Sometimes the correct play is very difficult to tease out.

Dissecting an Angle

  • Boards
  • Print
Author Image

The letter F!irst off, a few apologies about errors in my last article. In Exercise 1, you were meant to be on 11 life rather than 13 to make the blocking of the Ashcoat Bears necessary. Most people figured this out for themselves, but I correct myself here to placate any irate or confused readers.

I had intended this week's installment to be a detailed draft walkthrough. However, due to the massive amount of response from both friends and the forums about my last exercise, I will discuss it in detail. It is not often you come across an example that appears so simple and yet is so deep. To my knowledge, no single play has been analysed in this much depth. To start with I present a slightly edited version (as I forgot to include some details last time).

The Brave Seer

You have three cards of average worth – an okay trick, a solid creature, and a good creature. You and your opponent are both at 16 life. Neither of you, to simplify things, is splashing a third colour. All plays leading up to this game state have been correct and no information could possibly have been gleaned from them – in short, this exercise is designed to focus on card possibilities rather than on information, and the previous turns and how the state wound up as this are of no concern to us. It is your opponent's turn. He has drawn a card and attacked with his Fathom Seer. He has not made a land yet. What is the correct play?

16 life
Three cards in hand
In play: 3 untapped Islands, 3 untapped Mountains, an untapped 1/1 Goblin token, and a tapped and attacking face up Fathom Seer

16 life
Three cards in hand
In play: 3 tapped Forests, 2 tapped Mountains, untapped Nantuko Shaman and Ashcoat Bear

I deliberately left the knowledge of your hand ambiguous. I did this because without this knowledge, speculation as to how you would play becomes even more important. When this board position occurred, I was the blue-red player – your opponent. I had ascertained that my opponent's hand was comprised of a trick (not necessarily a combat trick, but a spell that was not a creature) and two creatures, one of which he was trying to protect, and I had to work out whether or not I could attack. At the time, I deemed it safe to attack with the Seer.

If you examine the exercise from this new angle, it soon becomes apparent that you should not know the exact make up of your hand. Once I had worked out that it was my Seer I wished to attack with, it was up to my opponent to work out what I could be holding so he could work out how to block. Having certain cards in hand makes blocking certain, whilst others mean no blocking is an option. By leaving them unknown, you had to work out your course of action given each scenario, but more on this later.

I've run this situation via several other pros, and I will endeavour to show their take on what should be done. It should be worth noting that many of them differed in what they would have done.

Let's look at the options you face:

A) No blocks
B) Double block
C) Block with the Ashcoat Bear
D) Block with the Nantuko Shaman

A) No Blocks. This is certainly an option. You are on a very high life still, and it is obvious that your opponent has a trick. If he is bluffing, then you lose out, but this is such a bizarre play he is unlikely to be bluffing and you have little to lose. Julien Nuijten, Jelger Weigersma and Olivier Ruel chose this option. I will be going into their reasoning later.

B) Double block. I want to state here and now that no one should take this route. If the attacker has either a Brute Force or a Bogardan Rager, your whole board position crumbles. It is not worth gambling whether you win or lose this game over one point of damage. Unsurprisingly, all pros immediately dismissed this as a viable play.

C) Block with the Ashcoat Bear. This option is much like not blocking. However, it distinguishes itself in one important aspect. It asks, "Now what?" The Bear is unlikely to force a trick out but might make the opponent reveal the reason for his attack – be it a Grapeshot, a Subterranean Shambler, or what not. However, it is very likely that it will not ask the question forcefully enough for him to actually do something. If he does nothing, then you save yourself a point of life. If he does something, you find out what his trick is. Sam Gomersall, Tiago Chan, and I, initially, chose this option.

D) Block with the Nantuko Shaman. If blocking with the Bear asked a question meekly, this shouts it. This block forces your opponent to make a play or put his Seer in the bin. This is the most obvious block, and your opponent must have considered this when he made the attack. He expects your gut reaction to be this. The result of this block will most likely put you Shaman in the bin along with either a trick from your opponent or the Seer. Is it worth trading the Shaman for a trick? Is it worth trading for the Seer? Can you afford to give your opponent that option? Rich Hoaen, Kenji Tsumara, Rogier Maaten, Jeroen Remie, Frank Karsten, Big Oots, Raphael Levy, Mark Herberholz and, eventually, I thought this was the correct option. We will see why later.

A few things to consider: You want to find out why he attacked as soon as possible so that you will not make your future plays in ignorance. The Bear is currently 'turned off' by the Seer, so a dead Seer will be a good thing. Any creature he makes after combat will likely dwarf the Bear anyway. There is not much value difference between the Shaman and the Bear – sure, one has extra power and is thus more handy in dealing with cards like Subterranean Shambler, Coalstoker and Viscerid Deepwalker, but both are likely to soon be outclassed by both decks' six drops. Why didn't the Goblin attack or, rather, why did the Seer attack alone?

Now that we have examined our options, let us examine what our opponent's motivation could be for this attack, especially given our only other piece of information – that the Goblin token did not attack. Here is a list of the cards we suspect he might be holding (there are others, but they can be dismissed very quickly so are not listed, e.g. Word of Seizing):

Fury Charm
Brute Force
Piracy Charm
Fire Whip
Subterranean Shambler
Crookclaw Transmuter
Bogardan Rager

Fury Charm, Brute Force, and Ovinize

These are all reason why the Seer would attack over the Goblin, because the Goblin is always blocked by the Bear and none of these tricks is worth trading for the Bear. You should feel happy trading the Shaman for any of these cards, especially as it might cause a loss of tempo for him, forcing him to cast a weaker threat as well as increasing the chance your Bear will be able to attack back next turn. If you block with the Bear, he will not play any of these. Not blocking is inferior to both plays.

Piracy Charm

Will trade for the Bear if they block – not a problem. If the Shaman blocks then you happily two for one him. Not blocking again sucks. Some people have pointed out that he might have Piracy Charm and Viashino Bladescout. Although this is an unlikely combination that would result in him trading the Charm for the Shaman, this is again actually worth it because you have forced his play and left him with one card in hand and you with the active Bear.

Fire Whip

Interestingly, this card has not been factored in by anybody. Sure it might be timeshifted, but it is another explanation why the Goblin did not attack (although there are many common reasons also). If you block with the Bear, you lose it; block with the Shaman and you trade for the Seer. Not blocking sees you in the clear. However, as this card is effectively rare it is much harder to take into account, especially as its impact is at best still a one for one and at worst a dead Bear for information gained. It is also not a sole reason to attack – he is better off playing Fire Whip on the Seer and passing, then trading the Goblin for whichever creature attacks.

Subterranean Shambler

Several people have ruled out his holding this based on just the Seer attacking. So let's take a closer look. If he had attacked with both, it is likely you would block the Goblin with the Bear and taken one, then the Shambler takes out the Bear but he has to pay echo next turn and your Shaman will happily trade for it a turn later. If he had attacked with just the Goblin, you would have probably blocked with the Bear, leaving exactly the same scenario except you have one more life.

As it is, he has attacked with the Seer. If you bounce it off the Bear, then the Shambler will leave the board once more of 3/2 versus 2/3. If you block with Shaman, the board ends up as Bear and Shambler. Blocking with the Bear here is the better scenario, but all the eventualities are good as he will have a near-irrelevant 2/3 and have to pay echo next turn, leaving you plenty of time to turn the game around. Not blocking is again inferior to the other options, as you want him to cast the Shambler.

It is, however, unlikely that he has the Shambler, as the best play would probably be to send in both creatures. This way, you can either take two and he can play something else, or you can block both and be destroyed by the Shambler, or the board position looks like it would had you sent one or the other into the red zone. So unless he has a very specific reason why he wants the Shaman dead and the Seer to not be alive, we can safely rule out a Shambler here.

So far, none of these options would be good enough for your opponent to warrant attacking with the Seer. Thus, by themselves alone, we can for now rule them out. Now for the candidates that warrant a much closer examination.

Snapback and Grapheshot Snapback is the best reason to assume that this might be a bluff attack. Your opponent is just trying to see if you'll take one before he bounces and re-morphs the Seer. There is no reason to think this as he has no reason to attack here if this is the case. He can stay on defence and double block whichever of your creatures attack next turn – probably both will attack, and then probably he will block the Shaman and lose nothing if you have a combat trick to keep the Shaman alive. He would also look really stupid if your grip contained a Verdant Embrace or Undying Rage.

Grapeshot alone is nothing to be scared of. He needs to empty his hand to remove your board if you block. If you do not block with the Shaman he could have Dreamstalker to savagely ruin you, but this is unlikely as the risk of the Shaman blocking should deter this play. He might have Coal Stoker, and hence play a sizable enough threat to make this play worthwhile. It is unlikely that Grapeshot alone is his play, as most options reflect negatively for him.

This is why I have lumped Grapeshot and Snapback together. No matter how you block, he will be able to bounce, morph and unmorph the Seer, re-lay an Island, and Grapeshot your team to death. This suggests not blocking as the best course of action. Of course, if you do not block, he is free to make another play with his last remaining card, which has to be a good creature to deal with yours, otherwise the attack makes no sense.

Crookclaw Transmuter

This is one of the few cards where blocking with the Shaman seems bad. Look at the damage race if he doesn't have the Transmuter – he doesn't attack in his turn, you attack back with both of your guys, the Bear gets blocked by the Seer and he takes 3 – versus if the Transmuter is the only card in either of your hands – only the Bear attacks back and he only takes 2. He could block the Shaman with the Seer in your turn and trade it thanks to the Transmuter, but you could easily have a combat trick to save the Shaman (not to mention that he would be adding to the storm count for any potential Empty the Warrens or Grapeshot you might have). Both plays risk the Transmuter to any sorcery speed removal you might have or increase the incentive to play a Prodigal Pyromancer or Flowstone Channeler during your second main phase rather than a better creature.

If he needs to be rid of the Shaman because he has a powerful creature with three toughness in hand, then the attack makes sense. It could also be possible that he has a Grapeshot too, which would leave your board empty facing a Transmuter and a Goblin, but any removal in your hand or Penumbra Spider would neutralise the board again. Neither situation is too unrecoverable from your 16 life to warrant not blocking with the Shaman.

If you do not block, he can deal you an additional two, but will be facing taking five back in return. If you block with the Bear, he is highly unlikely to trade here, making the Bear the safer block as he fails to gain maximum usage out of his Transmuter.

Only blocking with the Shaman is good for him. Which begs the question: If he has the Transmuter, then why is he attacking? An easy answer is that the Shaman is the correct block in every other scenario so far, so he expects you to block here too. Otherwise, it would seem he must have another spell in addition to the Transmuter, even if it is just a blocker for the Shaman. Fire Whip would cover the Bear blocking too. Almost every other card is an acceptable trade for the Bear. If the his trick is Transmuter, then he must have something else to play. He need not play the Transmuter if something happens other than the Shaman blocking, so he is free to play whatever else he has in his hand.

Bogardan Rager

Here we arrive at what is most likely to be the worst-case scenario. When evaluating what the correct play will be it is often easier to work out what the play is if you start with the worst that can happen. Often there is nothing you can do about it and you have to shrug and move on. Let's see if this is the case with the Rager.

Some people dismissed the Rager as a poor card that most pros are unlikely to play. This is class-A drivel. Even pros have drafts that go wrong and are short a few cards. Aside from this, I think Bogardan Rager is a good card. I am tempted to say very good, but I might get slammed by the critics. Suffice to say, I am always happy to run one and am often found with two and sometimes even three in my decks. Anyhow, back to the conundrum.

If you block with the Shaman, you force him to play the Rager. Again, the Shaman trades with the Seer, and again, you find out what his trick is. It is unlikely you will be able to attack back into it, unless you are holding a Brute Force/Might of Krosa/Lightning Axe/Shivan Meteor/your own Rager, and he is likely to block in any case, so a bluff attack will not work. But you will know what you are dealing with and will have many options to deal with it, even something simple like making another 2/2 and double blocking if he attacks.

If you block with the Bear, he doesn't play the Rager because it is not worth it. If you do not block then the attacker is faced with the option of holding it back or nugging you for five. This is quite a large life swing, which will leave him with the dominant board position and two cards still in hand. Or he can choose not to play it and to pass the turn. You now have no information about what to do and probably cannot attack back as he can play the Rager, boosting the Goblin, and block either or both of your attackers. It seems that you cannot afford to not block, nor block with the Bear, so once more the Shaman is the correct block.

The question we need to ask ourselves is whether it is in our opponent's interest to attack with the Seer rather than just pass the turn with six mana up. This would probably mean either Rager, Transmuter, Cancel, Spell Burst, Dismal Failure, Draining Whelk, Teferi, Mage of Zhalfir, or Sulphurous Blast. Given these options, you will probably still attack with both creatures, as only Rager and Teferi are bad for you. Teferi you can do nothing about, but if you have a removal spell, you can use it to boot Teferi after combat, at the acceptable price of a two for one.

Against the Rager you can have Might of Krosa, Brute Force, Strength in Numbers, Lightning Axe, or a Rager of your own if you draw a land. This is quite a large spectrum, and if your opponent has put you on one of these cards, it might be a good enough reason to play the Rager in his turn rather than yours. One very good thing going for him playing the Rager in his turn is that it neutralises a lot of Strength in Numbers' power, because he will be safe to block if you attack back if you have it. The biggest attraction he has of playing the Rager in his turn, if you block with Shaman, is that it keeps the tempo in his favour – you will have to play something to deal with the Rager, which will probably tap you out, so that he is again free to enter combat on his terms. Thus, him having the Rager is a distinct possibility.

If the trick in our hand is one of the ones mentioned capable of dealing with the Rager, then not blocking is a fine option.

Consulting with the Pros

So where does all of this leave us, now that we have assessed all of our options in the dark, regardless of knowledge of our hand? Most people's first instinct, mine included, is to block with the Bear, as it is safe, prevents 1 damage, and might find the reason behind the attack. It probably won't though. Blocking with the Shaman, in every case, forces him to make use of his trick. There are only a few options he can have that end favourable with the Shaman block – Snapback and Grapeshot, Crookclaw Transmuter or Bogardan Rager.

As Jeroen Remie put it, "The Shaman is always a straight trade." This is very much true. No matter the cards in our opponent's hand, the end result will almost certainly be the death of the two combated creatures. There is one very important factor of note that I have left out till now – the Seer is probably worth more than the Shaman in value!

Mark Herberholz put it best when he said, "You want to risk trading a 3/2 for a 1/3 to take the Seer off the table... Why let him have a chance to draw Tolarian Sentinel, Vesuvan Shapeshifter, Snapback, Wipe away, or Dream Stalker?" The Seer has tremendous value in its potential reuse, not to mention its currently stellar blocking value. Both of these pros advocate a good enough incentive to block with the Shaman, as it is almost always correct to block like this.

Tiago Chan pointed out another very relevant aspect of the attack – it seems in every way that your opponent wants you to block with the Shaman. He knows what he is holding and it might be some bizarre combination, maybe even multiple rares, and he knows that you should block with the Shaman, yet he attacked anyway. Tiago also made a very apt comment, "I would block with the Bear, and accept a warning for slow play."

Tiago acknowledges that this scenario would send him deep into the tank. I was worried there wouldn't be enough meat for this article to be interesting, but now I find it hard to limit what I write. Imagine being at the table early into the first game of your match – this decision will take time! He chose to block with the Bear as he couldn't fathom the perfectly correct play, and this play asks at least some questions of the attacker rather than not blocking.

Olivier Ruel added his own sage wisdom into the mix. "Sometimes, when you can't find the answer, it's important to realise that your opponent knows why he did something that looks weird to you." In your average tournament, blocking with the Shaman is one hundred percent correct. But in a match against a high level pro, there are other things to take into account, which is why Oli opted not to block.

As it happened, the cards you had in your hand at the time of this decision were Rift Bolt, Herd Gnarr and Dragon Whelp. None of these were a good enough reason to stop the Shaman from blocking (the only good reason would be if one of the cards successfully dealt with Bogardan Rager).

For those interested, I misplayed the attack. I thought about it for a very long time and figured that is was very unlikely my opponent (the red-green mage) would block with the Shaman because the attack looked like I wanted him to, his hand was strong, and he was on a high life total. My hand was strong enough so that if he did block with the Shaman he was still in trouble, but my hand was so strong that even were he only to block with the Shaman 10% of the time, my attack was still wrong. Because I held a Dream Stalker. I had no reason to gamble on him trading for the Seer, as there was almost no way I could lose the game with those additional two cards that bouncing the Seer would give me.

As it happened, he blocked with the Bear as I had expected him too – after much deliberation. I Dream Stalkered the Seer, made and unmorphed it, re-laid an Island and cast a Fire Whip on the Goblin token to finish the Bear off. I had no need for the Crookclaw Transmuter I had also held in my grip.

The correct play for the turn would have been to Dream Stalker the Seer and say "Go." Assuming he drew nothing of great relevance, he would have laid the Dragon Whelp and passed. I would have untapped, played Transmuter to switch the Whelp's stats, and then made, used, and sacrificed the Fire Whip to finish the Whelp. The Transmuter would then get Rift Bolted, and I would probably go on to win off the two extra cards the bounced Fathom Seer would provide.

I hope this scenario has proven helpful. It's amazing how complex something so simple can become. Next week I will bring you the first of what will hopefully become a monthly draft walkthrough. Until then, never underestimate Limited.


  • Planeswalker Points
  • Facebook Twitter
  • Gatherer: The Magic Card Database
  • Forums: Connect with the Magic Community
  • Magic Locator