Overview

Tibalt's Trickery is a ridiculous card. In case this is your first time seeing it, and you're having a some trouble with the wall of text on the card itself, what it does is counter a spell, but then let its controller cast a random spell from the top of their deck for free. There's a bit more nuance to it than that, but that's basically what you get.

I think the idea with this card is that it would be a sideboard option for aggressive decks against control decks. At 2 mana it's pretty easy to leave open, and because control decks are packed with counter spells (which probably can't even target anything if flipped by Trickery) and have relatively few high impact threats, if you counter some big draw spell or board wipe you're almost certainly going to be better off with whatever random free spell they get.

But that's not what we're doing here. With this deck, the plan is to cast Tibalt's Trickery on our own 0 cost spell on turn 2, and hopefully cast something strong enough to win the game.

If you're thinking that this combo will be way too unreliable with only 4 Tibalt's Trickery and 8 0 cost spells (4 each of Tormod's Crypt and Stonecoil Serpent), you're mostly right. But the odds are a lot higher than you're probably thinking. I wrote some simple python code to simulate playing with this deck, and he odds of hitting the combo in the first 5 turns of the game (if you follow my mulligan rules) are around 62%.

To be clear, that's not high enough to make this deck competitive. You aren't going to win every game where you get the combo to work, and if your opponent is running duress, negate or anything they can play on turn 1 or 2 to disrupt you, the odds absolutely plumet.

It is however more than enough to make the deck fun to play. Games with this deck, win or loose, are very quick. So if you spend 20 min to half an hour playing it you should get a least a couple absolutely absurd Genesis Ultimatum hits, or a Chromatic Orrery that lets you play Prismatic Bridge from hand, or some other silly combination of cards that lets you just absolutely obliterate your opponent before they have a chance to do much of anything.

Mulligan Strategy

The key to getting the combo consistently in this deck is aggressive mulligans. But aggressive doesn't mean simple. The optimal mulligan strategy for this deck changes with each mulligan we take.

• For the first 3 hands we see
  • Kick anything that doesn't have both Tibalt's Trickery and one of Stonecoil Serpent or Tormod's Crypt
• For hands 4 and 5
  • We still kick anything that doesn't have Tibalt's Trickery in it, but we can keep without a 0 cost spell now
    For the 6th hand
  • We're getting desperate at this point, and we're better off keeping any hand with a 0 cost spell and at least one land, even without Tibalt's Trickery
  • Obviously, we still keep anything with Tibalt's Trickery in it, regardless of the land count
• For the 7th and final hand
  • We're technically allowed to mulligan this hand and then return all 7 cards you draw to the bottom, but we shouldn't
  • If we hit a Tibalt's Trickery on this last hand, that will be the card we keep, and at that point we've still got about a 25% chance to hit the combo by turn 4 if we're going second or turn 5 if we're going first
  • If we didn't hit Tibalt's Trickery our odds are so much lower we're better off conceding and moving on to the next game
     

Python Simulation

As I hinted at above, I depended heavily on python simulations to maximize the chances of getting this combo to work as early and consistently as possible. If even one person leaves me a comment asking to see my code, I'll put it up on github and leave a link somewhere. But until then, I'll just give an overview of what I did and discuss my results here.

I simplified the problem somewhat by considering only 7 different card types, rather than every possible card I could put in the deck:

• High CMC threats that can hopefully win the game by themselves
• Untapped lands that can produce red mana
• Tapped scry lands that can produce red mana
• Turntimber Symbiosis
• Tibalt's Trickery
• Tormod's Crypt
• Stonecoil Serpent

I then made a bunch of decks with different numbers of each of these cards, and wrote a program that played out the first 5 turns of a game of magic, 1,000 times for each deck and recorded how often the combo went off. To simulate those 5 turns of magic, I had the program step through this list of actions

• First, mulligan according to the mulligan strategy laid out above
  
  • I didn't use the simulation to inform the muligan strategy, I just did a bunch of odds calculations
• Then, each turn
  • Draw a card if it's not the first turn
  • Play a land if you canIf it's a scry land, look at the top of the deck, and put the card on the bottom if it isn't a missing piece of the combo or a land you need to cast Tibalt's Trickery
  • Check to see if we have the mana and cards available to play the comboIf we do, simulate casting Tibalt's Trickery by flipping 1-3 cards off the top of the deck and then seeing if we hit a High CMC Payoff card before another Tibalt's Trickery or a 0 cost spell
  • If we sucessfully cast a payoff spell, check how many payoff spells we have in our hand + the top card of our deck (I used this to determine how usefull extra mana would be and therefore help me choose which High CMC payoff cards to include)

The first clear result I got was that decks with 4 copies of Tibalt's Trickery vastly outperformed those with less. No surprise there, but it's a good sign my code does what it's supposed to. So, I kicked out all the decks with less that 4 Tibalt's Trickery and had the software crank out 10,000 attempts with all the remaining decks.

With the increased attempt count, it became obvious that decks with 4 copies of Turntimber Symbiosis, Tormod's Crypt, and Stonecoil Serpent were ahead of those without. Again, not terribly surprising, but good to know. But, more importantly, this narrowed the field of decks enough that I could crank the number of attempts per deck up to 1,000,000 and still have the simulation finish in an hour or two, so I did that.

And at the end of all those simulations, the most consistent deck had 19 High CMC payoff cards, 16 Untapped lands, 8 Scry lands, and 4 each of Tibalt's Trickery, Tormod's Crypt, Stonecoil Serpent, and Turntimber Symbiosis. With these card counts, the combo went off on turn two 47%, going up to 54% on turn 3, 60% on turn 4 and ends up at 65% on turn 5 when the simulation stops.

However, that's not the deck I ended up going with. The counts in this deck are instead 29 payoffs, 14 untapped lands, and 8 scry lands, and the percentages are 40% turn 2, 50% turn 3, 56% turn 4, and 62% turn 5. You'll probably notice that these odds are lower. But I chose this arrangement over the most optimized version, because this is the best result I got with 29 payoff cards in the deck. And if you're wondering why that's important, then please join me in the next section.

 
Deck Construction

The key challenge I had when building this deck was finding enough cards that are strong enough to win the game by themselves. Of course this is much easier in older non-rotating formats. Even historic has a couple big pickups like Ulamog, the Ceaseless Hunger over standard, but older formats also have Thoughtsieze and Spell Pierce. So even if our deck is more powerful, our opponents have much better options for stopping our combo, so we're better off in standard.

And in standard, there really aren't that many good options. There's Koma, Cosmos Serpent, Ugin, the Spirit Dragon, Kiora Bests the Sea God, maybe Dream Trawler, and Genesis Ultimatum. All in all, there's just enough good cards to fill out the 19 payoff slots in the most optimized version of the deck my code found. This 60 card version of the deck would absolutely get the combo to go off more often than the 67 card version I've decided to go with, and most likely have a higher win rate, but that's not really what this build is about.

The problem with the 60 card deck is that it will only ever get to cast one powerful spell. After that you just have to spend 2-3 turns attacking in and praying your opponent doesn't have a way to remove whatever you managed to cast. Maybe one game in 10 you hit a good Genesis Ultimatum, and things are a bit more interesting, but for the most part, out of the 5 turns you're likely to play with this deck, exactly one of them is fun.

So I wanted a way to keep playing the game past that first turn when the combo goes off. And, armed with the knowledge from my simulation that you'll have an average of 3 High CMC threats in your hand after casting the combo, my plan was to include a bunch of high end mana generation in my payoff cards. The three best cards I could find to fill this role were Chromatic Orrery, Goldspan Dragon, and Nyxbloom Ancient. And with these cards in place, I had to start thinking about the mana costs for my other payoff cards.

When Chromatic Orrery comes into play, you get 5 mana to spend immediately, so we want at least a few 5 mana cards we can play. The good news is that Goldspan Dragon costs 5 mana, so we're partway there already. The bad news is that Prismatic Bridge is basically the only other 5 mana spell we're ok with hitting off Tibalt's Trickery, so that's where our 5 cost spells have to end. The more important number for us is 7.

If we get Goldspan Dragon on turn 2, then by turn 3 (after combat) we will have generated 2 treasure tokens, each worth 2 mana of any one color. This combined with the 2 land we know we have because we cast Tibalt's Trickery last turn and the 3rd untapped land we'll be able to play this turn 75% of the time gives us 7 mana. With Nyxbloom Ancient we'll have 9 mana with an extra land, and with Chromatic Orrery We'll have 7 mana whether we manage to play a second land or not. All told, if we hit a mana generator off Tibalt's Trickery, we'll have a better than 80% chance to have 7 mana available the next turn. And 7 mana is enough for an Ultimatum.

So we run 8 Ultimatums in this deck. 4 Genesis Ultimatum and 4 Emergent Ultimatum. Both of these are absurdly powerful in this deck. Genesis Ultimatum needs only a brief explanation: it's a way to get multiple threats on the board at once, and ramp up 2 lands on average as well. Emergent Ultimatum is a bit trickier to explain, but no less powerful.

The key thing to know about this card in this deck is how it interacts with double faced cards like Esika, God of the Tree and Valki, God of Lies. When you search your library for Mono-colored cards, you can find both Valki and Esika because their front face is mono-colored. However, when you go to cast them you have the option to choose either face, because Emergent Ultimatum puts no restriction on what kind of card you can cast with it, only what kind of card you can search for with it. So when we cast it, the plan is to find Esika, God of the Tree, Valki, God of Lies, and Vorinclex, Monstrous Raider. This way our opponent is given two absolutely horrible options when deciding which one we don't get.

If they give us Vorinclex, Monstrous Raider and Tibalt, Cosmic Imposter, then Vorinclex will double the number of loyalty counters Tibalt enters the battlefield with, allowing us to use his ultimate immediately. And since this started with us casting Tibalt's Trickery on a Stonecoil Serpent or Tormod's Crypt, both those cards are currently in our graveyard, just waiting to be cast again.

If, on the other hand, they kick back one of those cards, then we get to resolve Prismatic Bridge, and that's the strongest card in our deck. As an enchantment it's pretty hard to remove, especially in the early turns, and it gets us a never ending stream of game ending threats at start of every turn.

If you're unlucky enough to have drawn or milled one of Tibalt, Cosmic Imposter or Vorinclex, Monstrous Raider, then you'll have to pick backup targets like Kiora Bests the Sea God, Goldspan Dragon or Nyxbloom Ancient. Because of these backup options, you'll always be able to cast two payoffs, meaning that even if we don't get the exact three cards we're hoping for, Emergent Ultimatum is still better than any single card we could cast.

And with all these choices established, here's where we come to my decision to increase the number of payoff cards, (and even the total number of cards) beyond those in the optimal result I got from my simulations. If you're going down this road of including mana generation and 8 ultimatums, then 19 payoff cards doesn't leave enough left over for any other cards. The most efficient way I found to get down to 19 was to cut Emergent Ultimatum along with the cards that support it, and one Nyxbloom Ancient. But at that point you're left with only 7 cards in the deck that can actually damage our opponent, none of them stronger than a 5/5. That means if our opponent plays Lovestruck Beast, that's actually too strong for our deck to beat anymore.

Therefore, we either have to give up on using the optimal card distribution, or give up on casting more than one spell per game. And I know which of those is more important to me.

Conclusion

This deck is in no way consistent enough to be competitive. If you play it, be prepared to lose just under half your games to bad draws, and about one in 5 to early removal, discard or counter spells. But losses are very quick with this deck, and there's not much you need to think about while playing it, so for me at least it's really easy to move on from my losses with this deck.

And when you win with this deck, it is always a spectacular win. Just as an example, let me recount the highest high roll I've gotten with this deck to date.

After Mulligans, I had 4 cards in hand: Tibalt's Trickery, a Scry land, a pathway, and a Stonecoil Serpent.

I'm going first so turn 1 I play the scry land, see Esika, God of the Tree on top and keep it there.

Oponent plays a mountain and passes.

Turn 2, I hit Chromatic Orrery off the combo, and play Prismatic Bridge from hand.

Opponent plays Robber of the Rich, hits me for 2 and passes.

Turn 3 I hit Nyxbloom Ancient with Prismatic Bridge and draw Genesis Ultimatum. I tap Chromatic Orrery for 15 mana, and play Genesis Ultimatum hit 2 land, Emergent Ultimatum and Koma, Cosmos Serpent. With the rest of my Mana I cast Emergent Ultimatum grabbing Tibalt, Cosmic Imposter, Kiora Bests the Sea God and Vorinclex, Monstrous Raider. My opponent puts Tibalt back in my deck, Kiora, Bests the Sea God gets two counters immediately so I get an 8/8 Hexproof Kraken and my opponent's Robber of the Rich get frozen. I swing for 6 with Vorinclex and pass the turn.

Koma, Cosmos Serpent makes a 3/3 snake, and my opponent plays a Anax, Hardened in the Forge,  and passes back to me. At this point my opponent is spamming "Nice" at me repeatedly, but they're still in the game and they're not roping so I assume that means they're also just laughing uncontrollably at the absurdity of the situation. Props to them.

Turn 4. Koma makes another serpent. I low-roll and hit Esika, God of the Tree off Prismatic Bridge, but it doesn't matter. Kiora Bests the Sea God Triggers and steals the Anax, Hardened in the Forge my opponent played last turn. They now have no blockers and 14 life, while I've got 30 power worth of attackers ready to swing in. And just in case that wasn't enough, I've got 4 lands on the field. With Nyxbloom Ancient, that's enough mana to draw 5 cards with Chromatic Orrery and still have 7 mana leftover to cast the Turntimber Symbiosis I pick up, and put a Goldspan Dragon onto the field.

And honestly, that was easily enough fun for me to justify playing a deck with a negative win rate for a couple hours.