Ironically I played this mirror last week as well. I realized the way to winning this matchup is to ultimate Jace and steal his or her elixir and just never crack his. Once you can recycle your deck with your own elixir he will just end up decking himself. Side note, I actually lost my mirror because in response to me cracking elixir at his end step, he responded by a quicken-planar cleansing and I had no counters in hand (scrubbed out in that regard). Obviously I just milled myself in the end.
Elixir gets shuffled into your library in that scenario since it's in your graveyard when its ability resolves.
Deck is built around creating a recurring source of card advantage with Pack Rat and Chandra's Phoenix. Pitch two Phoenixes to Pack Rat, and you get two Rats. Burn the opponent, and you get the Phoenixes back and get to make two more Rats. Prophetic Flamespeaker and Chandra help provide more cards, and Return, an already brutal card, is just nuts when you get a Phoenix back with it. Pitching Soul of Shandalar to Pack Rat is also adorable. The card advantage engines are supplemented by evasive damage and additional reach from the burn spells.
Cyclonic Rift is cheaper and more versatile--important traits for a maindeck card. It's also the only way for the deck to interact with Detention Sphere at instant speed. That said Jest seems very good in creature-centric matchups, including the mirror. It can be used in response to a Master trigger to cause the Elementals to enter as 1/0s and die to state-based action, for example. Just be sure to Jest in response to Master's trigger, not after it reslolves; otherwise, the Elementals will all become 1/1s and survive. Jest can also shut down Polukranos--if you activate it in response to Monstrous, then the Frog will get the Monstrous counters and not fight anything because it no longer has Polukranos's triggered ability. Again, the timing is specific here--you can't Jest in response to Polukranos's fight trigger, because it will just become an X/X Frog and then kill your Master, etc. Also keep in mind that if you Jest in response to Monstrous and they have enough mana, then they can just activate Monstrous again in response to Jest, so be wary of that. Another thing to note is that an opposing Thassa can make critical creatures unblockable in response to Jest, preventing combat blowouts. Another cute trick is that if you have seven mana, you can Jest, then Domesticate a big creature such as Desecration Demon, Polukranos, or an active Thassa. Domestication's end step trigger will see a 1/1 Frog and won't sacrifice itself, and then when the turn ends the Frogs will go back to being that big beater, which you get to swing with next turn.
The main problem with Jest is that is only good in combat. If you're not killing a bunch of Frogs then it's not doing a lot for you, and if your opponent plays around those situations then it's just going to sit in your hand, useless. The more widely played it is, the less effective it is. Not to mention that an opponent can just wait until blockers are declared and then use spot removal to protect the most crucial Frogs. Punking Mistcutter Hydra and Skylasher is great though. Now, the wombo-combo with Izzet Staticaster does seem pretty promising... too bad it's atrocious vs control. And I don't mean Domestication-terrible, as in "this is a dead draw but it isn't going to through off your G1 percentages that badly and worst case can still be used on your own creature to activate Thassa." I mean, "you are drawing so many dead cards because of this package that your already low chances to win G1 are now abysmal". It does seem like it has the potential to be a very potent sideboard package though.
AEtherspouts is almost certainly too slow. This is not a durdly defensive deck.
Thursdayisgod, can you explain what you mean when you say U/W/R topdecks poorly? If anything, U/W/R seems like it can be the most proactive of the control variants. It can bring in AEtherling, Keranos, Assemble the Legion, Jace, Memory Adept, and more if it chooses, and has twelve Temples to help dig for action. Proactive is where I think U/W/R's sideboard should be, by the way (similar to Splinter Twin sideboarding in Modern); thanks to Counterflux, U/W/R never has to worry about losing the counterspell war in the mirror when it is necessary to answer a specific threat, like an AEtherling or a huge Revelation. This means that the deck can afford to spend more mana during its own main phase, because it doesn't need to leave up additional mana for extraneous countermagic. As such, the deck can just jam threats when it has Counterflux backup, which depletes the opponents own counterspells and puts them on the backfoot if they don't answer quickly. Once their reserves are depleted it becomes very easy to win the fights you want to put the effort into winning. This isn't even mentioning how backbreaking it is to Counterflux an Elixir and subsequently resolve your own. Proactive threats are also good in other matchups, depending on how difficult or unprofitable it is for a deck to interact with them. For example, neither Monoblack nor Jund can touch a Keranos, and Monoblack additionally can't do much against Assemble the Legion.
By the way, I am firmly in the "don't run Anger" camp. If it's not being played on turn 3 or 4 then it's generally not worth it, and in order to do that consistently you need about 19 red sources (a few less after accounting for scry and card draw), which the manabase simply cannot handle. It's better to just run a high amount of spot removal; Mortars is great in particular in that it can pull a pseudo-Wrath in the lategame if necessary, and also stomps on the problematic pro-white creatures in midrange matchups.
Also, has anyone considered running a 28th land in three-color builds? Reasoning being that it provides another source of colored mana to make casting your spells easier, and also provides room for another basic land, giving you more flexibility in regards to how you sequence your land drops, and making you take less damage overall. The twelve Temples also help mitigate flood. 28 lands probably sounds like a lot, but I'm sure 27 lands also sounded like a lot before it became commonly adapted. The primary reason I have preferred straight Azorius to U/W/R in the past is how clunky twelve Temples can make your play, but with more comes-into-play-untapped lands that becomes less of an issue.
Perilous Vault is a potent card, and while most likely not maindeck viable due to how it conflicts with Monoblack's primary gameplan of accruing board presence while restricting the opponent's options, Vault does do things that Monoblack generally does not have the capability of doing, and has the potential to be a good sideboard card. It removes enchantments and artifacts, cards that Monoblack can typically only remove beforehand with discard spells, and is also able to reset overwhelming board states. It can also just sit on the board dissuading the opponent from committing any cards for fear of losing them, meaning they either try to hit a critical mass of threats that forces you to activate the Vault for the minimum of value, or only commit one or two threats to the table at a time, which plays right into your removal spells. With a Vault and an Underworld Connections in play, it isn't difficult to sit back and gradually rack up overwhelming card advantage if the opponent just twiddles their thumbs in fear of over-committing and losing it all to Vault. Vault is also moderately difficult to remove once resolved (Detention Sphere, Banishing Light, Vraska the Unseen, Reclamation Sage, and Putrefy being the most common answers), and some decks, such as Monoblack itself, can't touch it at all.
It seems at its best vs Monoblue, where it easily answers their plan A of swarming the field, removes the otherwise extremely troublesome Thassa and her Bident, and is difficult for Monoblue to interact with once resolved. It's good against Hexproof as well, crippling their entire offense and likely leaving them with insufficient resources thanks to your other interaction. There's also potential against the mirror and Jund; in the mirror, it answers Underworld Connections, Erebos, and Whip of Erebos, all permanents that usually can't be touched if they slip through your discard, and it even deals with Pack Rat to boot; and against Jund, it mops up Xenagos and all of his Satyr tokens, Domri, and the big monsters. However, Jund has a much easier time answering Vault, since if they weren't already maindecking Vraska and Putrefy, they will almost certainly bring them in postboard, making the potential for a tempo blowout high.
While the benefits sound good, there is a serious conflict of interest with Monoblack's tools and Perilous Vault, a conflict that might prove to be too much to board it in against some decks where it would seem beneficial--Vault actively lessens the value of your other permanents while in play. Dropping an Erebos while you have Vault out may not seem appealing, since the opponent could choose to commit enough cards the next turn to force you to pull the trigger, costing you your Erebos. Similarly, Whip, Liliana, and Connections all seem worse. Whereas you might get multiple activations out of each in a normal game, in a boardstate with Vault you might only get one or two activations. Furthermore, in a boardstate where Vault is the only thing in play, while you might normally be the aggressor in such a situation, your own Vault creates similar tensions. You can commit multiple creatures to pressure your opponent, but then your Vault is just going to sit there since you don't want to blow up your own army, in which case it is essentially a dead card, whereas an Erebos might have been fueling your aggression and putting the game out of reach for the opponent. And if you do nothing, you give the opponent more windows to drop a Connections or Erebos of their own, two-for-oneing you at worst. This isn't even touching on the prohibitive mana investment required; sometimes Vault is just too slow, and sometimes you just can't afford to spend nine mana to answer one threat.
So, taking this into consideration, the only decks I would definitely board Vault in for at the moment are Monoblue and Hexproof. Other matchups where it shows potential would need more testing, as it might simply be more optimal to go all-in on Connections, Erebos, Liliana, and Whip for your sources of card advantage, and just hope to bury the opponent and 1-for-1 and 2-for-1 them enough that you don't get put into a position where Vault would have been your only out.
Elixir gets shuffled into your library in that scenario since it's in your graveyard when its ability resolves.
4 Pack Rat
4 Prophetic Flamespeaker
1 Soul of Shandalar
4 Stormbreath Dragon
2 Chandra, Pyromaster
2 Dreadbore
3 Magma Jet
2 Rakdos's Return
3 Stoke the Flames
3 Thoughtseize
3 Ultimate Price
1 Mana Confluence
8 Mountain
4 Mutavault
1 Rakdos Guildgate
2 Swamp
4 Temple of Malice
1 Urborg, Tomb of Yawgmoth
2 Doom Blade
3 Duress
3 Magma Spray
2 Mizzium Mortars
1 Perilous Vault
1 Rakdos's Return
3 Slaughter Games
Deck is built around creating a recurring source of card advantage with Pack Rat and Chandra's Phoenix. Pitch two Phoenixes to Pack Rat, and you get two Rats. Burn the opponent, and you get the Phoenixes back and get to make two more Rats. Prophetic Flamespeaker and Chandra help provide more cards, and Return, an already brutal card, is just nuts when you get a Phoenix back with it. Pitching Soul of Shandalar to Pack Rat is also adorable. The card advantage engines are supplemented by evasive damage and additional reach from the burn spells.
The main problem with Jest is that is only good in combat. If you're not killing a bunch of Frogs then it's not doing a lot for you, and if your opponent plays around those situations then it's just going to sit in your hand, useless. The more widely played it is, the less effective it is. Not to mention that an opponent can just wait until blockers are declared and then use spot removal to protect the most crucial Frogs. Punking Mistcutter Hydra and Skylasher is great though. Now, the wombo-combo with Izzet Staticaster does seem pretty promising... too bad it's atrocious vs control. And I don't mean Domestication-terrible, as in "this is a dead draw but it isn't going to through off your G1 percentages that badly and worst case can still be used on your own creature to activate Thassa." I mean, "you are drawing so many dead cards because of this package that your already low chances to win G1 are now abysmal". It does seem like it has the potential to be a very potent sideboard package though.
AEtherspouts is almost certainly too slow. This is not a durdly defensive deck.
By the way, I am firmly in the "don't run Anger" camp. If it's not being played on turn 3 or 4 then it's generally not worth it, and in order to do that consistently you need about 19 red sources (a few less after accounting for scry and card draw), which the manabase simply cannot handle. It's better to just run a high amount of spot removal; Mortars is great in particular in that it can pull a pseudo-Wrath in the lategame if necessary, and also stomps on the problematic pro-white creatures in midrange matchups.
Also, has anyone considered running a 28th land in three-color builds? Reasoning being that it provides another source of colored mana to make casting your spells easier, and also provides room for another basic land, giving you more flexibility in regards to how you sequence your land drops, and making you take less damage overall. The twelve Temples also help mitigate flood. 28 lands probably sounds like a lot, but I'm sure 27 lands also sounded like a lot before it became commonly adapted. The primary reason I have preferred straight Azorius to U/W/R in the past is how clunky twelve Temples can make your play, but with more comes-into-play-untapped lands that becomes less of an issue.
Perilous Vault is a potent card, and while most likely not maindeck viable due to how it conflicts with Monoblack's primary gameplan of accruing board presence while restricting the opponent's options, Vault does do things that Monoblack generally does not have the capability of doing, and has the potential to be a good sideboard card. It removes enchantments and artifacts, cards that Monoblack can typically only remove beforehand with discard spells, and is also able to reset overwhelming board states. It can also just sit on the board dissuading the opponent from committing any cards for fear of losing them, meaning they either try to hit a critical mass of threats that forces you to activate the Vault for the minimum of value, or only commit one or two threats to the table at a time, which plays right into your removal spells. With a Vault and an Underworld Connections in play, it isn't difficult to sit back and gradually rack up overwhelming card advantage if the opponent just twiddles their thumbs in fear of over-committing and losing it all to Vault. Vault is also moderately difficult to remove once resolved (Detention Sphere, Banishing Light, Vraska the Unseen, Reclamation Sage, and Putrefy being the most common answers), and some decks, such as Monoblack itself, can't touch it at all.
It seems at its best vs Monoblue, where it easily answers their plan A of swarming the field, removes the otherwise extremely troublesome Thassa and her Bident, and is difficult for Monoblue to interact with once resolved. It's good against Hexproof as well, crippling their entire offense and likely leaving them with insufficient resources thanks to your other interaction. There's also potential against the mirror and Jund; in the mirror, it answers Underworld Connections, Erebos, and Whip of Erebos, all permanents that usually can't be touched if they slip through your discard, and it even deals with Pack Rat to boot; and against Jund, it mops up Xenagos and all of his Satyr tokens, Domri, and the big monsters. However, Jund has a much easier time answering Vault, since if they weren't already maindecking Vraska and Putrefy, they will almost certainly bring them in postboard, making the potential for a tempo blowout high.
While the benefits sound good, there is a serious conflict of interest with Monoblack's tools and Perilous Vault, a conflict that might prove to be too much to board it in against some decks where it would seem beneficial--Vault actively lessens the value of your other permanents while in play. Dropping an Erebos while you have Vault out may not seem appealing, since the opponent could choose to commit enough cards the next turn to force you to pull the trigger, costing you your Erebos. Similarly, Whip, Liliana, and Connections all seem worse. Whereas you might get multiple activations out of each in a normal game, in a boardstate with Vault you might only get one or two activations. Furthermore, in a boardstate where Vault is the only thing in play, while you might normally be the aggressor in such a situation, your own Vault creates similar tensions. You can commit multiple creatures to pressure your opponent, but then your Vault is just going to sit there since you don't want to blow up your own army, in which case it is essentially a dead card, whereas an Erebos might have been fueling your aggression and putting the game out of reach for the opponent. And if you do nothing, you give the opponent more windows to drop a Connections or Erebos of their own, two-for-oneing you at worst. This isn't even touching on the prohibitive mana investment required; sometimes Vault is just too slow, and sometimes you just can't afford to spend nine mana to answer one threat.
So, taking this into consideration, the only decks I would definitely board Vault in for at the moment are Monoblue and Hexproof. Other matchups where it shows potential would need more testing, as it might simply be more optimal to go all-in on Connections, Erebos, Liliana, and Whip for your sources of card advantage, and just hope to bury the opponent and 1-for-1 and 2-for-1 them enough that you don't get put into a position where Vault would have been your only out.