I don't know if I set a record last weekend, but it sure felt like it when I watched the recording this week.
For those that did not see the play in question, I am referring to a Brainstorm I cast on camera during game 3 of the semifinals at #SCGNJ last weekend. It was the last relevant play of an intense match, where a correct play could steal a game in which I was very far behind, having to fight through a Chalice of the Void with one counter, a Gaddock Teeg, and a Thalia, Guardian of Thraben with Storm. One slight error in sequencing could be the difference between winning and losing. While I came tantalizingly close to finding a winning line, my desperate Ad Nauseam from three life did not turn up a mana source soon enough to take the game and the match.
Given how interesting that play and the rest of the game that led up to it was, this week I will be recounting the game from my perspective at the time, as well as adding some insights I have gained in hindsight.
First, the players:
Me, playing this Storm list:
And my opponent, Tom Keating, playing the following Punishing Abzan deck:
Given Tom's large amount of permanent-based hate and relative lack of discard and/or countermagic, in sideboarding I wanted to cut as much of my own discard package as possible. The discard spells certainly facilitate a fast, easy kill by stripping Tom of his hate piece before he can even play it, but they are less reliable than simply having an answer to one that is on the battlefield. Since I did not fear Tom's ability to pressure me effectively, I preferred the more reliable solution and prepared myself to play a longer game if necessary.
To those ends, I made the following changes to my deck for game 3:
Tom's deck didn't have many creatures, so the Empty would be an easy way to take a quick game should I have a hand with a lot of fast mana, or provide additional threat density in case he took away my Past in Flames with Leyline of the Void, which occurred in game 2.
I added the Tropical Island without taking out a land because Tom's deck had a viable land destruction plan with Wasteland, Thalia, Guardian of Thraben, and Knight of the Reliquary. I have found that players tend to underrate the benefits of making three or four land drops with Storm. There is little fear of flooding when the deck wants to make a lot of mana on its combo turn, not to mention the presence of four Brainstorms to shuffle away excess lands. Moreover, having that many lands on the battlefield allows you to answer a hate piece and still go off that turn, which is excellent against a deck like Tom's that is likely to have a replacement piece of hate but has no way of interacting on its opponent's turn. Lastly, the extra land should help mitigate the effect of an early Thalia, allowing me to still cast the necessary draw spells to assemble a kill in a timely fashion.
As for Tom, I expected him to sideboard close to this:
The most questionable of those additions is certainly the Maelstrom Pulse, as it only serves as an answer to Empty the Warrens or artifacts I am unlikely to play before I need to because he only has two copies of Thoughtseize and it is certainly worse at the former than Golgari Charm, which is cheaper and can protect a Thalia or Canonist from Abrupt Decay. Without bringing in the Pulse I would expect Tom to leave in one of his Sylvan Libraries, which is a reasonable although risky card since the extreme life loss makes it easier for me to generate a lethal storm count through hate.
Now that we have that information, let's get to the game itself. If you would like to follow along, you can do so here:>
I am on the play, and I draw this opening hand:
Not exactly the hand I was looking for since the odds of a turn 2 kill are low. Still, this hand checks off a lot of the boxes I am looking for:
- A big spell to close the game
- A way to dig for missing pieces
- Not vulnerable to Wasteland
- Answer to potential hate cards
What it is missing is any form of fast mana, but that category is broad enough that Ponder is likely to find some. Another issue is the hand can be weak to a Chalice of the Void on one, but if I find Cabal Rituals and Lion's Eye Diamonds, it is certainly possible to win through a Chalice, or at least Ad Nauseam into an Abrupt Decay. Moreover, in game 1 when Tom was on the draw, he cast a Chalice on zero as his first play, hoping to stop a turn 2 kill and leave his mana to cast a Dark Confidant. Here he could do the same and also have the option of playing a Thalia or Canonist, which would be a strong opening. In that case, only a Dark Ritual would be needed for me to cast an Ad Nauseam on turn 3 with a healthy life total, almost certainly securing victory.
A major issue with this hand is its weakness to Thoughtseize, but Tom only has two of them and Pondering into more cantrips can help me recover fairly easily. After considering all these issues, I still think the hand is borderline, but I decided to keep and am fine with that decision in hindsight.
Fortunately, Tom mulligans twice and starts the game on five cards while also scrying to the bottom.
I start the game by sacrificing a Misty Rainforest for an Island and casting Ponder. Looking back I think I could have found an Underground Sea, since I already have a non-basic in my hand and I am likely to want to find Tropical Island or Volcanic Island to enable my off-color spells so it will be difficult to turn on my opponent's Wastelands.
Making a play like this--finding a non-basic to play around Wasteland--is counterintuitive and one that I feel many do not make when they should. Having all dual lands on the battlefield is a disaster if your opponent is flooded with Wastelands, but that is unlikely, and if they only have one or two it will leave your colors intact, whereas fetching a couple basics may not. You have to be able to fully commit to finding basics for that plan to work, and given my four color configuration after sideboard, I was unlikely to be able to do that, even if I had the capability of finding three basics with my draw.
However, given that my plan was to win on turn 3, I can find a Swamp with a Delta on my next turn and then be entirely insulated from Wasteland for the length of the game. This is appealing because my opponent is on five cards and is likely to Wasteland me early in desperation, hoping to catch me with a land-light hand to buy time to draw out of his early disadvantage. I also have enough total lands to play through a Wasteland or two even if I find a basic early, so in this instance I think finding the Island is worth it, but it is important to note that this situation is rather unique and the vast majority of the time I will find a dual land in that spot.
Moving on, my Ponder shows me a disappointing selection of Gitaxian Probe and two lands, so I quickly shuffle and draw Infernal Tutor before passing the turn. Infernal Tutor is not a great draw because I need mana, but it does provide insulation against Thoughtseize and can find a second copy of any ritual I draw to set up a turn 3 Ad Nauseam, so my hand is close to complete even though I can no longer sculpt my draws.
Tom merely plays a land and passes, which is a relief since these Life from the Loam decks are much less scary when they do not open on Mox Diamond. My draw step yields a Dark Ritual and now I can see the finish line. Unless Tom has a Chalice of the Void on his next turn, I will be able to cast my Ad Nauseam on turn 3 as planned. The decision here is whether or not to cast the Infernal Tutor main phase to find a second Dark Ritual.
Doing so seemingly gives me more mana on my Ad Nauseam turn, but if you consider my opponent's possible plays, I think holding the Tutor is correct.
- If Tom plays Chalice of the Void, I have another dead card in my hand.
- If Tom plays Thoughtseize, I no longer have a win condition.
- If Tom plays Thalia, Guardian of Thraben, I break even on mana. This one is a little tricky to see, so allow me to explain:
Say I hold the Tutor. Tom plays Thalia and I have two mana to cast a Chain of Vapor on his end step. On my turn I can play a third land, Infernal Tutor for a second Dark Ritual, and cast both Rituals into Ad Nauseam with no mana floating.
If I Tutor on my turn, I have to Chain of Vapor on my main phase, which costs the same two mana that Infernal Tutor cost in the previous scenario, and from there the situations are the same. Essentially both the Chain of Vapor and the Infernal Tutor will cost two mana, and it does not matter much when I cast them. The tie-breaker is that leaving a Dark Ritual in my deck makes my draw step a tiny bit better, emphasis on tiny.
If Tom plays an Ethersworn Canonist, a Gaddock Teeg, or draws Mox Diamond to enable a Green Sun's Zenith, then I will wish I had played the Infernal Tutor on my main phase since I can Chain of Vapor on turn 3 and still cast Ad Nauseam with a land untapped; however, this is the least likely scenario, and the gain is only one mana in a scenario where I am very likely to win regardless of the extra mana, although it may require passing the turn. I may even pass the turn and Ad Nauseam in response to the replayed hate bear, allowing me to keep every card going into my combo turn, cast another answer card, and go off comfortably on turn 4.
I pass the turn, hoping to see a creature from my opponent, but alas he has the Chalice of the Void. They never make it easy.
A Tendrils of Agony does nothing to help, and I am left to pass the turn. Tom plays a Green Sun's Zenith for Gaddock Teeg, which is not particularly threatening since I presumably have to find an Abrupt Decay for the Chalice, at which point I already have Chain of Vapor for the Teeg. However, it does start a clock while preventing a natural Ad Nauseam.
This becomes relevant immediately as I draw a Lotus Petal that would have allowed me to cast such an Ad Nauseam. At this point, my mind starts to fade a bit and I play the Lotus Petal thinking that I actually can cast Ad Nauseam, but I plan to do it on his turn. Upon realizing my mistake I silently curse myself for exposing the Petal to a removal spell, as well as potentially robbing myself of one storm later in the game.
The extra storm is important given that I have the Tendrils of Agony in hand and my opponent's hate means that I will be winning on a small margin.
Tom plays Dark Confidant on his next turn, which increases the clock and means it will not be long before he finds another piece of hate, which should seal the game. I need to draw something good on my next turn, and my deck delivers with a timely Abrupt Decay.
Unfortunately, I am one mana short of casting Decay on Chalice, Chain on Teeg, and then Dark Ritual into Ad Nauseam. Had I cast the Infernal Tutor on turn 2 to find a second Dark Ritual, I would have an Ad Nauseam from fifteen life here with a Lotus Petal on the battlefield and a land drop left to play, making me a heavy favorite. But it is dangerous to engage in this kind of hindsight, and I still believe that not casting the Tutor was best at the time.
Without a kill, I pass the turn hoping to dodge two draw steps from my opponent and set up an Ad Nauseam from eleven life, where I still think I am a favorite.
As Patrick Sullivan noted, Tom has cut himself off of several of his hate pieces, as his Chalice prohibits him from casting a Thoughtseize and the Gaddock Teeg stops any more Chalices from being cast, as well as a Leyline of the Void. So I just need to dodge Thalia and Ethersworn Canonist.
Spoiler Alert: I didn't. I am not sure if the Thalia came from Tom's draw step or he had it but wanted to play Dark Confidant first given that Chalice and Teeg were more than enough to buy another turn, but that is ultimately irrelevant unless I am looking for reasons to tilt.
With Thalia on the battlefield, I need to conserve mana, so casting the Abrupt Decay immediately is imperative, and also opens me to drawing cantrips to dig for another answer or more mana if I use the Infernal Tutor to find a second Chain of Vapor.
My draw step shows a Gitaxian Probe, which is not worth casting until the Thalia is safely away in Tom's hand. I cast Infernal Tutor for a second Chain and pass the turn hoping that my opponent is finally out of hate and I get a little lucky.
Tom has only a Knight of the Reliquary to threaten lethal, and I am finally able to cast my Ad Nauseam, now from a psychotically low life total of five. In order to set up my Ad Nauseam, I am willing to sacrifice my Lotus Petal to play the first Chain of Vapor on Tom's end step, essentially losing one mana on my next turn to save two mana.
Without a good draw step this still leaves me tapped out after casting Ad Nauseam, since I need to cast Chain of Vapor, leaving me with three lands and Dark Ritual for a total of five mana. Note that because I planned on tapping all of my lands, I found a Swamp with the last fetchland so that I could still draw blue, black, or red mana with a land drop if need be. This is a miniscule difference because I am likely to need Lotus Petal anyway, but I may reveal a string of Dark Rituals and Lion's Eye Diamonds with a land to complete a natural Tendrils.
I don't think I need to explain that casting an Ad Nauseam from five life with no available mana and three storm is unlikely to work, so I am hoping for a miracle. My draw was perhaps the most interesting card possible in Brainstorm. At this point we have the following gamestate:
Drawing Brainstorm is interesting because it gives the potential for not winning with Tendrils or Ad Nauseam. It would take quite a Brainstorm to do so, but I am already in a very difficult spot. It is also possible that I find enough mana to cast both Tendrils and Ad Nauseam, thus giving me much more life to work with and setting up a Past in Flames or Empty the Warrens to finish my opponent.
Also of note is that if I draw two artifacts off of the Probe and Brainstorm, I can cast them both, use Chain of Vapor to return them to my hand along with his Gaddock Teeg, and re-cast them before casting Dark Ritual and Tendrils of Agony. A quick storm count reveals that such a Tendrils would be for eighteen life (Probe, Brainstorm, Artifact, Artifact, Chain, Artifact, Artifact, Ritual, Tendrils), at which point Tom's Dark Confidant stands a good chance of killing him, especially with the extra time bought by the life gained from Tendrils.
Unfortunately, Tom had a Liliana of the Veil to kill his own Dark Confidant so I would only have had one turn to kill him with this line, and I suspected he had a Liliana even before casting Gitaxian Probe since he asked me multiple times how many cards I was holding on the previous turns. Still, one shot at drawing a non-Chalice spell are odds I would gladly take from that position.
Given the great potential offered by the Brainstorm, I decided to cast it after Probe to gain the most information. However, casting the Brainstorm first has a lot of merit if I draw a Lion's Eye Diamond. Consider the case where I draw Lion's Eye Diamond and two lands. I can make the following sequence:
2- Cast Lion's Eye Diamond
4- Cast Lion's Eye Diamond
7- Cast Tendrils of Agony for fourteen life (BB Floating)
This line is obviously short on storm, but there is three extra mana left over at the end as you have six mana floating with the Gitaxian Probe on the stack and an untapped land. So our Brainstorm has the potential of finding more castable cards to increase storm (say, Abrupt Decay or more mana cards).
2- Cast Lion's Eye Diamond
3- Cast Lion's Eye Diamond
5- Cast Lion's Eye Diamond
6- Cast Lion's Eye Diamond
8- Tap Tropical Island, cast Tendrils for sixteen life
10- Cast Ad Nauseam at nineteen life, knowing that your top card is a land you can play (B floating)
Note that this line also works if I draw a Cabal Ritual and a Lion's Eye Diamond, although it generates one fewer storm and two fewer life from the Tendrils. By casting the Gitaxian Probe first, we limit our ability to use one of the best possible draws in Lion's Eye Diamond. But we deny ourselves a card by casting Brainstorm first. Note that the third scenario can at worst mimic the first if I cast the Probe first, and if either of those Diamonds is a Lotus Petal, casting both Tendrils and Ad Nauseam is not possible.
This decision seems very close to me, even with the benefit of hindsight. Ultimately, I think casting Probe is the safer play, and seeing the most cards is quite valuable since the lines available at the beginning of the turn are so poor. Essentially this hand needs a lot of help, so you should dig as deep as possible.
So now we finally come to the Brainstorm. The Gitaxian Probe yielded a Cabal Ritual, an excellent draw, and the Brainstorm shows Infernal Tutor, Polluted Delta, and Abrupt Decay. At this point, my hand is the following:
My first instinct was to try to empty my hand for the Infernal Tutor, so I considered putting back the Tendrils of Agony and Ad Nauseam, then searching for Volcanic Island with Polluted Delta and attempting a combo. Unfortunately, that line is one mana short:
2- Cast Cabal Ritual (BBBBBB floating)
5- Cast Infernal Tutor (BBB floating)
At this point, even with the Volcanic Island, I cannot leave a mana floating after casting Past in Flames, and a Tendrils of Agony from that position is only for sixteen life (the five spells listed + Gitaxian Probe and Brainstorm yield seven extra copies). There is some chance of the Dark Confidant killing Tom from three life (Knight of the Reliquary, Leyline of the Void, Liliana of the Veil, and Maelstrom Pulse), but I felt like I could do better.
The other option is to use the Infernal Tutor to find a second Cabal Ritual, which nets both a mana and an extra storm. To do this most efficiently, I definitely want to put back Abrupt Decay since I only need one answer to the Gaddock Teeg and Chain costs one fewer mana.
The other card to put back has to be one of Polluted Delta, Ad Nauseam, or Tendrils of Agony. Of these the most obvious is Ad Nauseam so I considered it first. From there I play the Polluted Delta and have the following sequence:
4- Cast Cabal Ritual (BBBBBBBB floating)
6-Cast Tendrils of Agony for sixteen life
This is the same as before so I continue to look for a better line. If I put back the Tendrils of Agony, I can cast an Ad Nauseam with BBB floating and a Delta on the battlefield, but with only three life that is highly unlikely to work (I would essentially need to find a ritual and a Lion's Eye Diamond before almost any other spells), so the last card to consider putting back is the Delta.
With a similar sequence to the above, I get to the point where I have eight black mana in pool, a storm count of seven, and a hand of Tendrils of Agony and Ad Nauseam. From here I can cast the Tendrils for sixteen life as in the previous scenarios, or go for the immediate kill and cast Ad Nauseam from three life. With a storm count of eight, I have plenty of good draws. Any Lotus Petal or Ritual wins me the game, as does the combination of land and Lion's Eye Diamond. Moreover, hitting any playable land allows me to simply stop and Tendrils of Agony for eighteen life, forcing my opponent to survive a Dark Confidant trigger at one life, which is significantly harder than from three.
Of all the lines I went through, I think this one is clearly the best, and it unfortunately missed, revealing Abrupt Decay, Misty Rainforest, Cabal Ritual in that order. And with that the world's longest Brainstorm failed, but from that spot I believe I put myself in the best chance to win.
But was there an earlier point at which I misplayed?
After scouring the tape for the last few days, I believe the answer is yes.
Look back to all the lines I considered before drawing the Brainstorm. Because I was still clinging to the idea of casting Ad Nauseam, I favored conserving mana over maximizing my storm count. But if you look at many of those lines they had mana to spare, but were short on storm. Therefore, I think casting the first Chain of Vapor on Tom's end step was a mistake.
Think about what happens if I wait until my turn. I untap now with the same four lands and a Lotus Petal on the battlefield and this hand:
Even without a draw step, I have the following line available:
2- Cast Gitaxian Probe for two life
3- Replay Lotus Petal
5- Replay Lotus Petal
If either my draw step or the Gitaxian Probe yield another artifact, then I can play it twice by sacrificing another land with the second Chain of Vapor, thus making the Tendrils for eighteen life a much better spot than I was in. Any Ritual gives me a Tendrils for sixteen, which was about the baseline for the previous set of lines. And there are various runner-runner outs that flat out win the game. (Lion's Eye Diamond and Infernal Tutor, for example.)
Waiting on the Chain of Vapor is essentially saying that Ad Nauseam is a dead card, and my best chance of winning is through a natural Tendrils. I think the information I had at the time supports that conclusion, but I had been tunnel vision'ed on Ad Nauseam from the first turn of the game and could not get off of it.
Tunnel vision is one of my primary failings as a Magic player, as I often fail to realize the implications of my draw step and proceed with a previously planned but inferior line. In fact I lost a game to Elves in the quarterfinals in much the same manner, drawing a Dark Petition in the final turn of game 1 but failing to move myself off a line that ended with Infernal Tutor to a line that simply used Infernal to find another Ritual to set up a Dark Petition for Past in Flames.
Concluding that holding the Chain of Vapor was correct should ultimately be enough to end my analysis of this game, but because like most Magic players, I have masochistic tendencies, so I had to see if it would have made a difference given my exact draws. Fortunately, that is not possible given that I sacrificed a Polluted Delta to cast the end step Chain of Vapor so I will never know.
But why would I let something like that stop me? Say I decided to crack the Delta just to thin my deck, which is reasonable on the last turn of the game when the life will not matter since I am not casting Ad Nauseam. If my draws were exactly as they were on the final turn, I had a winning line by holding the Chain of Vapor. Starting from my main phase:
2- Cast Gitaxian Probe for two life
4- Cast Lotus Petal
6- Cast Lotus Petal
8- Cast Cabal Ritual (BBBBBB floating)
10- Cast Tendrils of Agony for twenty life
Looking back over the game, I noticed that the only reason I had the option of casting an end step Chain of Vapor is because I mistakenly cast the Lotus Petal on turn 4. Had I not done that, I would have drawn Abrupt Decay on turn 5 and passed the turn with seven cards in my hand, then cast an end step Decay, and upon casting Infernal Tutor for the second Chain of Vapor, had been left with only one mana.
Having the Lotus Petal in my hand would have given me the option of gaining an extra storm at the cost of one mana by playing the Lotus Petal before the first Chain of Vapor on Thalia, although I doubt I would have, given how tight on mana those lines were.
Still, it is interesting to see that a seemingly innocuous mistake on turn 4 rippled its way through the game by allowing me to take a subpar line on turn 6 and leaving me in a worse position to combo on turn 7.
Normally I would be upset with myself for missing the best line and mope for a day or so before moving on to the next tournament, but this game was so intricate that missing the line and going to back to find it was exciting. It is games like this that give me hope that I will never get bored of Magic, even if my excitement level, and thus effort level, wane for brief periods of time. This is the kind of game that makes me realize why Magic is so popular and rediscover why it is perhaps the greatest game ever made.
And all it took was an unreasonably long Brainstorm.