Indeed - my local meta is filled with Esper and Jeskai (because everyone in Seattle thinks they're the best control player in the world), but I don't expect those two to be anywhere near as prevalent at the GP, so Geist will probably be less great there.
A few tweaks from the last time I posted a list:
I've moved Hurkyl's to the main since I would like that fifth copy and I also expect a decent amount of Eldrazi Tron so having a maindeck answer for Chalice is good.
Finally switched over to the set of Nets and down to just one Kite Shield, as I expect a lot more matches where Reach is relevant over the jump from 4 to 5 toughness.
Things I'm still not sure on:
14 vs. 15 lands - I've had a lot of land-induced mulligans lately (by which I mean no lands, not one landers, but plenty of those too). Will probably end up dropping the 21st equip (last Kite Shield) for another Fountain but I'm still unsure. Is nayone else still on the 14 land train or has everyone moved over?
There's been a fair amount of discussion about using other fetches to imitate something else if possible. Has that been working out for folks? Does it outweigh the knowledge we give observant opponents that we have no basic islands? (though I suppose decks that play Blood Moon are still going to trot it out ASAP regardless).
I've still been very happy with Geist against Control matchups (and am happy with it vs. BGx decks as well since they really only have Lili as an answer, especially once you start suiting it up and make it a 2/unkillable by blocks). Should I keep it or replace it with a third Paradoxical/some other alternate threat (none of which I'm too happy with at the moment)?
Should I swap out the Wear/Tears with Fragmentizes? I feel like the answer is yes but I do really like instant-speed answers in the deck.
I have recently built this deck, and now I really want your help guys. I'm going to my first tournament tomorrow, to test it but I'm not really sure of the deck list. This is what it looks today:
Now, before you say anything: I know the sideboard looks a little weird right now. The idea is to have 3x Leyline, I can have that tomorrow and take the Island and something more out. I know Bone Saw is better than Kite Shield, I haven't got any yet though. And also, I will play a Horizon Canopy instead of something in main, when I get it.
Here is my issues about this deck list though:
Do I really play the Goblin package (Goblins+Boots) or is it win more? The thing is that it needs Puresteel on board to work.
Swan Song in main? In expense of what? The goblins? Serum Visions? I like the idea of having more than one win con in the deck, but what?
I got myself some Paradoxial Outcomes. What is it good against? Why do people play it in the sideboard? It feels too expensive?
What is the best cards to board out? Can I take a Sram out because it is legendary? Or do I board out Serum Visions and cards like that?
Anything else I can do to tweak the list? Thanks in advance!
EDIT: Repeal in main instead of Serum Visions or something? Bouncing those annoying Stony Silence and Chalices would be awesome G1, wouldn't it?
The Gaveleer plan doesn't really work with the all-in build anymore, as it's basically just win-more and takes up slots we could better use. I'd cut it (and Greaves) entirely. You can put more filtering/protection in those slots.
Additional wincons in the main aren't really necessary, as we're situated well against a decent number of decks game one. The board is where you want them.
Paradoxicals are for BGx decks/decks where they have a high concentration of interactive cards and a slow gameplan. The idea is that it allows you to get back in the game (and often win straight up) off of it if they've dealt with your starting engines, drawing 5-7 cards at the opponent's end step and then going off on your next turn.
Boarding depends on the matchups, but generally the flex slots (protection/filters) are the first to be boarded out. Don't board out any engines - they're what make the deck work, why would you board one out?
Also small note on the manabase - Glimmervoids can be a little awkward with the deck (especially if you hit the two glimmervoid opener, in which case you have to waste an artifact to play your land. The value of it also isn't really needed in the deck, since we only care about red at the very end. I'd recommend switching them over to more Seachrome Coasts or Fountains.
No problem, I love to see people using Powder in just about any deck, so I'm glad to share any wisdom I have about it.
I played several matchups the other night with a friend; some thinngs that happened:
1) I opened with 5 lands on 4 separate occasions, and kept all of them because they also had an engine. I lost each of those games horribly. I think it might be right to mulligan 4-5 lands even if the other cards are perfect.
2) I bricked off Paladin+Sram with 24 cards left in my deck. I think I had played something like 10 equipments and at least 1 Retract. I also cast SV as a last-ditch effort as my final spell. My list has 3 Noxious, didn't see them or the other Retract or even Opal+Grapeshot.
An engine is necessary but not sufficient for an opener. I would probably not keep a 5-land, engine, 1 random card hand. The 4-land hand definitely depends on what the two cards are.
As for the two-engine brick, it's rare but does happen. Was SV one of the last two cards you drew when you bricked? If not casting it mid-combo is usually better to help continue the chain/dig for the Retract/Opal piece you need to finish the combo.
Whiffing on finding a Retract has been the most common point of failure for me so far, so I'm not entirely against one main Hurkyl's. I do end up with a lot of combos where I have only U, and playing Hurkyl's does mean that you can't SV to help continue the chain if you're mana-light just in case. It will drastically improve the kill rate when you have 4+ mana, but is often dead if we're going off T2 or T3. I've been running some in the board for non-interactive matchups; I'll consider trying one main tonight's modern event.
You need Noxious to generate infinite mana, as your Opals will all be tied up in the loop (1 for Retracts, 3 for Elixir). You can get infinite life without it, though (as well as one infinitely large Grapeshot). If you're only casting one Grapeshot, this isn't relevant since you can just use the Elixir Opals at the end for Grapeshot mana, but if you need to cast multiple Grapeshots (in the case of Abzan when they have the combo on board) then you either need sufficient extra mana from lands/Mantles or you need the Noxious to make you a 5th Opal in the loop, generating infinite mana.
As long as you're giving a master class here, if we're able to draw through our entire deck with one engine in play, what are the conditions under which we could still pull of the loop? Can it be done?
One engine is certainly trickier, since as is the case with the normal win drawing your whole deck with just one engine isn't all that likely, but so long as the following are true I believe you should be able to loop regardless of how many cards are left in your deck:
1. You have 4 mana available (including Opals) to cast/activate Elixir and play Retract
2. You have at least as many Cheerios in your hand + on the battlefield as you do cards in your deck + graveyard.
3. You still have a Retract in hand (or Noxious giving you one).
At that point, you can play Retract, play Elixir, activate Elixir, play enough Cheerios to draw the deck, play Opals for mana (Noxiousing one back and playing it again to generate a net mana in the loop), and repeat. After the first repetition, the board state should be the same as the end of the first cycle (only Opals + NR + Retract in yard, no cards in deck, Elixir in play, storm count increases and you've generated a mana) and you can thus show it's a loop to shortcut (and since we're drawing the whole deck every time and not dependent on its order the randomness of shuffling shouldn't be an issue in demonstrating the loop like it is for Four Horsemen).
If instead of a Retract, you have an Elixir, you can try to go for it and hope to draw a Retract after the shuffle with your remaining Cheerios in hand, but there's definitely the fizzle possibility there.
You need Noxious to generate infinite mana, as your Opals will all be tied up in the loop (1 for Retracts, 3 for Elixir). You can get infinite life without it, though (as well as one infinitely large Grapeshot). If you're only casting one Grapeshot, this isn't relevant since you can just use the Elixir Opals at the end for Grapeshot mana, but if you need to cast multiple Grapeshots (in the case of Abzan when they have the combo on board) then you either need sufficient extra mana from lands/Mantles or you need the Noxious to make you a 5th Opal in the loop, generating infinite mana.
You are wrong, if the win infinite life but you create infinite damage, you win.
Also I think am artifact IS less risky than a creature.
kirtash is correct here that Elixir allows us to beat the Abzan combo, though not for the reason they seem to think (infinite damage does not automatically beat infinite life - it depends on the circumstances of how the life is gained, as well as who is the active player in some cases). We certainly can't win with a single infinitely large Grapeshot while they have the combo out - they just gain more life than the number we give in response. However, while we're generating the infinitely large Grapeshot, we can first kill a piece of the combo and stop them from being able to change their life total any further. Once that's done, we demonstrate and then shortcut the loop to make the storm count large enough to kill them. The mana for casting multiple Grapeshots while spending three for the Elixir loop shouldn't be an issue if you have Noxious in the deck - you can recur Opal each loop to generate a mana, so you can demonstrate infinite mana (as well as life and damage).
With all that said, I'm still not convinced this is necessarily a great option for Abzan combo. Keep in mind that once they do have the combo online, not only can they gain infinite life but they also get to choose what card they draw next due to the infinite scry 1, so they'll easily be able to find all their hate/most effective cards. You really only have the one turn after they go off then to try and go off
So, at this point I think folks are generally happy with their mainboards (the general consensus seems to be only a few cards off, and that's usually a difference in two of land/cheerio/cantrip count, so it's not that big a difference).
At this point, I'd really like to start a larger discussion about sideboarding, since that seems to be the most nebulous thing right now (I know my sideboard feels very loose right now, and I'm sure I'm not the only one). Instead of trying to come up with 15 cards, I think the better method is identifying what we need for each matchup, and subsequently figuring out how many cards we want for those matchups. then we can try to consolidate. As an example:
Control: This matchup hinges on settling a threat through countermagic and holding it through the combo. Silence seems strongest here, as it allows you to hold the whole turn, but other 1 CMC countermagic/protection could be considered. Alternate threats which dodge their plans (such as Geist) are also options. Recommended cards: Silence, Geist, cheap countermagic (Dispel?).
Burn: Post-board, they're bringing in more spells which can hit creatures, as well as Eidolons if they're not already main. Protecting the combo is important, but dealing with Eidolon is critical as we actually can't win with it in play. Recommended cards: Echoing Truth, Path to Exile, Leyline?
BGx: Post-board, they're pretty much guaranteed to have either hand disruption or removal in their opener. We have two routes here: either go for a critical mass of threats to try and top their 1 for 1 spells or board in spells which put you back in the game at a moment's notice. Recommended board cards: Monastery Mentor (for the former plan), Paradoxical Outcome (for the latter plan), Silence (for both), Leyline?
Other decks: To be frank, in two weeks of live tournament play with the deck I haven't lost a match against any deck that wasn't one of the three archetypes above, and in almost all cases the matches against non-interactive decks were straight up rolls so boarding didn't matter too much. If we actually want to go into detail, though: if they're doing their own thing (Tron, Merfolk, Infect, Affinity, etc.) and can't interact, we're generally just faster than them. Expect some spot removal to come in, but generally they won't have too much against us. Recommended cards: Hurkyl's Recall (to help ensure we don't fizzle, especially since you'll likely have the extra turn where you have sufficient mana for it).
I also strongly recommend boarding in at least one Echoing Truth in most cases. Between Eidolons, Leylines, Chalices, and other random things, it's almost always worth bringing one in to hedge.
Curious what other cards people like for specific matchups (most notably for the three mentioned above, as those are the ones we need the most targeted cards for).
Has anyone considered Ojutai's Command in the side as a singleton? Might help against the removal heavy match-ups. Instant speed engine from the graveyard + countering a snapcaster or drawing into more cheeri0s might be the right ticket. It costs 4, but as a 1-of it shouldn't stress the mana-base too much.
This is what Paradoxical Outcome is trying to do for the BGx matchups. Given there isn't always an engine in the yard (due to Path), being able to draw a lot of cards off Paradoxical seems slightly better to me (since usually at the point where you'd hit that SB slot you're just holding a bunch of Cheerios and waiting for something).
Won a GPT with the deck today (same list I posted a few days ago but -1 Silence, +1 Hurkyl's in the board). In the swiss, wins against Infect, Esper Control, and Skred Red, with losses to Grixis Control and Junk. Ended up in 8th and survived being on the draw against Grixis Delver and Elves (finals was split since the other player couldn't go to Vancouver).
Observations/Notes from today:
Once again, Geist of Saint Traft out of the board was an MVP vs the control decks when I expected they boarded out sweepers (especially when you get it down on T2 with Opal and then suit up with a Shield to stop Snapcaster blocks). Highly considering going up to two.
The Sram,Paladin,Opal,Land,3xCheerio hand came up quite a bit today (at least 5 times), and it's certainly something worth discussing. I snap kept it on the draw against the non-interactive decks like Infect and Elves, but in the dark or against other decks it's much closer. On the play in the dark is also an interesting discussion since you can dodge bolt with the T1 Paladin if you have a Shield Cheerio.
SV is infintely better in the opening hand, and in almost all cases it's also better when comboing off, as the potential of removing two bricks/ensuring you get what you need is worth as much if not more than a random card in hand IMO. This is especially true if you have multiple engines out, where ensuring you chain and draw two more than makes up for the one random card. Yes, you can play Repeal on an Opal, but that just effectively makes it a worse Cheerio.
Silence and Swan Song both accomplish what Judge's Familiar does, but better. There's been a lot of discussion about Mentor and Myth so I'd dig through the last 10 pages or so of the thread and see the discussions there (tl;dr: Mentor has the same issues that our engines do since they die to the same cards. Myth is better but also dies to a lot of the same like push and path)
Ascendancy doesn't actually get us up on cards (which we need to do to combo off), and looting can be problematic if you're trying to combo off with Sram and don't have enough cards in deck because of Ascendancy (unlikely to happen but possible). The cost/mana requirement is also pretty rough, especially for a deck that already has trouble getting to two land. Further, it's mostly dead mid-combo, since you usually have no or one open mana (aside from the retract opal mana), and when you do, just play a second engine copy to ensure you get there. The pump is almost never useful, and neither is the untap (except for the sole case of Paradise Mantle on a non-summoning sick engine, but then that means the engine already lived a turn so you're probably fine anyways). Outside of the 3-land, ascendancy, three-Cheerio opener, I'd never be happy to see it in my opener (and even that scenario I'd still be meh about). The most important thing, though, is that we don't desperately need another combo piece. I've certainly mulliganed plenty of hands due to not having an engine, but that's part of the deck, and eight copies means that between your 7, 6, and 5 you should be able to get one. If you want an alternate win-con for board, it should be something that attacks from a different angle (see below on why Mentor isn't this).
Mentor suffers from the same issues that Sram and Paladin do - it dies to just about all the same cards. At that point, you're just playing a more expensive card with the same risk, which doesn't really do much for us. Cards like Geist, Bastion Inventor, and Ghirapur Aether Grid, however, all attack from a different angle, and usually dodge the answers your opponent brought in.
In the "alternate threat" sideboard slot I tried Geist today, and I sided it in vs. control figuring they might have boarded out Verdicts in favor of cheaper single-target spells (they did). Played it turn two on the play with Opal and rode to victory after equipping with a shield equipment to survive Snap blocks. Going to try it again next event and try to see how I feel about it in the long run, but I'm pretty happy with it so far over other options. Might go up to two, we'll see.
I'm also slightly warming up to the Burn matchup. Eidolon is obviously very rough post-board, but the matchup certainly doesn't feel unwinnable, especially if you dodge the Eidolon (certainly the matchup is skewed against us but so far all of my games have been such that the person on the play wins and the loser is one turn off, which gives me hope in the matchup if you win game one).
I also highly recommend actually representing as UWR/Esper control until you go for it game one (or it otherwise gets revealed). Hollywood them when they're casting their turn two/three spells and pretend you're thinking about Leaking/some other counter on their spells and try to bait them into the tap-out so you can combo without fear (especially against Burn which might not play double spell turn two unless you help guide them into it).
Still debating land count (my current main is the same as Shinx's but -1 Sleight, +1 Cheerio), and I may switch that 21st equip to another fetch since I have been mulliganing a lot due to lands (and keeping some sketchy one-land with SV on the draw as a result).
4 Puresteel Paladin
4 Sram, Senior Edificer
Artifacts: 25
4 Mox Opal
4 Accorder's Shield
4 Cathar's Shield
1 Kite Shield
4 Spidersilk Net
4 Bone Saw
4 Paradise Mantle
Spells: 13
3 Serum Visions
2 Grapeshot
1 Noxious Revival
2 Swan Song
1 Hurkyl's Recall
4 Retract
4 Seachrome Coast
3 Hallowed Fountain
1 Sacred Foundry
1 Plains
3 Flooded Strand
2 Arid Mesa
4 Silence
3 Echoing Truth
3 Path to Exile
2 Paradoxical Outcome
1 Geist of Saint Traft
2 Wear // Tear
A few tweaks from the last time I posted a list:
I've moved Hurkyl's to the main since I would like that fifth copy and I also expect a decent amount of Eldrazi Tron so having a maindeck answer for Chalice is good.
Finally switched over to the set of Nets and down to just one Kite Shield, as I expect a lot more matches where Reach is relevant over the jump from 4 to 5 toughness.
Things I'm still not sure on:
The Gaveleer plan doesn't really work with the all-in build anymore, as it's basically just win-more and takes up slots we could better use. I'd cut it (and Greaves) entirely. You can put more filtering/protection in those slots.
Additional wincons in the main aren't really necessary, as we're situated well against a decent number of decks game one. The board is where you want them.
Paradoxicals are for BGx decks/decks where they have a high concentration of interactive cards and a slow gameplan. The idea is that it allows you to get back in the game (and often win straight up) off of it if they've dealt with your starting engines, drawing 5-7 cards at the opponent's end step and then going off on your next turn.
Boarding depends on the matchups, but generally the flex slots (protection/filters) are the first to be boarded out. Don't board out any engines - they're what make the deck work, why would you board one out?
Also small note on the manabase - Glimmervoids can be a little awkward with the deck (especially if you hit the two glimmervoid opener, in which case you have to waste an artifact to play your land. The value of it also isn't really needed in the deck, since we only care about red at the very end. I'd recommend switching them over to more Seachrome Coasts or Fountains.
An engine is necessary but not sufficient for an opener. I would probably not keep a 5-land, engine, 1 random card hand. The 4-land hand definitely depends on what the two cards are.
As for the two-engine brick, it's rare but does happen. Was SV one of the last two cards you drew when you bricked? If not casting it mid-combo is usually better to help continue the chain/dig for the Retract/Opal piece you need to finish the combo.
One engine is certainly trickier, since as is the case with the normal win drawing your whole deck with just one engine isn't all that likely, but so long as the following are true I believe you should be able to loop regardless of how many cards are left in your deck:
1. You have 4 mana available (including Opals) to cast/activate Elixir and play Retract
2. You have at least as many Cheerios in your hand + on the battlefield as you do cards in your deck + graveyard.
3. You still have a Retract in hand (or Noxious giving you one).
At that point, you can play Retract, play Elixir, activate Elixir, play enough Cheerios to draw the deck, play Opals for mana (Noxiousing one back and playing it again to generate a net mana in the loop), and repeat. After the first repetition, the board state should be the same as the end of the first cycle (only Opals + NR + Retract in yard, no cards in deck, Elixir in play, storm count increases and you've generated a mana) and you can thus show it's a loop to shortcut (and since we're drawing the whole deck every time and not dependent on its order the randomness of shuffling shouldn't be an issue in demonstrating the loop like it is for Four Horsemen).
If instead of a Retract, you have an Elixir, you can try to go for it and hope to draw a Retract after the shuffle with your remaining Cheerios in hand, but there's definitely the fizzle possibility there.
kirtash is correct here that Elixir allows us to beat the Abzan combo, though not for the reason they seem to think (infinite damage does not automatically beat infinite life - it depends on the circumstances of how the life is gained, as well as who is the active player in some cases). We certainly can't win with a single infinitely large Grapeshot while they have the combo out - they just gain more life than the number we give in response. However, while we're generating the infinitely large Grapeshot, we can first kill a piece of the combo and stop them from being able to change their life total any further. Once that's done, we demonstrate and then shortcut the loop to make the storm count large enough to kill them. The mana for casting multiple Grapeshots while spending three for the Elixir loop shouldn't be an issue if you have Noxious in the deck - you can recur Opal each loop to generate a mana, so you can demonstrate infinite mana (as well as life and damage).
With all that said, I'm still not convinced this is necessarily a great option for Abzan combo. Keep in mind that once they do have the combo online, not only can they gain infinite life but they also get to choose what card they draw next due to the infinite scry 1, so they'll easily be able to find all their hate/most effective cards. You really only have the one turn after they go off then to try and go off
At this point, I'd really like to start a larger discussion about sideboarding, since that seems to be the most nebulous thing right now (I know my sideboard feels very loose right now, and I'm sure I'm not the only one). Instead of trying to come up with 15 cards, I think the better method is identifying what we need for each matchup, and subsequently figuring out how many cards we want for those matchups. then we can try to consolidate. As an example:
Control: This matchup hinges on settling a threat through countermagic and holding it through the combo. Silence seems strongest here, as it allows you to hold the whole turn, but other 1 CMC countermagic/protection could be considered. Alternate threats which dodge their plans (such as Geist) are also options. Recommended cards: Silence, Geist, cheap countermagic (Dispel?).
Burn: Post-board, they're bringing in more spells which can hit creatures, as well as Eidolons if they're not already main. Protecting the combo is important, but dealing with Eidolon is critical as we actually can't win with it in play. Recommended cards: Echoing Truth, Path to Exile, Leyline?
BGx: Post-board, they're pretty much guaranteed to have either hand disruption or removal in their opener. We have two routes here: either go for a critical mass of threats to try and top their 1 for 1 spells or board in spells which put you back in the game at a moment's notice. Recommended board cards: Monastery Mentor (for the former plan), Paradoxical Outcome (for the latter plan), Silence (for both), Leyline?
Other decks: To be frank, in two weeks of live tournament play with the deck I haven't lost a match against any deck that wasn't one of the three archetypes above, and in almost all cases the matches against non-interactive decks were straight up rolls so boarding didn't matter too much. If we actually want to go into detail, though: if they're doing their own thing (Tron, Merfolk, Infect, Affinity, etc.) and can't interact, we're generally just faster than them. Expect some spot removal to come in, but generally they won't have too much against us. Recommended cards: Hurkyl's Recall (to help ensure we don't fizzle, especially since you'll likely have the extra turn where you have sufficient mana for it).
I also strongly recommend boarding in at least one Echoing Truth in most cases. Between Eidolons, Leylines, Chalices, and other random things, it's almost always worth bringing one in to hedge.
Curious what other cards people like for specific matchups (most notably for the three mentioned above, as those are the ones we need the most targeted cards for).
This is what Paradoxical Outcome is trying to do for the BGx matchups. Given there isn't always an engine in the yard (due to Path), being able to draw a lot of cards off Paradoxical seems slightly better to me (since usually at the point where you'd hit that SB slot you're just holding a bunch of Cheerios and waiting for something).
Observations/Notes from today:
Once again, Geist of Saint Traft out of the board was an MVP vs the control decks when I expected they boarded out sweepers (especially when you get it down on T2 with Opal and then suit up with a Shield to stop Snapcaster blocks). Highly considering going up to two.
The Sram,Paladin,Opal,Land,3xCheerio hand came up quite a bit today (at least 5 times), and it's certainly something worth discussing. I snap kept it on the draw against the non-interactive decks like Infect and Elves, but in the dark or against other decks it's much closer. On the play in the dark is also an interesting discussion since you can dodge bolt with the T1 Paladin if you have a Shield Cheerio.
Mentor suffers from the same issues that Sram and Paladin do - it dies to just about all the same cards. At that point, you're just playing a more expensive card with the same risk, which doesn't really do much for us. Cards like Geist, Bastion Inventor, and Ghirapur Aether Grid, however, all attack from a different angle, and usually dodge the answers your opponent brought in.
List is as follows (sideboard is still somewhat rough and will almost certainly be a few cards different by Thursday):
4 Puresteel Paladin
4 Sram, Senior Edificer
Artifacts: 25
4 Mox Opal
4 Accorder's Shield
4 Cathar's Shield
3 Kite Shield
2 Spidersilk Net
4 Bone Saw
4 Paradise Mantle
4 Serum Visions
2 Grapeshot
1 Noxious Revival
2 Swan Song
Lands: 14
4 Seachrome Coast
3 Hallowed Fountain
1 Sacred Foundry
1 Plains
3 Flooded Strand
2 Arid Mesa
4 Silence
3 Echoing Truth
3 Path to Exile
2 Paradoxical Outcome
1 Wear // Tear
1 Geist of Saint Traft
1 Hurkyl's Recall
Notable takeaways from today:
In the "alternate threat" sideboard slot I tried Geist today, and I sided it in vs. control figuring they might have boarded out Verdicts in favor of cheaper single-target spells (they did). Played it turn two on the play with Opal and rode to victory after equipping with a shield equipment to survive Snap blocks. Going to try it again next event and try to see how I feel about it in the long run, but I'm pretty happy with it so far over other options. Might go up to two, we'll see.
I'm also slightly warming up to the Burn matchup. Eidolon is obviously very rough post-board, but the matchup certainly doesn't feel unwinnable, especially if you dodge the Eidolon (certainly the matchup is skewed against us but so far all of my games have been such that the person on the play wins and the loser is one turn off, which gives me hope in the matchup if you win game one).
I also highly recommend actually representing as UWR/Esper control until you go for it game one (or it otherwise gets revealed). Hollywood them when they're casting their turn two/three spells and pretend you're thinking about Leaking/some other counter on their spells and try to bait them into the tap-out so you can combo without fear (especially against Burn which might not play double spell turn two unless you help guide them into it).
Still debating land count (my current main is the same as Shinx's but -1 Sleight, +1 Cheerio), and I may switch that 21st equip to another fetch since I have been mulliganing a lot due to lands (and keeping some sketchy one-land with SV on the draw as a result).