Have you ever had the feeling that there is a powerful deck out there, but people don’t yet believe how potent it really is? I bet you have, because that is how almost every tier one decks starts out when it is not built around hype cards. For example, when Birthing Pod was spoiled everybody knew there would be something out there, but it took a while until anybody took the Illusions deck that later evolved into UW Delver seriously. The Legacy Omniscience deck might be such a deck on its way to tier one. Of course it might just as well not be, but it’s the most promising candidate I have seen up close in 18 years of playing Magic.
Origins of the Deck
When I started playing Legacy at the start of the year, RUG Delver appealed to me. I like Delver in Standard and RUG Delver seemed a fine choice in Legacy as well, especially since we have a a player in town who is extremely proficient with the deck. Playing these highly interactive decks also helps to get your game to the next level. RUG is fairly powerful and I had some success with it right from the start. I always imagined I would be playing that deck at GP Ghent, but it didn’t come to that. Helvaults were shattered, Demons were freed, and Omniscience was unleashed upon the world.
Sneak and Show on the other hand was never a deck I considered. S&S felt like the Belcher-style, cheesy deck that players bring to a tournament in the hopes of evading any meaningful interaction. When Avacyn Restored added Griselbrand to the equation, the picture changed. Instead of playing four [card emrakul, the aeons torn]Emrakuls[/card] and four [card progenitus]10-point burn spells[/card], you would get to play eight real threats now. As we started testing for GP Ghent, I imagined that between reanimating Griselbrand and putting him into play via Show and Tell or Sneak Attack, S&S would be the superior strategy, as it is not as susceptible to hate.
Reanimator is more powerful pre-board because it has a greater threat diversity and more consistency, but post-board you rely on the other player being unprepared. We might actually not have given Reanimator enough credit for its ability to overcome hate, but this was our first assessment.
Some playtesting showed that S&S was indeed very powerful, and we figured that the deck is just underappreciated because everybody agrees that it is so dumb. Few self-respecting players play Belcher either. A couple of weeks before Ghent, I brought my S&S version to our regional, monthly tournament in Dortmund. It was a bit hard on me to rub a turn one [card emrakul, the aeons torn]Emrakul[/card] into the face of my opponent as the first thing to do in the morning, but I believe it might have been worse for him. I mean do you really get up early on a Sunday, drive a hundred miles, and pay a dozen bucks to have your opponent put an Emrakul into play before you even made your first land? Anyway, for three rounds everything went according to plan.
After a draw to Reanimator, things fell apart. I lost games I shouldn’t have. For example, I lost to control after putting Griselbrand into play and drawing seven cards. Twice. The deck is actually not as reliable as it looks on paper. It is also more fragile than we had assumed. Heavy countermagic is tough and cards like Blazing Archon or Ensnaring Bridge are basically unbeatable. Bringing this to Ghent was not an option.
A few days before the Dortmund tournament M13 had been spoiled, and Carsten Kötter had already come up with an Omniscience deck. When I saw it, I ridiculed the concept—after all why would you want to cheat a card into play that doesn’t even win? Simply drop Emrakul and get the job done! The deck idea was in my head, but going for the Show and Tell, Omniscience, wish for Petals of Insight kill seemed suboptimal. Three-card combos don’t cut it.
Maybe there was something better out there? Couldn’t you for example play a bunch of Whispers of the Muse? You would cycle them when not needed and draw your deck when you cheated Omniscience into play. I checked. “You can pay additional costs such as kicker costs.” Fine. In my opinion the best thing about an Omniscience-powered deck was that you could kill immediately, allowing you to load up on Pact of Negation. There were a lot of options to consider, though. Would we try mono-blue? Or add red for Burning Wish as in Kötter’s version? Or stay blue but add clunky Cunning Wishes? I discussed the matter with Simon Görtzen who is a good friend of mine and lives in the same city. Simon suggested Personal Tutor and wanted to get rid of the Whispers. He always wants to play the most powerful cards, and Whispers is not a good card on its own.
We tested some builds and the results were encouraging, but after a couple of days Simon became suspicious of the combo: Might it be possible that it doesn’t actually work?
I beg you pardon, it does not?
I guess not—Oops!
Ghent was about a week away at that point, and our deck had just evaporated. That was a bit irritating as you might imagine. For a short time we were unsure how to proceed, but Simon quickly came up with a new list that had Kötter’s kill as well as a set of Griselbrands. He also had some ideas about playing Dream Halls, but I find the card horrible and a bit of goldfishing on his part convinced him that I was right in my assessment. Most of the remaining days were spent on determining the counter package and the tutors we wanted to have.
Simon had changed his mind on Personal Tutor, while I advocated that it be played once or twice. We also shaved Pact of Negation from the deck, as it is useless when you have to protect Show and Tell into Griselbrand. The deck turned out to be so explosive that Daze was a hard counter to protect the combo most of the time. At other times Daze is just a nice way to interact with your opponent without burning your resources, unlike Force of Will.
While I had done a lot of brainwork on the first versions of the deck, most of the final polish came from Simon. When he gets rolling the best thing you can do is try to keep his momentum up by giving him some additional perspective on his ideas. I don’t know how, but in the end he came up with the idea that we should include two Jace, the Mind Sculptors. On that one I was a bit skeptical, because I didn’t see the connection to the rest of the deck, but I supported the idea on the basis of David Ochoa’s GP Atlanta report that was basically a rave about the potency of Jace TMS in Reanimator. We finally registered the following list for GP Ghent:
Ghent and the Aftermath
Simon, I, and another friend from Aachen, took the deck to Ghent and started out really well. The Omniscience deck felt powerful and took people by surprise. I don’t want to go into too much detail, but Simon finished Day One at 7-2, I was at 8-1, and Julian had lost the last round for Day Two, finishing 6-3. Day Two was rough for us as we both only managed a 3-4. The result was especially disappointing as we both felt we had played very inconsistently during the tournament, making good plays as well as poor plays.
To give you a brief impression, I played Belcher in round 6 and somehow took away a close game one. In game two I killed his first wave of Goblins with Pyroclasm. I then had Force of Will plus double Daze in hand, and the combo assembled. Can you imagine I lost that game and the match?
Well, on his last turn he plays Lotus Petal, which I let resolve and then another Lotus Petal. He has a Simian Spirit Guide and an unkown card left, and passes the turn. While going into my turn it hits me: What if he actually drew Goblin Charbelcher? He played Land Grant earlier, and there are just two draw steps over which I am not sure what he drew, but what if? In what amounts to a panic reaction I go for the Show and Tell anyway. The chances of him having drawn Belcher are not that high after all, but of course he has the Belcher. I even had a Jace in hand which I could have cast instead. If I had stayed calm, I might have figured that his line of play makes only sense if he actually has the Belcher, but I didn’t get it.
On the other hand, in the following round I played against something like Esper Stoneblade with a bunch of discard. We were tied in games and I was on the play in the third, but had to mulligan. My hand is Ponder, Jace, the Mind Sculptor, Volcanic Island, Flooded Strand, City of Solitude, and Ancient Tomb. This is not what you hope for either, but Jace is quite good in this matchup and going to five probably won’t help. I start with land, Ponder, and see Lotus Petal, land, Griselbrand. What would you do with this? Shuffle? Griselbrand is one part of the combo, but you know that you won’t find the second part for a substantial time. I thought about it for a bit and decided that I would immediately draw Griselbrand and put Lotus Petal on top. If my opponent would Thoughtseize me then, he would have to fear that I actually hid Show and Tell. Thus he would have to take the Griselbrand away. I would then draw Petal, drop a turn-two Jace and hopefully he would not have a Force. That went according to script and when my opponent killed my Jace with his Jace on turn four I had already pulled so far ahead that the game was out of reach for him.
These are just two examples on my part, but there were more—according to Simon’s reports he had similar ups and downs. At the end of the weekend, finishing just outside the money was a real bummer, as I knew the deck could have gone 9-0 on Day One, and who knows what would have happened on Day Two without screwing up there.
After Ghent Simon was approached by Travis Allen, who had read about Omniscience in the coverage and Simon gladly briefed him on our deck. Travis took the deck to SCG Open Buffalo and promptly rode it to the quarter-finals. Afterwards there was a week of tranquility. At some point Cedric Phillips apparently started to play the deck on MTGO, streamed that, and suggested a couple of modifications. Unaware of that, I took Omniscience to the monthly Legacy tournament in Dortmund. By now people knew what the deck was about, but the deck would not be stopped that day. I did not drop a single game, finishing a perfect 7-0 with 14-0 in games. Meanwhile, people had indeed taken notice. While I would have loved to take Omniscience to the Legacy MOCS I was not qualified. Watching the WMC coverage I had all but forgotten about the MOCS, too. It took me until Monday night to figure out that our deck had delivered for real this time. Congratulations to jacksad for winning the MOCS with Omniscience!
Notes on a Few Card Choices
Yesterday, Matt Nass published an article on combo decks on this very website, and he has some valid points regarding our card choices. Most cards in the Omniscience deck are obvious inclusions, but a few cards are debatable.
At first I couldn’t think of any card choices to discuss—everything fell into place so well—but I would like to give you a different perspective on the points Matt raised. When discussing individual choices you should keep in mind what makes this deck tick. The main strength of Omniscience over regular Sneak and Show is that Omniscience is way more resilient to hate. Humility or Peacekeeper are Game Over for Sneak and Show and several other cards are seriously annoying, but there is just not one card that hates Omniscience effectively.
When you consider moving all-in on the [card petals of insight]Petals[/card]/Grapeshot kill, think about what you are doing there. You are killing what makes this deck strong—its resilience. Suddenly, Thalia means a world of endless pain for you. Originally we had considered playing Sparkcaster to make the deck even more resilient, but that proved unnecessary. Usually you beat Thalia by casting Emrakul with Omniscience. The addition of Jace also came later and added another angle of attack, further obsoleting Sparkcaster.
Daze also increases this deck’s consistency and resilience to hate. This time you are not evading the hate that is aimed at you, but the generic attacks on mana bases that are a part of many strategies in Legacy. Adding a third color to remove Daze makes Omniscience more susceptible to Wasteland and Blood Moon. It might be worth it, but in my opinion this decision is not a no-brainer.
If you decide that not having a third color is for you, then including Daze is the logical choice. The deck needs more protection and disruption than just Force of Will. Pact of Negation is obviously not an option, and including multiple Misdirections is horrible. Mostly you want to have a Force of Will to protect your combo, and Misdirection doubles as that, but consider that almost any hand with mulitple pitch counters would just be better if you replaced one of the pitch counters with a random blue card. Pitch counters are such a drain on your resources that you can almost never afford to cast two.
But Daze is not only the logical choice, it also fits your game plan very well, as it allows you to run through your cantrips and not be naked at the same time. It allows you to protect your combo in the early turns when you could not even have afforded the mana to cast a Thoughtseize. Also the mere presence of a super-explosive combo deck puts pressure on your opponent. He might be dead any turn and often he cannot afford to wait on casting his spells. Last but not least, Daze is a blue card. Thougtseize does not pitch to Force of Will. All things considered, including Thoughtseize has a lot of appeal, but for me Daze has always pulled its weight in this deck. The one thing I am not sure of is whether you want three or four Daze, as drawing multiple Daze is rarely a good thing.
Playing the Deck
Omniscience is fun to play, as there is some variety in how you actually close the game and you never know in advance what will happen. If you play Magic for a whole day you will have difficult decisions to make, and Omniscience can’t take that away from you, but in general, it is one of the easier decks to play. Unlike other combo decks, there is little calculating to do, as you either have the goods or you don’t. The best advice that I could give you is that Omniscience doesn’t need to take a lot of mulligans, but when your hand is not good, don’t shy away.
The deck is very consistent due to a million cantrips and only needs a few specific cards to win. Consequently, the deck mulligans rather well and keeping one of the few bad hands is just not worth it. In-game, the general principles of playing combo decks apply. Assess when your opponent might be able to kill you and go off before that. When there is no imminent danger, consider whether prolonging the game is more likely to yield you additional protection or your opponent more disruption. Most of the time when you have burned all your cantrips and the combo is assembled you should assume that your chances will not increase further.
Before we take a look at the matchups, I would like to present an updated list:
|
Some people have cut the alternative win condition that Living Wish + [card emrakul, the aeons torn]Emrakul[/card] provided, but I am not convinced that you want to do that, as it greatly reduces your resilience to hate cards. On the other hand, I get that people want to have more sideboard slots.
It just struck me that it is not necessary to go down the Wish route. You can put a Time of Need into the sideboard and find your [card emrakul, the aeons torn]Emrakul[/card] or Griselbrand from the main deck. You probably don’t want Karakas any more if you cut Living Wish, but it never felt very important to me, and going for infinite turns via Emrakul + Karakas is usually overdoing it.
Regarding the inclusion of Massacre, I like the idea, but I don’t quite get why you would want to have it. A singleton might be nice to conveniently get rid of [card thalia, guardian of thraben]Thalia[/card], but jacksad had three. Why? There are no creature decks besides Maverick that have Plains. If Thalia is such a big deal, wouldn’t it be better to have a singleton Tremor in the board? It’s one mana more expensive, but Massacre adds a color and is even more narrow. If they see it coming, the Maverick player can also occasionally play around it by never making a Plains. On the other hand, Tremor is better against Belcher as you can Wish for and play Tremor on turn two with a single Lotus Petal.
Matchups
I have played about a 100 games with the deck, but in Legacy few decks ever have a metagame share of more than 10%, meaning that I have not played many games against the same deck. I will thus give you some general advice on the five decks I think are most common, and round out the article with a few musings on what might be intersting to try out.
RUG Delver
As it happens with matchups where RUG Delver is involved, they are always very interactive. You are a clear favorite here, but you have got to pay attention to what they are doing. Sometimes they are a goldfish, but at other times they have an early Delver and a bunch of countermagic. Don’t go too low on life as you have to assume that they have a pair of [card lightning bolt]Bolts[/card] in hand, but otherwise take your time. Their Daze and Spell Pierce get worse as your mana develops.
Sideboarding:
-2 Jace, the Mind Sculptor
-2 Daze OR -1 Daze, -1 Personal Tutor if they have Surgical Extraction
They interact via counters, thus it is just logical to bring as many Blasts as possible. Shooting an early Delver also gives you time, which is nice against their soft counters. If they don’t do a lot, again just try to wait out their Dazes and Spell Pierces. You should eventually be able to fight through their Forces and Blasts. After all, you have 50% more library manipulation than they do, and should thus be able to shape your hand more efficiently. Finally, pay attention to whether they have Surgical Extraction. You don’t have to be paranoid, but if you can avoid being caught by Extraction, don’t give away free games.
Maverick
They have almost only proactive hate, so you can just watch what they do and go off when they threaten lethal damage. As they have no meaningful ways to improve their clock spontaneoulsy, this is just a matter of calculating the boardstate. The only thing that you have to consider is what they might put into play via Show and Tell. [card thalia, guardian of thraben]Thalia[/card] is most likely, but Quasali Pridemage or Karakas might come into play as well. If you can go Omniscience into [card emrakul, the aeons torn]Emrakul[/card] they are just dead, but sometimes you don’t have the nuts and they have the perfect answer. It sucks, but it is not that likely. Move on and crush them in the post-board games.
Sideboarding:
+1 Pyroclasm
+1 Pithing Needle
The worst that can happen post-board is that you try to go off with Omniscience and they have Krosan Grip while you try to Wish for something. But even if they had all the answers, they still have to win a rock-paper-scissors game against you when you are Show and Telling, and if the don’t win they are dust. You can also counter their stuff at leisure, giving you time and additional ways to dodge their hate.
Stoneblade
This matchup is probably the hardest to describe, as their builds vary considerably, and they attack from different angles. Discard is annoying, but fortunately they cannot afford to slowroll that. Otherwise you would have to be very conservative with your Brainstorms. In Jace and CounterTop they have ways of closing the game, and if you go off blindly you risk running into Force of Will. It all sounds horrible, but you are still a favorite here. Their problem is that they have a couple of good cards, but they can’t kill you on the spot. What will happen if they run out a Jace? You might just Daze it and/or use the window to blow them out and that goes for most of their threats. Even when they know your hand due to Thoughtseize, it is hard to assess what you will be able to do due to all the cantrips.
Sideboarding:
+1 Pithing Needle
+4 Red Elemental Blast
-4 Daze
-1 Personal Tutor
Daze is not that good as you have to plan for a long game. Post-board, one of their more likely routes to victory is getting one of your Show and Tells into the graveyard and then Extracting that. You might want to be a bit more conservative with your Burning Wish to have the fourth Show and Tell as backup. Playing a longer game also means that the chances of a topdecked discard spell are higher. Hence slowrolling Brainstorm makes more sense post-board.
Show and Tell Combo
You basically play against an outdated version of your own deck. Sneak Attack is more expensive than your combo and they cannot go for Show and Tell as they can’t kill you right away and might not live to see another turn. If they play Show and Tell, putting Griselbrand into play is the best option. If they have Griselbrand nothing happens, but if they don’t you draw 14 cards and win. [card emrakul, the aeons torn]Emrakul[/card] is not as good, but still pretty good because if they have Griselbrand and you have [card emrakul, the aeons torn]Emrakul[/card], they will have lots of cards, but they lose their board if they don’t draw all their Lotus Petals. Omniscience is the worst, as you will probably lose to their 14 extra cards if they have Griselbrand. Omniscience still beats Emrakul most of the time, though. On the other hand if you want to go off, you have to be able to kill them immediately as the same principles apply. You don’t want to pass the turn, just to be attacked by Emrakul.
Sideboarding:
+3 Red Elemental Blast
+1 Pithing Needle
-4 Daze
This is fairly awkward as you don’t know what they will do. They cannot realistically win via Show and Tell, but will they board Show and Tell out? If they do, they are cold to Pithing Needle. Also they cannot expect to protect Sneak Attack with multiple Pyroblasts, as this costs more red mana than they can ever hope to assemble. Try to figure out what they are up to, but in the end they will probably just stick to their original plan as they don’t have have a lot of options.
UW Miracle Control
Besides the Forces they have a bunch of actual hard counters. That means that unlike Delver they might benefit from the game going longer as more mana means more Counterspells to them. If they have Sensei’s Divining Top, it should be even more obvious why you don’t want to drag this out longer than necessary. On the other hand, they have just four Force of Wills to protect themselves in the early game. Thus, if you can go off early your best bet is probably to just do that. Emrakul is a bit lame as they might have Karakas or Brainstorm into Terminus, but if you get Griselbrand into play, drawing some extra cards will help to win even if you lose him.
Sideboarding:
+4 Red Elemental Blast
+1 Pithing Needle
-4 Daze
-1 Emrakul, the Aeons Torn
You are much better equipped to play a long game post-board. Their counterspells come online gradually, but your Red Elemental Blasts do as well. As in the Stoneblade matchup, they can never afford to tap out, as they are almost assured of being blown out right away. You can just run through your cantrips freely, though. The Pithing Needle is a miser’s choice to hate their Sensei’s Divining Top. Without that they are pretty far behind and you can afford to cut a fattie as you have enough time to find one anyway.
Cards to Consider
We had a hard time assessing whether Personal Tutor is good in this deck. In the end we played one and it was okay, while never being an all star. If you want to fiddle with the numbers, cutting or adding Personal Tutor is a natural point to start. I’m not sure what I would want to have in the deck if I cut the Tutor, though. On the other hand if you cut a Preordain for a second Personal Tutor, you could run silver bullets in the board even more effectively. If you have pairs of sorceries in the board you can bring one in, and have an actual copy in the deck plus four Burning Wish and two Personal Tutors. Besides Pyroclasm, I have not found a card I liked that much, but I restricted my search to blue and red cards. If you find something really nasty, think about including another Personal Tutor.
If you want to cut the Personal Tutor, a card to consider is Ulamog, the Infinite Gyre. Instead of having 8 Show and Tells + tutors and 8 targets, you go to 7/9—which is obviously worse as it reduces your consistency. The upside is that if you have Ulamog ready at your command, you make sure that they can’t hide. For example, in Ghent, I had to win a game against Ensnaring Bridge and Thorn of Amethyst. Jace did it, but the game was very close. If I had had an Ulamog that I could even have tutored with Burning Wish for Time of Need for Ulamog, things might have been way easier. If you don’t want to add Ulamog to the deck, but want still want to have a catchall you can of course put a gruesomely slow Vindicate into the sideboard. Should I decide to have the answer, I would prefer to have Obliterate or even Decree of Annihilation in the board.
A few other cards have popped up as sideboard candidates as well. I already discussed the merits of Tremor. The card obviously looks ridiculous, but it might just do the job. People have suggested Sparkcaster, as it provides an immediate kill that is resilient to Thalia. Including him only makes sense if you want to stay on the Living Wish route, though. Then you have to use up six of your sideboard slots, limiting your actual sideboard options considerably. Another card that always gets added to the equation is Overmaster, as you can Wish for it conveniently. Right now I don’t know where I would want to do that, but I imagine that it makes sense when you face a lot of Red Elemental Blasts. You would then wish for Overmaster, cast Overmaster, and resolve an uncounterable Show and Tell, thus blanking their Blasts.
Finally you might consider adding another color to the deck. So far I don’t see the necessity, but hey, it worked for jacksad. Adding green has the bonus of enabling you to cast Time of Need, which is a nice bonus when you are Wishflooded in a long game. Unfortunately, we didn’t find any juicy green sideboard cards, but we didn’t check out all the 2000 green cards either. If you want to go black,jacksad’s list should suffice as inspiration. Finally, as long as Balance is indisposed and there is no sorcery named Silence, I don’t see why you would want to add white to the deck. The only good wish target is Morningtide, and white’s sideboard cards are generally not the most powerful. Maybe, just maybe you could want to Personal Tutor for Terminus, but that’s really the best I can come up with.
-Florian
