too much. Here, trying to cut something (like, say, taking the Desire and cutting black) is possible.
Compare that with this Magic 2011 booster pack randomly generated in the same fashion:
Sign in Blood
Act of Treason
Eleven of these fourteen cards I'm happy to have in my deck. Pegasus, Adept, Basilisk, Drake, Slime, Druid, and Maze are all legitimate first-picks if you look solely at power. Sign, Negate, Excommunicate, and Vampire are all cards I'm happy taking after my first couple of picks and playing with. Act of Treason isn't great but fine in the decks that want it. Alluring Siren is a card I don't usually want to make my deck, and the Cocoon is unplayable.
Let's say you want to try cutting a color. You take the Stormfront Pegasus and ship all of the good blue and green. Then the next pick, you end up being passed a similar pack, except it has two or three good white cards. Fair enough. There's hope for cutting left. The next pack has two decent white cards, but it also has a card in another color that's better than either white card.
At this point, not only is cutting impossible, but you're actively harming yourself by doing it. If you'd just taken a different card first pack and been more flexible in your next couple of picks, you could've had several strong cards. Even if you'd passed a few cards in a different color, you'd still be able to dry up that color in pack 2 and then picked up plenty of cards in pack 3.
The short version is this: packs are way deeper than they used to be. Where before cutting actually sent consistent signals, there's no guarantee that's going to happen today.
Can cutting work? Sure. The problem is that the packs have to break perfectly, and that's not something you want to rely on. On top of that, even in the drafts where cutting does work out, your decks only end up marginally better. You end up with far more playables – but as a result, you have to cut several of them. Maybe two or three of your cards are upgraded, but a lot of the rest of your deck remains the same.
Furthermore, the card quality is just so high in general that even if you get cut, you'll still likely end up with plenty of playables.
Mike Flores widely claimed throughout Magic 2011 drafting that he was uncompromising in drafting blue. He'd always draft blue, no matter the circumstance. On the surface, this sounds like a terrible idea. However, Mike managed to stick to his plan every single draft and always ended up with a reasonable blue deck. Even when wedged between two other blue mages, he could still support his blue decks.
How is that even possible? Let's look at it in detail.
Assume in each pack, there are two good blue cards as a baseline. Sometimes there will be more, sometimes less, but two seems like a good midpoint. Mike opens his pack and first-picks one. Then he gets passed a pack and second-picks another. The same thing happens in packs 2 and 3. That's six good blue cards.
On top of all that, Mike can round his blue out with cards like Negate, Cancel, and Phantom Beast which will come later on, and many neighboring players will pass on. Even if your last few cards are some of the weaker blue cards, they're not that much worse than other cards you could've drafted – plus you still have your six good blue cards to work with.
Next, there's the issue of picking another color.
The traditional plan with signaling was that you would let your signals dictate your first color, ensuring you'd be hooked up with plenty of it in the second pack, and the signals of players to your right would provide your second color. That's no longer the case.
Back in sets like Odyssey, that may have worked. Compared to today, a vast majority of cards were on the weaker end of the power scale. As a result, you had to cooperate and read signals to end up with a strong deck.
As you might have guessed, that's not true anymore.
It used to be that you were done for in a draft if you didn't have your direction firmly set by the end of pack 1. Today, you don't even have to know what colors you are as you open pack 2. Over two-thirds of the average pack is probably playable. You can just take the best cards that are being passed to you which roughly fit your colors and end up with a good deck. I've switched both of my colors multiple times after pack 1 and still ended up with a perfectly good deck.
Is Magic better this way? Some Draft pundits will say yes; others will say no. In general, I'd say it's just different. Drafting is a lot less about cooperating with your neighbors and much more about knowing the format and figuring out how to draft archetypes.
Now, this isn't all to say signals are useless and should be avoided. They have their place. However, far too often in pick orders, I hear about players taking Lightning Bolt over Doom Blade just because there was a Gravedigger and Black Knight in the pack. The value of that signal is significantly lower than the increased value of Doom Blade over Lightning Bolt.
The one area where I feel signals are important is when you're passed two consecutive packs mid-pack with a lot of strong cards of one color. When you see the second pack with two or more good cards in a color, it almost always means that color is open.
However, I often don't feel that's really signaling. Maybe in the technical sense of the word but not in how it's actually applied. There are many instances of “signaling” that I feel is actually just taking the best card for your deck. For example, when a card such as Volition Reins wheels the table, it should be pretty clear that blue is open – but is that more of a signal, or just knowing to take the best card?
Finally, I want to leave you with one of my all-time, favorite quotes. It came from Hall of Famer Alan Comer at a casual draft I was at years ago. Being the naïve, signal-focused youth I was, I made some offhand comment about the importance of figuring out how to send signals during a draft. Alan smiled in a way only a seasoned veteran could and replied while busy making his pick, as though he had recited the sentence hundreds of times.
“Funny, isn't it, how everyone is always so focused on sending signals, but so few focus on reading them.”
Think about that one.
In any case, I look forward to reading your comments and thoughts. I'm sure there are many out there who have something to say on the application of signals in modern Draft formats, and I'd love to hear what you think! Please either post in the forums, tweet at me @GavinVerhey, or send me at e-mail at Gavintriesagain at gmail dot com with your thoughts.
Talk to you soon!