As someone who is more or less disinterested (but not uninterested) in this thread from a magic perspective, but interested because it has cross-format appeal to any game where play order matters (in particular, I'm thinking video games which alternate between double blind picks and counter-picks), I'm curious to hear posters' views. Note the scenarios aren't mutually exclusive.
What is the actual problem you're tying to solve?
A: Game advantage - You're trying to remove any gameplay advantage from going first (in any given single game). If this is the case, then the only solution is to change the game's rules, like scry 1 on the draw, lands enter tapped on the play, non-equal life totals, etc.
B: Tournament advantage - You're trying to remove long-term advantage on average (across say, a tournament or multiple tournaments). If this is your goal, the first thing to do is to recognise that a pure random method achieves this on average. What people are suggesting is methods of reducing variance (so the average is achieved more reliably). However, the problem with variance-reduction strategies is usually (in addition to logistics) that they merely shift the issue. Consider a tournament where each player is guaranteed to alternate exactly between play and draw. Even though both players get the same number of plays and draws, the person going first in the grand final gets an advantage compared to the person going first in round 1. The problem still exists, but in a different form.
Now to throw additional problems in the mix, consider the above circumstance, but where intentional draws are a thing: anyone who gets assigned to "draw" on the intentional draw round gets a long-run advantage by getting an extra "play" in the remaining games. It gets worse if you can actually manipulate the order of play vs draw by throwing games.
C: Set Advantage - You're trying to eliminate advantage across a given set (e.g. best of three). Here's where things get interesting in my opinion, and the problem with the above scenario becomes much more apparent: Should a Bo3 go:
random, looser picks, looser picks;
random, looser picks, random;
Play-draw-play (G1 randomly assigned);
Play-draw-draw (G1 randomly assigned);
random,random,random;
random,random,opposite if G1 and 2 are the same?
Once you consider sideboarding, things get more difficult again- the deck used in game 1 is not the same deck as the deck used in game 2, and if a player wins G1 and G2, there is no G3. Under 'looser picks', winning G1 guarantees the play in G3, which is an 'extra advantage' compared to winning G2. But G1 is a non-sideboarded game. Is winning without sideboarding (and importantly: without being subject to sideboard hate) worthy of this 'extra advantage'? Does this 'favour' decks prone to sideboard hate, like affinity and dredge?
In terms of evaluation of Claim // Fame my thoughts are:
It is the only 1-mana spell that 'protects' against a T1 thoughtseize. For that reason alone I'd play it. In this respect, it is a more powerful postmortem lunge, which was already playable, and compares favorably with (but is less flexible than) noxious revival.
The main problem, as with all protection spells, is actually drawing the paladin and the protection spell at the same time. Claim earns some merit here since you don't have to draw it before you draw the paladin, in contrast to most protection, but it still suffers from not helping us filter the deck (in contrast to, say, remand).
In theory, Claim could also be used to help filter if you use it to loop snapcaster mage and cantrips. I personally don't own snappy, and I am not investing hundreds of dollars to experiment with it, but I could see it working.
As a downside, it does not protect against path to exile, nor does it work if leyline of the void or rest in peace, etc. is thrown at us, but that's kinda narrow, and I don't see it doing any real harm in any practical sense.
Fame is gravy
@Lunicyl:
I dabbled with thopter sword for a very short time. I found I simply could not actually assemble the combo. I tried running Whir of invention, but the triple blue was a problem. Also, even when I did assemble it, I didn't find that it actually won me the game, either due to artifact hate, or because I didn't have enough mana to sink into the combo to generate enough tokens/life to pull back from the inevitably bad board state I was in after spending two turns setting it up. That's not to say its a bad combo, and I wouldn't discourage it. I just personally had no success with it.
I've been away, so not testing or playing, but based on the posts in this thread, I'd be extremely comfortable saying that the answer is jamming more threats. Serum visions is simply not good enough to help against DS. It neither digs deep enough, nor does it give actual card advantage to combat hand shredding.
Mentor is the obvious choice, but I think the 3-cost is a dealbreaker personally. Personally I would use maverick thopterist, but I'm partial to the guy so this may be my bias than a real answer. bastion inventor feels kind of irrelevant in a board stuffed with big butt goyfs and shadows. I also strongly think that real card draw is the go-to answer vs discard: namely, thoughtcast and reverse engineer. This does slow down the deck and potentially interrupt the combo though, but if we can't win without them its better to go with a bad plan B than a terrible plan A.
i don t like maverick. too inconsistent.
you create a bunch of little guys... and then? they still will be dead by tarmo
what is you suggest?
I don't find it inconsistent at all, but it does need to be supported in the right build (colours of mana, and you want to be digging pretty hard since thopters don't last long if your plan is to block).
Option 1: block till you find your combo engine then combo off
Option 2: run bone saw and sigil of distinction and a few golem-skin gauntlets and beat face with vigilant fliers. Note that the thopters are good with paradise mantle for additional mana.
It does run a slow clock and while my current testing is showing wins, it is showing wins by 1 turn faster than the opponent. If my opponents are actually misplaying (quite probable) then it could be that this is simply not viable.
Glint Hawk and Golem-skin Gauntlets is worth a try if you're lookin for more threats for beat down mode, but against Lili unless you have haste or generate multiple creatures in one play it's rough.
Glint hawk sits in that awkward spot of being really terrific value for money, but not actually being a threat: i.e. it is not a good topdeck. The substantial value-added with storm entity or monastery swiftspear, erayo, etc. make it pretty great in some builds.
On the one hand, I do think cantrips are the way to go in G1. It fits the format of every other decent Ux combo deck to-date, and is currently the most successful Cheeri0s version we've seen in our admittedly limited sample. It's also the clearest variance reducer at our disposal. This doesn't include a singleton (or maybe double) Riddlesmith. Smith is actually an engine that helps us win games. He's also better than Paladin against Chalice (cast trigger), and triggers off Opals. I think adding 1-2 might be correct, even at the expense of some number of SV/NR/GS.
On the other hand, I do think the idea of adding other threats (Inventor, Thopterist, Geist, Grid, etc.) is important in G2-G3, depending on the matchup. This is where my issue of sub-optimal sideboards comes back. I think most SBs are just really awkward right now and we need to figure out how to build them better. This will necessarily include deciding on threats vs. protection, and what particular cards of each to run.
I'm thinking a transformative sideboard is the way to go, with, say 10-12 cards dedicated to the really bad matchups. I don't like the idea of silver bulleting; we have enough trouble sculpting hands as-is.
Great seeing the reports! Thanks for the updates My cards have been lost in the mail, so no physical games for me, but I've done quite a lot of testing on the computer, and checking the results with the tourney results being posted, I am fairly confident in the following analysis:
The only way to reduce variance and improve the bad matchups in this deck is to increase the threat count. How you do that is up to you, there are multitudes of builds, my general idea is use threat to kill them, and combo off as a backup plan while they try and deal with the threat. The key here is to pick threats that synergise and don't cannibalise your hand. Have to be very careful on that last point.
Cantrips and protection can't solve all the deck's issues. Cantrips simply aren't good enough- you either run a small number and then have a chance of not drawing them, or you run a large number and you gaurantee drawing them, which fizzles the combo plan just as often as it enables it. You hit a steady state of still-not-quite-being-low-enough-variance. Likewise, protection is "do nothing" on its own, so while it helps sometimes, equally as often it just doesn't do anything to advance your gameplan.
This is why I advocate for upping the threat count (while also keeping a small to moderate number of cantrips). Drawing the right threat forces your opponent to make plays and that alone makes it superior to cantrips or protection.
What I think needs more of in this thread is analysis of backup plan threats, coupled with "downside" analysis. Not upside analysis.
For example, assault formation is a bad idea: it does nothing unless you have creatures, and doesn't help you get, or protect, your creatures. And if you're only looking at the upside case where you already have creatures, you can already combo to win. As such, the card is lose-lose in the majority of cases. Compare doran, the siege tower. He is a creature, does the same effect, so is generally better in most cases than assault formation since it doesn't require anything else to be imposing. I.e. makes a reasonable topdeck. Dies to removal? Sure, but so what, assault formation doesn't work if the opponent has removal either. Doran's colours are bad? True, but there are some very highly synergistic cards in that colour combination, including ayli, eternal pilgrim and time of need. I've done builds like this in the past, but I'm personally not invested in Doran, so it's not where I prefer to sink my time.
Another example: storm entity and monastery swiftspear work very well T1 or T2 -> enraged giant. Good plan right? Maybe, the problem is that enraged giant -> swiftspear/storm entity is very bad over 2 turns, so you can randomly loose to ordering.
Another example: bastion inventor avoids removal, but as I've said before, and ktkenshinx repeats correctly above, guy gets chumped, double blocked, dies to liliana. Doesn't really help our jund matchup. Geist of saint traft falls into the same category (slightly better on the offensive, but worse on the mana requirements).
My personal preference *at the moment* (and subject to more testing) is to run maverick thopterist vs burn and jund, alongside thoughtcast and reverse engineer (the latter in low numbers). It kinda works, but I'm cutting it close in testing and I'm not sure if my opponent is playing optimally since I can't see their hand. Also does not work very well vs aggro decks, since the thopterist is a slow clock, and I'm not sure how transformative my sideboard can be.
I found the same thing with bastion inventor vs jund- Liliana is a pain to get through. maverick thopterist however works quite well in the right build, as it puts 3 bodies for 1 card which can apply pressure with bone saw and flying. Its slow, and it doesn't look fantastic on paper, but in practice I found it helps. Its best thing is that its usually never terrible. i.e. worst case it is less dead than alot of other cards, like leyline. Have to beware though that it is a nonbo with retract.
Has anyone tried Enraged Giant?, I'm playing an aggro version and I've been missing a haste effect like Mass Hysteria. Think its good or should I just get the enchantment?
My initial reaction to this was that bastion inventor was better than enraged giant as removal is out main weakness. But on more testing, I'm hitting a stumbling block in liliana of the veil which means that non-evasive hexproof has some weaknesses. So I did some very brief testing with enraged giant, my thoughts were that it is OK, but not stellar. Then I went back and jammed in other haste creatures, in monastery swiftspear and storm entity. threw in glint hawk for extra storm and beats. Deck worked pretty well. Had a niggle of consistency issues since sequencing is very important, and it really wants to draw a mox, but it poured out massive damage when it worked.
Bit awkward on the mana since it requires Green, but has Time of Need been looked into? It effectively runs as copies 9-10 (I don't think I'd run more than 2) of the engine, though it does necessitate the extra turn (but that still leaves T3 combo open so I don't think it's too terrible).
This sounds like a good idea to me. I don't think a single green splash is too hard on the mana base (provided you ease up on the blue, I wouldn't run green and serum visions, for example), particularly since green has plenty of mana-fixing cards available.
@slax01 @kirtash900 What do you mean my deck is less consistent than others? Could you explain what mine is lacking compared to others?
Mostly covered already. The land count is fine, my main thought is that it seems like you'd have trouble drawing both the paladin, the lands and the protection all together in enough turns to win against something with a fast clock, like burn or jund. That said I don't have advice on how to change, as I don't think there is any settled lists yet to refer to, I'd leave it to your testing and personal preference. For example, ktkenshinx prefers more cantrips, while I prefer more threats. Depends on the gameplan and your willingness to take risks - note also that theoretical consistency against bad matchups like jund and burn isn't necessarily something to be afraid of, it might be better to take a gamble with those games and win for sure vs decks with slower clock, possibly while relying on the sideboard more heavily, than to try and optimise against specific bad matchup decks.
Something I have not seen discussed, though I may have just missed it, is running actual protection cards like apostle's blessing.
Is that something that could be useful in countering Abrupt Decay and the like cards or is that to narrow for this list?
EDIT: reply to this dropped off; adding in.
Protection is fine, some people prefer more flexible cards which can serve dual roles, like silence and swan song (I prefer silence) since they also interrupt the opponent's plan. That said, protection is a perfectly reasonable way to go, though I would recommend gods willing as the better choice.
@shadowgripper: great to see some tourney results being posted, love detailed matchup reports and sideboard analysis, good work
My thoughts: I am surprised that a deck with only 3 serum visions 1 nox didn't brick after the first guy got removed: or get mulliganed to death. From a closer look I will say that it looks like you got a tad lucky in a few of those matches (not that luck is a bad thing, this deck will always have luck involved), with some good topdecks. I suspect/theorise that larger samples could reveal a bit of a consistency issue, it will be good to track how you go in future. It also appears that the matchups posted other than Jund are highly favourable (possible exception of lantern, sounds like your opponent misplayed? I'm not sure, not familiar enough with the interactions with our deck/boardstates) because we are too fast for opposing combo or the opponent's clock was too slow even if they have control/hate.
@Wesste: I think all three of those are viable sideboards. I personally would lean to sideboard #3 but -geist and +bastion inventor, but then again I'd just have bastion inventor in the main.
After having the chance to test, I think I've settled on a non-combo version of the deck as my preference, since it seems to have the most overall resilience by relying on hexproof and huge toughness to beat removal and win combat, and sheer card draw to scuplt the hand and combat discard. I've gone to golem-skin gauntlets as my wincon, rather than grapeshot, since in this mana-heavy version I'm drawing over multiple turns, rather than all on one turn. It is obviously much much slower than the combo version, but has a degree of inevitability that the combo doesn't have. Still plenty of refining to do (a sideboard and fixing the mana base) but I'm pretty happy with the build. The sideboard could be particularly interesting as it is quite possible to convert to combo for game 2 (or conversely, to run combo maindeck and convert to this postboard).
Sorry, I wrote the earlier post in a bit of a hurry so I didn't really say what I meant to:
Re: Stony silence:
What I meant to say was that I don't see this as being a particularly bad (good?) hate card, i.e. not one that really hoses us. The equip-blocking makes bolt a smidge better, but is otherwise inconsequential, except in aggro builds. The mana block from mox is a bit of an issue, but doesn't really cause too many problems if you think about it: A T2 Stony Silence means they're tapped out- this means we drop a paladin and draw a bunch of cards. Even if we don't combo off, that can give us fuel and sculpt our hand to go off next turn, not necessarily a loss. Note also this means that they have to remove the paladin in their turn, reducing their T3 plays. If they T3 s.silence, holding up removal, that means we can T2 combo, but also means that they've slowed themselves down when they could've been doing something productive on T3- again, not necessarily a bad outcome. And if you consider that we have up to 3 or 4 lands to play with, this isn't a terrible outcome either. Now to go a step further, consider what boarding in Stony Silence means- what have they taken out? Did they mulligan to get it? both of these can be a positive for us as we die to simpler strategies than stony silence. Put another way: stony silence is among the hate cards that people can throw at us, but is almost certainly at the low-end of the hate list. Compare Eidolon or chalice, which simply shuts us down the moment it hits the board. In any event, I'm not telling people not to board in cards to take out stony silence, since it doesn't require a taxing sideboard (and since the paladin draw engine still works, so we have good odds of actually getting the silver bullet), just curious that so many people are throwing it up as a hate card.
Re: gemstone mine (not caverns!)
In an ideal/theoretical world we won't tap in three times, but in practice it happens with sufficient regularity to be wary of it. At least as much is my experience. Put simply, we want to tap out every turn. Not an issue if we win by T3, which does happen (alot) but not always.
I've been testing vs. Burn (about 25 games done, 50/50 overall) and the cantrip version is just fine at handling removal. Eidolon is still a nightmare, but removal is manageable. If the % dips too heavily in later tests, especially against Jund/Abzan, I think we have flex to go -2 cantrips and +2 protection in the main. You have a surprising degree of resilience with 8 engines, 1 NR, and 7 cantrips to dig for more engines, so I'm not worried too much.
My testing was vs Jund (old builds, not sure what the new fatal push meta will look like, but probably even worse). Personally, I found jund unwinnable. Between discard and removal, there was no way to consistently win. Random wins? Sure, but consistent wins? None. Like 80-20 or some such nonsense. Burn was easily winnable with the right sideboard, although the matchup was much more borderline preboard in the pure cantrip version, I would say less than 50-50, but it depends on the opponent's list, as some burn is actually not necessarily that removal heavy.
There are some things in this thread which are confusing me:
Why are people concerned about stony silence? What games do we actually win by equipping, especially given that none of the decks above are running aggro cards? Is it really hurting the moxes *that* badly?
Why do people run leylines? While I accept that they protect against discard, they are card disadvantage, can't be reliably cast, and don't offer on-board protection. Has anyone done the math on how many games you'll actually draw leyline and paladin in the opening hand?
Re: Gemstone caverns: from past testing I've played many games where I've depleted these lands (cantrip, paladin, cantrip). Noxious revival gets them back, sure, but that's some bad card disadvantage.
Anyway, I've got some initial testing under my belt over the weekend, and thoughts as follows:
The cantrip-heavy build is reliable and fast. However, I can't see how it can also sport protection without becoming too dilute. Basically, I see it as a sideboardless deck. Its success will rise and fall with the amount of removal in the meta. In a removal heavy-meta (and vs opponents who know what they are doing), it should never, or rarely, win. Vs Aggro and combo, it will perform fine.
I am currently focusing on an aggro build. This takes me away from combo so I won't post the list here, as it risks turning into a different deck entirely (improvise midrange with a side of combo), but there's a heck of a lot more testing I need to do first before I settle on a build. All I do know is that bastion inventor and spirit link work wonders vs burn.
The deck has to be at least 3 colours. Blue for retract and cantrips if there used. White for the paladins and red for grapeshot. The mana from grapeshot normally comes off mox opals. I'm not sure how we would benefit from black or blue. With black we have hand disruption and green possibly assualt formation as a different win con but we don't really need either as goblin gaveleer and mass hysteria is good enough as a different win con.
Green is probably unnecessary but the argument for adding black to go Esper would be to slow down the deck and make it more control-y rather than super explosive by adding stuff like hand disruption, Push, and creatures like the boltproof scrying aetherborn. Since it's in-color, it's also possible to supplement Sram+Paladin with Riddlesmith for a third draw effect, and compensate the discard with Ovalchase Daredevil...
I don't think anybody's ever given this build serious consideration, however. It's possible but I'm not sure if it would be better or just different. Some testing would be required.
While the conclusion that the deck is WU is ultimately correct, I feel these posts don't justify the testing that's gone on in the past, or the considerations for each colour splash. Here's a detailed analysis of the colour issues:
Prelim: our primary concern with splashing colours is the number of lands in the deck. The core engine needs only two lands to go off, and consumes both lands to cast puresteel paladin. Since it is a very safe assumption that you're comboing the same turn paladin hits the board, you're realistically only going to have 1 spare mana up at any given time (the land on T1). Sometimes you will have an extra land on T3, but not always. The mana gained through mox opal is unreliable, but also typically needs to be held up midcombo (i.e. to cast retract). This means most of the analysis for a colour splash depends on what 1-cost cards are in that colour which support the combo, or what 3-cost plan B cards in that colour.
1: White: obvious; Puresteel Paladin costs WW. This means that white is essential to all builds. With fetches, we can reliably go tri-colour, with two Wx shocklands. Quad-colour can raise the question of whether to use fetches or rather rainbow lands like mana confluence, etc. The new spire of industry adds to that option.
2: Blue: retract costs blue, so we want blue. It isn't safe to assume that retract will always be cast off a mox- often you can only dig a couple cards deep before needing to fire off retract, so you may not have drawn a mox yet. Adding blue lands therefore adds to the consistency of the deck. Add to this that blue has cantrips, which, while not the best tutors around, are the easiest to slot in, and the case for blue is strong. Blue also has some backup plans by offering protection and (historically) riddlesmith. Because of this, blue can tend to crowd out the other colours. That doesn't mean the deck has to be built blue however, and significant consideration has been put into the other three colours, discussed below. The general consensus is that blue is superior, predominantly because the tutor effects in those colours are not strong enough to warrant diverging from blue.
3: Red: grapeshot costs red, however, grapeshot is typically cast off mox opal, so that's not the main reason red can be splashed. Instead, the main reason is that red has the best "plan B"s - i.e. alternate wincons outside of white. Examples are in the primer, but include ghirapur aether grid, goblin gaveleer, Jeskai ascendancy, the list is long. So overall, red is a very flexible colour choice that fits in with many different builds. However, red is the worst colour for tutoring out paladin, and doesn't protect paladin. As such, red considerations are typically about putting in alternative threats, rather than supporting the main plan. With Sram, the main plan got much more consistent, which is why you've seen the degree of red splash reduce in most people's lists.
4: Green: The main reason people don't run green is that a 4-colour build can't be reliably sustained on fetches and shocks. This colour cost is seen to outweigh the additional benefits green can bring. Specifically, green has the best 1-drop creature dig, in commune with nature. Other green options are of a similar ilk (e.g. oath of nissa), but the marginal benefit compared to, say serum visions is not huge, so rather than risking the extra colour, most people choose not to splash green. Also, green does not give us a sound "plan B". If it did, I feel that a green splash would be more than justified. By way of example, cards like managorger hydra are inferior to monastery mentor. Green's protection cards are also not superior to white's protection cards. Overall, green isn't a bad colour, but doesn't do much the other colours do. Assault formation is bad in this deck since we can't get creatures to stick, so it just does nothing most of the time. Unlike red's Plan B, which fit in naturally, this requires a build-around, which means a substantial departure from the combo. Green also has some midrange options, like congragation at dawn, but that's considered too slow.
5: Black. Black contains spoils of the vault, so was considered extensively. However, it's generally accepted that the risk of just randomly loosing a game isn't worth it. Other than that, black doesn't have protection and doesn't really dig. In terms of the Plan B's that are possible with black, there are some (painsmith, glaze fiend) but they are typically outclassed by red, or require an extensive build-around. ovalchase daredevil was experimented with a number of players, but that doesn't count as a black splash since we never cast it. Midrange options are relatively good in black, with the new fatal push and battle at the bridge helping us out, but that begs the question- why are we trying to prolong the game? That's not to say this option can't be good, but it requires dedicated deckbuilding.
6:multi-colour: I once dabbled with signal the clans I think most people agree that its not worth the two extra colours, but just thought I'd put it out there, as the only reliable 2-cost tutor.
On the 17th of February there will be two modern Grand Prix where the deck could have it´s break out. If so, i hope it doesn't escalate too much because in the B&R-announcement they said that 5 weeks after the tour there will be another.
Darn, I just realised I will be travelling at the time of the GP so I can't take this deck for a spin; I would've loved to run this at the Brissie GP. Alas, no more modern GPS in AU for the rest of the year...
Hey, I have recently played around with this deck and decided to get a bit more serious when Sram was spoiled. I wanted to ask about the general concensus on Paradoxical Outcome. Currently I play with 2 Outcomes and no Cantrips, hoping to either draw one engine card in my starting hand or playing a huge Outcome in my opponents EOT, trying to find one. Maybe I am too reckless after the boost in consistency offered by having 4 additional Puresteel Paladins, but I am not sure if this deck should be anything but greedy.
I side the Outcomes out for Monastery Mentor in creature-heavy matchups, but otherwise I thought they are still pretty good.
My analysis of Outcome is as follows:
A - 4 Mana is just a huge problem. I found I could cast it reliably with, and only with, lotus bloom or ritual shenanigans, but that diluted the deck and in any event, the bloom/rituals reduced the number of equipment I had in hand to bounce. Fun, but neither consistent nor reliable. Even if you do cast it, finding additional mana to then cast grapeshot or paladin is impossible. I found that it was much more viable with aetherflux reservoir, since you can outcome after reservoir, rather than the other way around, but the 4 mana on that too makes it clunky to say the least. And prone to artifact hate. reshape and whir of invention or the expertises *could* make this a better strategy, but it very much risks over-complicating the deck. Likewise, in a ritual-heavy version, empty the warrens becomes fantastic, since you need a minimal storm to win. But this is just not reliable, since it really only works if you have an optimal opening hand.
B - You don't really want to cast outcome quickly anyway, even if you can cheat it out. If you cast early, you're going to draw, say a max of 7 cards? Usually 3-4? That often isn't enough to win the game.
C - Dead mid-combo, so you don't want to draw it late either.
That said, there's really only one place for Outcome to do any work: if you combo and fizzle. If this is happening to you, outcome can be a viable way to recover a game. So my short answer is Outcome is viable as a one-or-max-two-of as a recovery tool in a deck with relatively high land count. While it could potentially be breakable, you'd have to warp your entire deck around the card, and I'm not sure that that is desirable.
What is the actual problem you're tying to solve?
A: Game advantage - You're trying to remove any gameplay advantage from going first (in any given single game). If this is the case, then the only solution is to change the game's rules, like scry 1 on the draw, lands enter tapped on the play, non-equal life totals, etc.
B: Tournament advantage - You're trying to remove long-term advantage on average (across say, a tournament or multiple tournaments). If this is your goal, the first thing to do is to recognise that a pure random method achieves this on average. What people are suggesting is methods of reducing variance (so the average is achieved more reliably). However, the problem with variance-reduction strategies is usually (in addition to logistics) that they merely shift the issue. Consider a tournament where each player is guaranteed to alternate exactly between play and draw. Even though both players get the same number of plays and draws, the person going first in the grand final gets an advantage compared to the person going first in round 1. The problem still exists, but in a different form.
Now to throw additional problems in the mix, consider the above circumstance, but where intentional draws are a thing: anyone who gets assigned to "draw" on the intentional draw round gets a long-run advantage by getting an extra "play" in the remaining games. It gets worse if you can actually manipulate the order of play vs draw by throwing games.
C: Set Advantage - You're trying to eliminate advantage across a given set (e.g. best of three). Here's where things get interesting in my opinion, and the problem with the above scenario becomes much more apparent: Should a Bo3 go:
random, looser picks, looser picks;
random, looser picks, random;
Play-draw-play (G1 randomly assigned);
Play-draw-draw (G1 randomly assigned);
random,random,random;
random,random,opposite if G1 and 2 are the same?
Once you consider sideboarding, things get more difficult again- the deck used in game 1 is not the same deck as the deck used in game 2, and if a player wins G1 and G2, there is no G3. Under 'looser picks', winning G1 guarantees the play in G3, which is an 'extra advantage' compared to winning G2. But G1 is a non-sideboarded game. Is winning without sideboarding (and importantly: without being subject to sideboard hate) worthy of this 'extra advantage'? Does this 'favour' decks prone to sideboard hate, like affinity and dredge?
It is the only 1-mana spell that 'protects' against a T1 thoughtseize. For that reason alone I'd play it. In this respect, it is a more powerful postmortem lunge, which was already playable, and compares favorably with (but is less flexible than) noxious revival.
The main problem, as with all protection spells, is actually drawing the paladin and the protection spell at the same time. Claim earns some merit here since you don't have to draw it before you draw the paladin, in contrast to most protection, but it still suffers from not helping us filter the deck (in contrast to, say, remand).
In theory, Claim could also be used to help filter if you use it to loop snapcaster mage and cantrips. I personally don't own snappy, and I am not investing hundreds of dollars to experiment with it, but I could see it working.
As a downside, it does not protect against path to exile, nor does it work if leyline of the void or rest in peace, etc. is thrown at us, but that's kinda narrow, and I don't see it doing any real harm in any practical sense.
Fame is gravy
@Lunicyl:
I dabbled with thopter sword for a very short time. I found I simply could not actually assemble the combo. I tried running Whir of invention, but the triple blue was a problem. Also, even when I did assemble it, I didn't find that it actually won me the game, either due to artifact hate, or because I didn't have enough mana to sink into the combo to generate enough tokens/life to pull back from the inevitably bad board state I was in after spending two turns setting it up. That's not to say its a bad combo, and I wouldn't discourage it. I just personally had no success with it.
Mentor is the obvious choice, but I think the 3-cost is a dealbreaker personally. Personally I would use maverick thopterist, but I'm partial to the guy so this may be my bias than a real answer. bastion inventor feels kind of irrelevant in a board stuffed with big butt goyfs and shadows. I also strongly think that real card draw is the go-to answer vs discard: namely, thoughtcast and reverse engineer. This does slow down the deck and potentially interrupt the combo though, but if we can't win without them its better to go with a bad plan B than a terrible plan A.
I don't find it inconsistent at all, but it does need to be supported in the right build (colours of mana, and you want to be digging pretty hard since thopters don't last long if your plan is to block).
Option 1: block till you find your combo engine then combo off
Option 2: run bone saw and sigil of distinction and a few golem-skin gauntlets and beat face with vigilant fliers. Note that the thopters are good with paradise mantle for additional mana.
It does run a slow clock and while my current testing is showing wins, it is showing wins by 1 turn faster than the opponent. If my opponents are actually misplaying (quite probable) then it could be that this is simply not viable.
Glint hawk sits in that awkward spot of being really terrific value for money, but not actually being a threat: i.e. it is not a good topdeck. The substantial value-added with storm entity or monastery swiftspear, erayo, etc. make it pretty great in some builds.
I'm thinking a transformative sideboard is the way to go, with, say 10-12 cards dedicated to the really bad matchups. I don't like the idea of silver bulleting; we have enough trouble sculpting hands as-is.
The only way to reduce variance and improve the bad matchups in this deck is to increase the threat count. How you do that is up to you, there are multitudes of builds, my general idea is use threat to kill them, and combo off as a backup plan while they try and deal with the threat. The key here is to pick threats that synergise and don't cannibalise your hand. Have to be very careful on that last point.
Cantrips and protection can't solve all the deck's issues. Cantrips simply aren't good enough- you either run a small number and then have a chance of not drawing them, or you run a large number and you gaurantee drawing them, which fizzles the combo plan just as often as it enables it. You hit a steady state of still-not-quite-being-low-enough-variance. Likewise, protection is "do nothing" on its own, so while it helps sometimes, equally as often it just doesn't do anything to advance your gameplan.
This is why I advocate for upping the threat count (while also keeping a small to moderate number of cantrips). Drawing the right threat forces your opponent to make plays and that alone makes it superior to cantrips or protection.
What I think needs more of in this thread is analysis of backup plan threats, coupled with "downside" analysis. Not upside analysis.
For example, assault formation is a bad idea: it does nothing unless you have creatures, and doesn't help you get, or protect, your creatures. And if you're only looking at the upside case where you already have creatures, you can already combo to win. As such, the card is lose-lose in the majority of cases. Compare doran, the siege tower. He is a creature, does the same effect, so is generally better in most cases than assault formation since it doesn't require anything else to be imposing. I.e. makes a reasonable topdeck. Dies to removal? Sure, but so what, assault formation doesn't work if the opponent has removal either. Doran's colours are bad? True, but there are some very highly synergistic cards in that colour combination, including ayli, eternal pilgrim and time of need. I've done builds like this in the past, but I'm personally not invested in Doran, so it's not where I prefer to sink my time.
Another example: storm entity and monastery swiftspear work very well T1 or T2 -> enraged giant. Good plan right? Maybe, the problem is that enraged giant -> swiftspear/storm entity is very bad over 2 turns, so you can randomly loose to ordering.
Another example: bastion inventor avoids removal, but as I've said before, and ktkenshinx repeats correctly above, guy gets chumped, double blocked, dies to liliana. Doesn't really help our jund matchup. Geist of saint traft falls into the same category (slightly better on the offensive, but worse on the mana requirements).
My personal preference *at the moment* (and subject to more testing) is to run maverick thopterist vs burn and jund, alongside thoughtcast and reverse engineer (the latter in low numbers). It kinda works, but I'm cutting it close in testing and I'm not sure if my opponent is playing optimally since I can't see their hand. Also does not work very well vs aggro decks, since the thopterist is a slow clock, and I'm not sure how transformative my sideboard can be.
My initial reaction to this was that bastion inventor was better than enraged giant as removal is out main weakness. But on more testing, I'm hitting a stumbling block in liliana of the veil which means that non-evasive hexproof has some weaknesses. So I did some very brief testing with enraged giant, my thoughts were that it is OK, but not stellar. Then I went back and jammed in other haste creatures, in monastery swiftspear and storm entity. threw in glint hawk for extra storm and beats. Deck worked pretty well. Had a niggle of consistency issues since sequencing is very important, and it really wants to draw a mox, but it poured out massive damage when it worked.
This sounds like a good idea to me. I don't think a single green splash is too hard on the mana base (provided you ease up on the blue, I wouldn't run green and serum visions, for example), particularly since green has plenty of mana-fixing cards available.
Mostly covered already. The land count is fine, my main thought is that it seems like you'd have trouble drawing both the paladin, the lands and the protection all together in enough turns to win against something with a fast clock, like burn or jund. That said I don't have advice on how to change, as I don't think there is any settled lists yet to refer to, I'd leave it to your testing and personal preference. For example, ktkenshinx prefers more cantrips, while I prefer more threats. Depends on the gameplan and your willingness to take risks - note also that theoretical consistency against bad matchups like jund and burn isn't necessarily something to be afraid of, it might be better to take a gamble with those games and win for sure vs decks with slower clock, possibly while relying on the sideboard more heavily, than to try and optimise against specific bad matchup decks.
EDIT: reply to this dropped off; adding in.
Protection is fine, some people prefer more flexible cards which can serve dual roles, like silence and swan song (I prefer silence) since they also interrupt the opponent's plan. That said, protection is a perfectly reasonable way to go, though I would recommend gods willing as the better choice.
My thoughts: I am surprised that a deck with only 3 serum visions 1 nox didn't brick after the first guy got removed: or get mulliganed to death. From a closer look I will say that it looks like you got a tad lucky in a few of those matches (not that luck is a bad thing, this deck will always have luck involved), with some good topdecks. I suspect/theorise that larger samples could reveal a bit of a consistency issue, it will be good to track how you go in future. It also appears that the matchups posted other than Jund are highly favourable (possible exception of lantern, sounds like your opponent misplayed? I'm not sure, not familiar enough with the interactions with our deck/boardstates) because we are too fast for opposing combo or the opponent's clock was too slow even if they have control/hate.
After having the chance to test, I think I've settled on a non-combo version of the deck as my preference, since it seems to have the most overall resilience by relying on hexproof and huge toughness to beat removal and win combat, and sheer card draw to scuplt the hand and combat discard. I've gone to golem-skin gauntlets as my wincon, rather than grapeshot, since in this mana-heavy version I'm drawing over multiple turns, rather than all on one turn. It is obviously much much slower than the combo version, but has a degree of inevitability that the combo doesn't have. Still plenty of refining to do (a sideboard and fixing the mana base) but I'm pretty happy with the build. The sideboard could be particularly interesting as it is quite possible to convert to combo for game 2 (or conversely, to run combo maindeck and convert to this postboard).
Decklist:
4 Puresteel Paladin
4 Sram, Senior Edificer
4 bastion inventor
Artifacts 20
2 golem-skin gauntlets
3 sigil of distinction
3 cathar's shield
4 bone saw
4 accorder's shield
4 mox opal
4 repeal
4 thoughtcast
4 reverse engineer
lands 16
4 flooded strand
4 windswept heath
4 hallowed fountain
2 island
2 plains
Re: Stony silence:
What I meant to say was that I don't see this as being a particularly bad (good?) hate card, i.e. not one that really hoses us. The equip-blocking makes bolt a smidge better, but is otherwise inconsequential, except in aggro builds. The mana block from mox is a bit of an issue, but doesn't really cause too many problems if you think about it: A T2 Stony Silence means they're tapped out- this means we drop a paladin and draw a bunch of cards. Even if we don't combo off, that can give us fuel and sculpt our hand to go off next turn, not necessarily a loss. Note also this means that they have to remove the paladin in their turn, reducing their T3 plays. If they T3 s.silence, holding up removal, that means we can T2 combo, but also means that they've slowed themselves down when they could've been doing something productive on T3- again, not necessarily a bad outcome. And if you consider that we have up to 3 or 4 lands to play with, this isn't a terrible outcome either. Now to go a step further, consider what boarding in Stony Silence means- what have they taken out? Did they mulligan to get it? both of these can be a positive for us as we die to simpler strategies than stony silence. Put another way: stony silence is among the hate cards that people can throw at us, but is almost certainly at the low-end of the hate list. Compare Eidolon or chalice, which simply shuts us down the moment it hits the board. In any event, I'm not telling people not to board in cards to take out stony silence, since it doesn't require a taxing sideboard (and since the paladin draw engine still works, so we have good odds of actually getting the silver bullet), just curious that so many people are throwing it up as a hate card.
Re: gemstone mine (not caverns!)
In an ideal/theoretical world we won't tap in three times, but in practice it happens with sufficient regularity to be wary of it. At least as much is my experience. Put simply, we want to tap out every turn. Not an issue if we win by T3, which does happen (alot) but not always.
My testing was vs Jund (old builds, not sure what the new fatal push meta will look like, but probably even worse). Personally, I found jund unwinnable. Between discard and removal, there was no way to consistently win. Random wins? Sure, but consistent wins? None. Like 80-20 or some such nonsense. Burn was easily winnable with the right sideboard, although the matchup was much more borderline preboard in the pure cantrip version, I would say less than 50-50, but it depends on the opponent's list, as some burn is actually not necessarily that removal heavy.
Why are people concerned about stony silence? What games do we actually win by equipping, especially given that none of the decks above are running aggro cards? Is it really hurting the moxes *that* badly?
Why do people run leylines? While I accept that they protect against discard, they are card disadvantage, can't be reliably cast, and don't offer on-board protection. Has anyone done the math on how many games you'll actually draw leyline and paladin in the opening hand?
Re: Gemstone caverns: from past testing I've played many games where I've depleted these lands (cantrip, paladin, cantrip). Noxious revival gets them back, sure, but that's some bad card disadvantage.
Anyway, I've got some initial testing under my belt over the weekend, and thoughts as follows:
The cantrip-heavy build is reliable and fast. However, I can't see how it can also sport protection without becoming too dilute. Basically, I see it as a sideboardless deck. Its success will rise and fall with the amount of removal in the meta. In a removal heavy-meta (and vs opponents who know what they are doing), it should never, or rarely, win. Vs Aggro and combo, it will perform fine.
I am currently focusing on an aggro build. This takes me away from combo so I won't post the list here, as it risks turning into a different deck entirely (improvise midrange with a side of combo), but there's a heck of a lot more testing I need to do first before I settle on a build. All I do know is that bastion inventor and spirit link work wonders vs burn.
Prelim: our primary concern with splashing colours is the number of lands in the deck. The core engine needs only two lands to go off, and consumes both lands to cast puresteel paladin. Since it is a very safe assumption that you're comboing the same turn paladin hits the board, you're realistically only going to have 1 spare mana up at any given time (the land on T1). Sometimes you will have an extra land on T3, but not always. The mana gained through mox opal is unreliable, but also typically needs to be held up midcombo (i.e. to cast retract). This means most of the analysis for a colour splash depends on what 1-cost cards are in that colour which support the combo, or what 3-cost plan B cards in that colour.
1: White: obvious; Puresteel Paladin costs WW. This means that white is essential to all builds. With fetches, we can reliably go tri-colour, with two Wx shocklands. Quad-colour can raise the question of whether to use fetches or rather rainbow lands like mana confluence, etc. The new spire of industry adds to that option.
2: Blue: retract costs blue, so we want blue. It isn't safe to assume that retract will always be cast off a mox- often you can only dig a couple cards deep before needing to fire off retract, so you may not have drawn a mox yet. Adding blue lands therefore adds to the consistency of the deck. Add to this that blue has cantrips, which, while not the best tutors around, are the easiest to slot in, and the case for blue is strong. Blue also has some backup plans by offering protection and (historically) riddlesmith. Because of this, blue can tend to crowd out the other colours. That doesn't mean the deck has to be built blue however, and significant consideration has been put into the other three colours, discussed below. The general consensus is that blue is superior, predominantly because the tutor effects in those colours are not strong enough to warrant diverging from blue.
3: Red: grapeshot costs red, however, grapeshot is typically cast off mox opal, so that's not the main reason red can be splashed. Instead, the main reason is that red has the best "plan B"s - i.e. alternate wincons outside of white. Examples are in the primer, but include ghirapur aether grid, goblin gaveleer, Jeskai ascendancy, the list is long. So overall, red is a very flexible colour choice that fits in with many different builds. However, red is the worst colour for tutoring out paladin, and doesn't protect paladin. As such, red considerations are typically about putting in alternative threats, rather than supporting the main plan. With Sram, the main plan got much more consistent, which is why you've seen the degree of red splash reduce in most people's lists.
4: Green: The main reason people don't run green is that a 4-colour build can't be reliably sustained on fetches and shocks. This colour cost is seen to outweigh the additional benefits green can bring. Specifically, green has the best 1-drop creature dig, in commune with nature. Other green options are of a similar ilk (e.g. oath of nissa), but the marginal benefit compared to, say serum visions is not huge, so rather than risking the extra colour, most people choose not to splash green. Also, green does not give us a sound "plan B". If it did, I feel that a green splash would be more than justified. By way of example, cards like managorger hydra are inferior to monastery mentor. Green's protection cards are also not superior to white's protection cards. Overall, green isn't a bad colour, but doesn't do much the other colours do. Assault formation is bad in this deck since we can't get creatures to stick, so it just does nothing most of the time. Unlike red's Plan B, which fit in naturally, this requires a build-around, which means a substantial departure from the combo. Green also has some midrange options, like congragation at dawn, but that's considered too slow.
5: Black. Black contains spoils of the vault, so was considered extensively. However, it's generally accepted that the risk of just randomly loosing a game isn't worth it. Other than that, black doesn't have protection and doesn't really dig. In terms of the Plan B's that are possible with black, there are some (painsmith, glaze fiend) but they are typically outclassed by red, or require an extensive build-around. ovalchase daredevil was experimented with a number of players, but that doesn't count as a black splash since we never cast it. Midrange options are relatively good in black, with the new fatal push and battle at the bridge helping us out, but that begs the question- why are we trying to prolong the game? That's not to say this option can't be good, but it requires dedicated deckbuilding.
6:multi-colour: I once dabbled with signal the clans I think most people agree that its not worth the two extra colours, but just thought I'd put it out there, as the only reliable 2-cost tutor.
Darn, I just realised I will be travelling at the time of the GP so I can't take this deck for a spin; I would've loved to run this at the Brissie GP. Alas, no more modern GPS in AU for the rest of the year...
My analysis of Outcome is as follows:
A - 4 Mana is just a huge problem. I found I could cast it reliably with, and only with, lotus bloom or ritual shenanigans, but that diluted the deck and in any event, the bloom/rituals reduced the number of equipment I had in hand to bounce. Fun, but neither consistent nor reliable. Even if you do cast it, finding additional mana to then cast grapeshot or paladin is impossible. I found that it was much more viable with aetherflux reservoir, since you can outcome after reservoir, rather than the other way around, but the 4 mana on that too makes it clunky to say the least. And prone to artifact hate. reshape and whir of invention or the expertises *could* make this a better strategy, but it very much risks over-complicating the deck. Likewise, in a ritual-heavy version, empty the warrens becomes fantastic, since you need a minimal storm to win. But this is just not reliable, since it really only works if you have an optimal opening hand.
B - You don't really want to cast outcome quickly anyway, even if you can cheat it out. If you cast early, you're going to draw, say a max of 7 cards? Usually 3-4? That often isn't enough to win the game.
C - Dead mid-combo, so you don't want to draw it late either.
That said, there's really only one place for Outcome to do any work: if you combo and fizzle. If this is happening to you, outcome can be a viable way to recover a game. So my short answer is Outcome is viable as a one-or-max-two-of as a recovery tool in a deck with relatively high land count. While it could potentially be breakable, you'd have to warp your entire deck around the card, and I'm not sure that that is desirable.