"Focus on what matters."
Certainly you've heard that phrase before, but how are you supposed to know what matters until someone explains it to you? I certainly didn't.
Well, let's start simple.
Example #1: They have a 3/3 and a 2/3, both with no abilities. Which do you kill with your lone removal spell?
Instinctively, you know the 3/3 is "better." However, what happens if you have Wave of Reckoning in your deck? With Magic, there are infinite variables, and you should be trying to consider them all. Of course, we're only human, so sometimes we miss things. If you are well practiced with whatever deck you're playing (and this could be as simple as memorizing the contents of the deck you just drafted), you should be able to make these decisions quickly.
Let's delve a little deeper.
Example #2: They are a relatively fast Boros deck and led with a Skinbrand Goblin. Do you kill it?
One of the most helpful things I've learned about Limited is to only use your removal on something that is literally causing you to lose the game. As Magic has sped up, that sort of conservative thinking doesn't exactly work anymore.
If you kept a slow hand with Knight of Obligation, Angelic Skirmisher, and a removal spell (let's say Orzhov Charm for this scenario), the damage Skinbrand Goblin would deal might cause you to lose. Assuming Knight of Obligation sticks, you're gaining some virtual card advantage since their Skinbrand Goblin will likely do nothing.
However, there are such things as bloodrush and battalion to consider when making decisions like this. Their creature might be worth more than the 2/1 body. If they have Martial Glory, a good bloodrusher, or a removal spell of their own, Skinbrand Goblin is going to continue to hit you. Over the course of a few turns, that damage adds up and will ultimately be the cause of you losing the game.
It's a bit off topic, but these days I prefer my removal to be used offensively rather than defensively. I also want it to be cheap. If I'm forced to Murder their two-drop, I'm losing a mana in the process. Something like Mugging is much more appealing to me than something like Angelic Edict.
Anyway, in the above scenario, I would snap kill that Skinbrand Goblin, hoping to buy time until my late game bomb can take over.
Example #3: You play a Nezumi Graverobber on the play against a good player. On turn 3, you attack, and he casts Candle's Glow before damage goes on the stack, giving you the opportunity to flip your Graverobber safely. What the hell?
The above scenario actually happened to me against Jon Sonne in a Grand Prix. At the time, I was thoroughly confused. His turn 3 Kitsune Blademaster further confused me. He had a way to brickwall my two-drop, so why would he "waste" a card and give me the opportunity to flip my Graverobber?
I quickly Rend Fleshed the Blademaster and continued attacking. His turn 5 Kumano, Master Yamabushi made me look incredibly foolish for using my only removal spell, and I lost a few turns later.
In hindsight, it was clear was Jon was trying to do, and I took the bait. He wanted me to skip my turn to transform my creature. He wanted me to kill his Blademaster. All of it was part of a setup to both buy time until he could drop his bomb and hopefully bait out the removal spell.
Oddly enough, there wasn't much I could do in the situation even if I sniffed out his plan. My deck had a pair of Rend Fleshes for removal, and I couldn't profitably attack around his Blademaster. If I save my Rend Flesh for his Kumano, I probably still lose to the Blademaster plus his other spells in a long game despite my Nezumi Graverobber.
Example #4: Your opponent is a good player, and they lead with Dregscape Zombie. You have a Knight of the Skyward Eye on turn 2. They play a second Dregscape Zombie, but they don't attack. Again, what the hell?
This above scenario happened at a later Grand Prix, this time against Bill Stead. Because I had played through the whole Sonne scenario years earlier, I quickly identified Bill's plan. He knew that two damage would likely not matter because he was counting on his bomb to win him the game against my removal light Naya deck. If he attacked with the first Dregscape Zombie and I had Magma Spray for the second plus another creature for a follow up, he would likely fall behind. At that point, perhaps he would be low enough on life that his bomb wouldn't save him.
Instead he sat back, hoping to preserve his life total and buy time until he could cast Flameblast Dragon. I went into hyperaggressive mode, offering worse trades for me (that could have been avoided by waiting) in order to get in more damage. He missed a key land drop, allowing me to take the game and match.
Example #5: You have a 3/3, and they have a 4/4. Is it worth it to bluff attack?
It depends. It rarely matters if they think you have a trick or not. Usually, it comes down to the value they place on their 4/4 and what role they think they have in the game. Do they think they are the control deck? Do they think they need to win by racing? Most of the time, it's impossible to determine that without seeing their hand or reading their mind.
If I was the one with the 4/4 and you attacked into me, I'd consider why would you'd be attacking, what tricks you could have, whether or not I could play around a trick once I got to untap, how important the damage is, whether I was in the controlling role or about to become the beatdown role, etc. Only after I considered those things would I make my decision.
If you can't figure it out from the 3/3 side of things, you need to reconsider your role in the game. Are you the beatdown? Are you going to be able to win a "fair" game of Magic without being tricksy? If not, you might have no other option aside from bluffing. If they're going to stabilize (because you've flooded out, for example), then sneaking in that three damage might be the only way for you to win the game.
It's also important to consider things like Spark Trooper and Lava Axe when making bold decisions like this. Sometimes that extra damage you get in will lead to victory if some of your topdecks have a lot of reach.
Even if there's a 1% chance they'll take it but you're going to lose if you don't, obviously you choose to attack. You can slightly increase your percentage by not agonizing over the decision. If you're forced into making a move like that, you have to sell it.
First of all, you need to consider what role you're in, similar to Example #5. If you're the beatdown, you're likely attacking, but you are never pigeonholed into that decision. Things like Ember Beast, Massive Raid, and battalion all make Skinbrand Goblin more valuable. Sometimes you need to slow down and rethink how valuable a card seems. Even trading with Syndic of Tithes, a card much better than Skinbrand Goblin, might not be worth it.
Chances are that they think they are the control deck and are considering things like Ember Beast, Massive Raid, and battalion and will make the trade. It boils down to one of those situations where even if your third turn play is going to be Ember Beast, it might be worth it to attack if your hand isn't capable of dealing twenty damage.
If you attack and they snap block, what does that mean?
Example #7: You're on the play against a G/W opponent who led with two-drop into Loxodon Smiter. Do you pull the trigger on Supreme Verdict now or wait, hoping they commit another creature to the board?
Unfortunately, Magic is in a state where your opponent's follow up will probably be something like Strangleroot Geist, Garruk Relentless, or holding up mana for Restoration Angel. If it's post-board, they could even have something like Rootborn Defenses or Nevermore. Pulling the trigger on Supreme Verdict is likely correct even if you think you could get some extra value from it.
Is it worth six damage to take the risk?
The power of various cards changes depending on what's going on. Careful, calculating players will save their counterspells for things that actually matter instead of just using them to trade one for one with things.
If you're at five life, do you use the Dissipate? What about twenty?
Example #8b: They are a midrange Jund deck, hoping to defeat your pristine U/W/R Flash deck. If they cast Liliana of the Veil on turn 4 after you've both been playing a draw-go game, do you counter it?
As you should know by now, there's not enough information to know for sure. If you can cast a Restoration Angel and attack it for three, I'd say that Liliana is basically useless for them. You'll both be discarding your worst cards, and in two turns they'll lose their Liliana.
How much is a planeswalker's loyalty worth though? Again, it depends. I've played games of Magic where I've attacked a planeswalker, then something changed and I attacked them, then something changed and I started attacking their planeswalker again.
In this case, the Liliana's loyalty is worth a decent amount. The first Restoration Angel attack means that Liliana won't be able to kill Restoration Angel. Past that, Liliana is mostly doing the same stuff, although can eventually kill the Angel or destroy half of your lands.
Typically, attacking a planeswalker until it's dead is a solid game plan, but everything changes if you think they are sandbagging a second copy of that planeswalker. Keeping Liliana at one loyalty with an Augur of Bolas while hitting them with Restoration Angel might prevent them from -2ing Liliana, casting another one, and -2ing that one, wiping your board.
In the end, everything has a worth that varies. Sometimes it's more valuable for you to not kill your opponent's things.
Example #9: You have a Tormod's Crypt or Nihil Spellbomb in play against Dredge. When do you blow the trigger?
The general rule is that when you have a finite number of ways to attack their graveyard you use it sparingly, much like creature removal in Limited. If they discard a single dredger, you could use it, but visualize what happens when you don't.
They'll mill into a Narcomoeba and a Bridge from Below. Those are tempting, but it's usually no big deal. They skip another draw step to dredge again, this time hitting more juicy things. Would you be dead if let them get another Narcomoeba? No? Then let them keep doing what they're doing.
Eventually, it will get to the point where they're casting Dread Return, hoping you'll blow your Crypt. At that point you can oblige them.
I've seen several games lost due to a lack of patience on the part of the Crypt player. It might suck to let them get value from dredging, but every turn you don't use it is basically another Time Walk.
Does it all make sense now? Well, it probably shouldn't because this type of stuff is mostly learned through practice. However, now that you know what to be looking for, what you should evaluate, and how, the process should become a lot easier.
At least, I hope so.
@G3RRYT on Twitter