Tezzeret control has been in the metagame for over a year now. Its consistent performance in the face of shifting competitors and restrictions is reminiscent of other great combo-control decks. Although there are a number of competitive strategies, I don’t believe there is a better deck you can take to a diverse metagame at this point in time.
While I’ve commented in a number of related threads, I’ve never put together my opinions on this archetype in a comprehensive way.
Please note that I’m just putting together my judgments here. I’m not claiming originality, as much of the subsequent discussion will show that most of my findings are empirical. I looked at what good players and deck builders did and incorporated it.
In that spirit, I’d like to acknowledge the key decklists that served as the basis for the list:
Luis Scott Vargas list from U.S. Nationals Side Event
http://www.morphling.de/printview.php?c=1104&d=1Jimmy McCarthy and Mike Solymossy lists from Chicago
http://www.morphling.de/top8decks.php?id=1095In a lot of ways there were other lists that were equally influential, if only to show what didn’t work. While this may seem like a backhanded compliment, these contributions are critical. For example it took seeing an optimized impulse list (like Owen Turtenwald’s from Chicago) to see why Dark Confidant was superior for my list. As I go through each card I’ll try to explain how these evaluations occurred.
The reason why I chose the title I did is because as the core of blue based combo control becomes more and more dependent on the restricted list, it becomes more and more about cards that search or tutor (as opposed to draw). The consequence of this is that single cards instead of 2, 3 and 4 ofs are more desirable and are more impactful. Magic is already a complicated game and trying to make sense of cards and how they’ll play is the most important skill in making these choices. The list below is the byproduct of careful observations about these emergent characteristics.
The breakthrough I had in constructing this list came after piling over list after list of other Tezzeret variants, lots of playtesting, and finally, some tournament experience at the GenCon prelim. The last piece was finding a sideboard configuration that was fully integrated with the abilities of the maindeck. Thankfully this discovery occurred at 5am the morning before the Type 1 Championship. That I misplayed my way out of the event and that my teammate came in 9th on breakers was certainly disappointing. However, seeing Hiromichi’s list and especially the performance of the archetype since is more than enough vindication.
Here’s how the deck came together:
1) Testing and observation of T8’s led to an assumption that optimizing time vault was the strongest strategy
2) With time vault as the pivot of a list, the next question is “how aggressive should the list be?” Our team had a transmute and grim monolith based list that was blazing fast and rolled most of the conventional Tezzeret lists. The problem was that it was much weaker against two common cards: null rod and mystic remora. The realization that turned our thinking was that time vault, the most broken combo in magic history actually allowed you to slow down and take a more controlling role.
3) Now with the idea of being more control and less combo, the priority was on rounding out the key components of a control list:
- draw engine
- answer suite
- mana base
Draw EngineI’ve long favored playing intuition/AK. When I saw Dark Confidant lists everywhere, it seemed a natural answer to have a more powerful draw engine and a tutor that found answers to the metagame’s most popular draw, Dark Confidant (darkblast) and win condition, time vault (ancient grudge). It didn’t work like I thought. The four color mana base gave the deck too much exposure against fish and stax. To fit the whole draw engine you also needed to cut on other broken cards or mana (which exacerbated the first problem). Not only this, it didn’t even succeed at what it was supposed to: it wasn’t better in the control matchup. With duress and the tempo gained from having a subsequently free draw engine the Confidant lists could actually go toe-to-toe with intuition. Trying to intuition for darkblast was a loser both in terms of card advantage and tempo.
The next idea was to try out the singular draw engine (restricted list, skeletal scrying, jace, etc). There was no unity to the deck. The cards did their job, but you weren’t able to plan for them and they weren’t powerful to stand on their own. Additionally, like intuition, they didn’t buy you any EV for non-control, non-combo matchups.
Impulse was the last thing I tried before Dark Confidant. It was pretty good. In some future metagame, I can certainly see using this as the supplement for the restricted list in a control shell. However, it wasn’t broken. Casting it didn’t turn the tide of games. Worse, sometimes you’d go into a black hole of impulsing for impulses and not find the cards you want or gain any card advantage. It was a reasonably solid card, but it didn’t make the deck better.
I usually don’t like Dark Confidant. It’s slow. It’s vulnerable. As a midgame topdeck it reads: “B1, next turn draw the top card of your deck”. I was making two mistakes. First, I was used to looking at it through the lens of the previous metagame. In a TFK world, there are simply better options and the control mirrors are faster. Second, I wasn’t looking at the card slot savings that this engine offered. What made it click was looking at the expected GenCon meta, looking at the 75 cards available and seeing how the sideboard plan worked out.
Actually playing with the card led me to understand much better how much is actually going on with the card. First, it automatically contributes a ~15% boost to your workshop and fish matchups. Second, if dropped early enough it’s plenty strong in the control and combo matchups. Third, in those matchups you’re not committed to more mana for your draw engine; there are tremendous long term tempo gains. Lastly, it’s a win condition.
This last item can’t be undervalued. Life is a precious resource in the mirror and for most ritual lists. It also hedges against stuff like Jester’s Cap/Sadistic Sacrament. For the recent Philly Open event we were all ready to roll out a transformational sideboard that used Sacrament against Oath, Tez and Ritual decks. The weird thing was, Sacrament, while great against Oath, didn’t do anything against Tez. As a Tez player you could lose your robot, time vault and Yawgmoth’s Will and use the card and tempo advantage the other player sacrificed to sacrament to simply out-advantage them and beat down with Confidant.
The sum of these things make for a surprisingly powerful card.
Answer SuiteThe printing of spell pierce complicates this section. However, at the time, it became obvious that duress and thoughtseize were the best option to supplement mana drain and force of will. The reasons are uncomplicated. They’re cheap, on-color, hit most of the crucial spells in the game and allow you to dictate the role. Relatedly, in a large, mostly unknown metagame like GenCon (or the Philly Open), these allow you to much better steer your future plays. There were contenders that didn't work out. Spell Snare while good, was just too narrow and hits neither Ancestral Recall nor Yawgmoth's Will. Counterbalance was too slow and inconsistent. Trinket Mage complicated the overall design and required playing weaker cards. Misdirection wasn't sustainable in multiples. Negate, etc didn't fit the curve. All of them are decent choices, but none as strong as duress/thoughtseize.
Filling out the rest of the answer suite brings us to a point where we can start talking about the list in total, slot by slot.
So far we have:
23 mana – looking at all of the Tezzeret variants from the restriction to GenCon there’s an obvious pattern that between 23 and 25 mana is necessary to make the deck work. I’m just starting, so we err on the low side.
3 Dark Confidant – running two makes it inconsistent enough to be pointless, four may be too many as they’re not particularly good in multiples and they’re lousy topdecks. This seems to be the sweet spot.
4 Force of Will
4 Mana Drain – I actually started with only three mana drains. It became more and more obvious that the format is slow enough so that this card remains very powerful (this is a pre-Zendikar evaluation, hold on for more)
2 Duress
1 Thoughtseize – I could see reversing these counts. I’m risk averse and didn’t want to be in a late-game situation where Confidant has brought me to a point where I can’t use these.
Let’s talk about the
restricted list. There are a few cards that will go into every combo-control list I ever play:
1 Ancestral Recall
1 Time Walk
1 Brainstorm
1 Merchant Scroll
1 Gifts Ungiven
1 Demonic Tutor
1 Vampiric Tutor
1 Yawgmoth’s Will
If you really have issues with any of these cards, please rethink your interest in vintage.
However, I’m sure there are cards that people would expect on that list that are missing. Let’s procede:
Thirst For Knowledge – this is a very powerful card, but at three mana and sometimes no card advantage, it’s certainly up for consideration versus other slots. It ends up making the cut, but mostly because of its strength in the midgame when you’ve drawn chaff and because it makes Merchant Scroll and Gifts Ungiven better.
Tinker – I know, how could you not like Tinker? Actually, it’s one of my least favorite ‘untouchable’ cards. It’s slow at three mana, it’s a sorcery and it costs you an artifact. These are all serious liabilities in the control and combo matchups. Time Vault makes this card much better, but its value is still largely dependent on the impact of getting a large robot into play. Right now I think this value is very low. This elaboration requires addressing the robot slot as well.
When I hear people say Inkwell Leviathan or Darksteel Colossus or Sphinx of the Steel Wind are good in matchup X, where X is not fish, I immediately tune it out. While the Tinker>robot play is acceptable in the Stax matchup (and admittedly good in the rare workshop aggro matchup), it is terrible against dredge, oath, ritual combo and certainly tezzeret. If you’re playing this list and losing to Tinker>robot, you’re either unlucky or you’re doing it wrong. If I could find the sideboard space, I’d move both Tinker and the robot out of the maindeck.
Which robot? Easy: Sphinx of the Steel Wind. Based almost exclusively on the rationale above, Sphinx is the obvious best choice. Game one it’s innocuous and pitches to Force of Will in those non-fish matchups. Against Fish it is the best card to play hands down. It’s susceptibility to Sower of Temptation and Echoing Truth is completely offset by its ability to allow you to win at a very low life level. It’ll even save you from your own Dark Confidant or Mana Crypt while taking infinite turns sometimes.
There are corner cases where Darksteel Colossus or Sundering Titan can kill faster or be more disruptive in matchups where Tinker is typically bad, but these are low-probability events. The pitchability of Inkwell and Sphinx are of paramount importance now that Thirst for Knowledge is restricted. Given Sphinx’s advantage in the matchup where it is most relevant seals the deal in its favor.
1 Time Vault
1 Voltaic Key – Thank you Mike Solymossy for showing me the error of my ways. Before I truly understood the power of Time Vault, I considered Voltaic Key as superfluous to a tightly built Tezzeret list. I was utterly wrong. This card does cute things like combo with Sensei’s Divining Top and fix artifact mana. However, it is very powerful at opening up lines of play that Tezzeret (the card) is too slow for. For me, the combo is not Tezzeret > Time Vault, it is Time Vault + Voltaic Key.
Tezzeret the Seeker – I’ve played many Tezzeret (the deck) lists where I failed to include this card. At five mana, sometimes it’s just too clunky (relative to its vulnerability) even though it can win the game. Unless the metagame changes, however, I’ve come around to this being a very good addition to the deck. If nothing else, it offers an additional win condition slot against control/combo/oath and allows the siding out of Tinker + robot, making way for better sideboard cards. It also steals games. Along with the nice things it does with Gifts Ungiven or in the late game as a one-card win condition, there will be games where you can vampiric for this or Black Lotus or Tolarian Academy and just push it through in the early turns. It also enables sideboard options with stuff like Pithing Needle, Tormod’s Crypt and Trinisphere if the situation lends itself to that.
1 Sensei’s Divining Top – Using Dark Confidant as a draw engine requires this card. When a quick Vault/Key combo is not available you’re often taking on a very controlling role. To effectively leverage Confidant into the long game, you need this in the deck. In addition to supplementing the draw engine, Sensei’s Top is a reasonably powered card in that it ensures land drops, hides powerful cards from duress and combos with Key, bounce spells and tutors. There are a myriad of tricks with this card that involve stacking either the draw effect or the look effect in combination with other responses. Going into details about this would require its own small primer. In lieu of that, I’ll just encourage people to treat this card with the same care that Brainstorm has earned; there’s a lot of options wrapped into this 1cc spell.
1 Rebuild – While I maintain the opinion that Tinker>robot is weak in a lot of matches, this doesn’t mean you should ignore it. You should be able to race this threat most of the time. However, there will be situations where you have significant advantage, but not enough tutors, etc to set up the combo. In these cases you’ll want to fetch out Rebuild. More often, though, this card will be solid game 1 against Null Rod or Workshop decks, or buy you a crucial turn against an opposing Vault/Key. It can often help ramp mana or get an extra card from Sensei’s Top. As it cycles, this card exemplifies one of the chief philosophies of my deckbuilding approach: whenever possible have no dead cards.
1 Fire/Ice – Another mini breakthrough was including Fire/Ice in the final list before GenCon. Darkblast, while tempting, always ran the risk of missing its mark. Fire/Ice is scroll-able, cycles, and answers some of the key threats to the deck. It also answers Tezzeret itself, and can take out an unexpecting opponent at low life.
Mana BaseEarlier I mentioned the sweet spot for a mana count seemed to be between 23 and 25. I always felt 25 to be too high for previous Tezzeret lists and 23 seems untenable in a metagame with any concentration of null rod or workshops. Going with 24 has resulted in the most consistent hands, the best results against decks that pressure the manabase and allows the right configuration to support three colors. It would probably work fine to run 23 (and against Drains/Oath/Combo I routinely side down to 22-23 sources postboard), but 24 has the right ‘feel’. I’d be happy if the statistically savvy out there have mathematical arguments to speak to this intuition.
8 Black Lotus, Sol Ring, Mana Crypt, Moxen
1 Tolarian Academy
These accelerators are obvious inclusions which leaves fifteen spots left to support colors, have sufficient basics, and the fetches to find them. I’ve settled on the following counts:
2 Islands
1 Snow-Covered Island
6 blue fetches
Having 9 cards to access a basic island is sufficient to combat wasteland. With six slots left, I split evenly between:
3 Underground Sea
3 Volcanic Island
This is what I’ve been running the last few months. To be honest, there’s wiggle room to switch a basic or Volcanic Island out for a 4th Underground Sea depending on the expected metagame and sideboard, but for now this has been working.
This leaves the list at:
1 Time Vault
1 Voltaic Key
1 Tezzeret the Seeker
1 Sphinx of the Steel Wind
4 Force of Will
4 Mana Drain
2 Duress
1 Thoughtseize
1 Rebuild
1 Fire/Ice
3 Dark Confidant
1 Sensei’s Divining Top
1 Ancestral Recall
1 Time Walk
1 Brainstorm
1 Merchant Scroll
1 Tinker
1 Gifts Ungiven
1 Demonic Tutor
1 Vampiric Tutor
1 Yawgmoth’s Will
8 Solomoxcrypt
6 Blue Fetches
3 Underground Sea
3 Volcanic Island
2 Island
1 Snow-covered Island
1 Tolarian Academy
= 55 cards
This, in my estimation, is how you build Tezzeret Control.
The remaining cards are metagame considerations and should be tweaked according to the sideboard, but these 55 cards I don’t expect to change for the foreseeable future.
Rounding out the list and sideboard
At GenCon I went with these five:
1 Echoing Truth
1 Misdirection
1 Ponder
1 Fire/Ice (2nd)
1 Mystic Remora
Echoing Truth – this card is usually a staple for me. It’s incredibly flexible and is the one ‘utility’ card that I feel is strong enough to not need to cycle. Still, in the most recent event I played in, I decided to cut it. My rationale was that it doesn’t answer Oath, it’s mediocre in the mirror and it doesn’t do enough against Stax. I was also hedging that my sideboard would be enough to contain the decks that Echoing Truth would be good against. While I’d be supportive of people running this card, I’ve come to view it as a crutch for me. The reality is that I don’t need to be able to answer any permanent to win games and other options are more flexible.
Misdirection – this is another fine card that I cut in the Philly Open event. At the time of GenCon I would have included this as a near auto-include. It’s great in the mirror and helps against aggressive decks. The tricks with Ancestral Recall steal games all the time. Post-Zendikar, however, I feel this is too risky. It’s rare that people throw out an unprotected Ancestral Recall (especially in the early game). Spell Pierce makes this gamble even worse.
Ponder – this card is really underwhelming if you can’t use it in multiples to cheat your mana count down. That being said, it is solid turn one for setting up your hand, it shuffles for Sensei’s Divining Top and it’s blue. More than anything this card excels at making your deck 59 cards.
Fire/Ice – Noble Hierarch based fish was being talked up a lot during GenCon and Dark Confidant was being played in Fish, Tezzeret and Stax.
Mystic Remora – This was probably a mistake, but this was my 60th card at GenCon. It could have easily been other things.
Sideboard
There’s no comprehensive way to do this. I sought to have coverage against a wide open GenCon metagame and to patch weaknesses in the maindeck. I also sought to construct both the maindeck and sideboard in a way that maximizes the control and combo matches in game 1 while having stronger board options for Null Rod, Workshop and Dredge decks.
The strategy here being that ritual combo (and belcher) and most of all Tezzeret are capable of stealing games from you with broken hands at a much higher rate than either null rod or workshop strategies. I also expected a larger portion of the field to be the former group, cards that can be maindecked against them have more overlap and the better players usually play these strategies. There’s lots of room for argument on these points, but this is what I stick by.
The sideboard I settled on at GenCon:
4 Leyline of the Void
3 Ingot Chewer
2 Mystic Remora
1 Pyroclasm
1 Darkblast
1 Red Elemental Blast
1 Pyroblast
1 Pithing Needle
1 Yixlid Jailer
My philosophy on sideboarding is to be flexible. There are patterns I keep depending on certain matchups, but I try to rethink my sideboarding every game of every matchup. I do this by shuffling the sideboard in and taking out what I feel are the weakest fifteen cards. The nature of Vintage is that small changes in decklists can radically shift the way matchups play and what cards are effective.
However, as a general guide, here are some examples:
Tezzeret mirror:
-1 Echoing Truth
-1 Island
-1 Tinker
-1 Sphinx of the Steel Wind
+1 Red Elemental Blast
+1 Pyroblast
+2 Mystic Remora
Stax:
-2 Duress
-1 Thoughtseize
-1 Gifts Ungiven
-1 Misdirection
-1 Mystic Remora
-1 Tinker
-1 Sphinx of the Steel Wind
-1 Fire/Ice
+4 Leyline of the Void
+3 Ingot Chewer
+1 Darkblast
+1 Pithing Needle
Fish:
-2 Duress
-1 Thoughtseize
-1 Mystic Remora
-1 Gifts Ungiven
-1 Misdirection
-1 Sensei’s Divining Top
+3 Ingot Chewer
+1 Pyroblast
+1 Red Elemental Blast
+1 Pyroclasm
+1 Darkblast
In addition to knowing why certain cards made the list, it's important to understand why certain cards go missing. The following were considered and found lacking:
Fact or Fiction - this card is extemely powerful and solid in the matchups I value most. However, I routinely found that I didn't get my return on investment when I sink four mana into this. A large part of this is that a good player will make splits that work largely to your disadvantage (usually based on cards they have that you don't know about). Additionally, with Tezzeret and Gifts Ungiven already occupying the high end of the curve, I found FoF unneeded. In games where I had acceleration early enough to make FoF relevant, I usually win with the cards that replace it. This being said, I perfectly understand people who consider this an auto-include.
Mystical Tutor - I find this card hugely overrated. Then again, I may be guilty of overvaluing card parity. Still, I hate wasting a draw step regardless of the card this is getting. Part of this probably has to do with my distaste for Tinker, but much of it has to do with an understanding I've developed. Certainly Ancestral Recall is important to the early game and one of the most consistent ways to achieve an advantage that will lead to victory. However, I've learned that there are important circumstances where allowing Ancestral Recall to resolve is the right play. These occasions are increased when the net effect of the play delivers one less card of advantage.
Imperial Seal - While powerful, this card is comprable to Mystical Tutor in that sometimes you never recover the card advantage. This has the added shortcoming of being sorcery speed which severely limits your responsiveness. Both of these are terrible topdecks.
Mana Vault - This card is pretty powerful. It makes early Gifts and Tezzeret victories that much more likely. However, the weakness against null rod and the inability to equalize the mana requirements without taking up another slot convinced me to drop it.
It's worth noting that the advent of Spell Pierce makes all of these cards riskier and therefore less powerful.
Post ZendikarThe new cards from Zendikar force a revisiting of certain logic. In general Oath is much more of a metagame force, Dredge is stronger and early game gambits are more risky.
five residual slots:
1 Thoughtseize
1 Repeal
1 Ponder
1 Sensei’s Divining Top (2nd)
1 Annul
Thoughtseize – This replaces misdirection in response to Spell Pierce and also serves to better combat Oath of Druids.
Repeal – This serves in the Echoing Truth slot. It’s not as effective against permanent based disruption and not as much of a catch-all. However, it’s better at responding to opposing Dark Confidants or Oath of Druids. It’s also never dead and does tricks with Sensei’s Top.
Ponder – see above.
Sensei’s Divining Top – I wanted more early game control and I wanted to make sure Confidant didn’t cost me any games.
Annul – This was entirely based on expectations of the Philly Open metagame. Oath has been gaining steam and the area is known for its 5c Stax players. This also bought me an additional slot in the sideboard.
I encourage people to take my sideboard from the Philly Open with a grain of salt. Dead/Gone was awful and the whole list was based on a risk that Dredge wasn't present in big numbers. That hedge did pay dividends in swiss, but I was very lucky to get the placement in the top8 that I did, not facing Dredge until the finals. I tweak my sideboard every time I play, but if I had to play the Philly Open again I'd run this:
4 Leyline of the Void
3 Ingot Chewer
1 Mystic Remora
1 Red Elemental Blast
1 Annul
1 Massacre
1 Pyroclasm
1 Greater Gargadon
1 Yixlid Jailer
1 Pithing Needle
I wasn't able to dedicate time to make this an exhaustive account, but hopefully this will stir some discussion on a deck that is the most commonly played, but not as often discussed.