Printable View
You do risk ending up being a poor version of either though. Nyx Fit's plan is strong enough to stand on its own. You just have to play differently from Rhino Fit. Instead of creating a board state and increase pressure you're trying to pull off a combo, basically. That requires a far more defensive and deliberate style of playing.
Active Deathrite only shuts down Rector if they have an open green source, remember. They often don't, especially 4 colour builds.Quote:
Originally Posted by drude1
Well yes, of course. But any decent legacy player with green Mana available is going to leave it up. In the mean time you are getting slammed in the face by delver, Leovold and any other creature they have on the board and there isn't much you can do about it. I just think it makes more sense to go after your opponent from more angles. Maybe a heavy Planeswalker build would be another option. That way you could still play the basic engine of nyx fit but not be so dependent on rector triggers.Quote:
Originally Posted by Navsi
Sent from my SM-N950U using Tapatalk
Nyx and Walkers are very different builds. We were discussing PW's briefly a bit in discord last night. One of the issues with a PW build, is that no one has ever figured out what the right build for it is. I definitely think there's one out there, but I think the current PW build which is SFM+combat walkers is the wrong way to go about it.Quote:
Originally Posted by drude1
And then there's the issue of mana curves. Running Rectors AND a good chunk of PWs bloats your curve way too much. At that point you digress to resolve a Veteran Explorer trigger around turn 2-3 or die.
Anyways, there also was some talk on the Discord about applying the SE method to Nyx Fit so let's see if we can get this off the ground. If we do, this is the itinerary for the coming week:
- Monday: Requirements modeling
- Tuesday: Requirements modeling
- Wednesday: Card evaluation and making a start for the list
- Thursday: Card evalutaion and brewing the main chunk of the list
- Friday: Finalizing the list and adding the SB
I'll be guiding the process and keeping track of requirements as we go.
A couple of pointers:
- Once the requirements model is defined, no concessions are made unless abolutely necessary
- This list isn't about what we want to play, it is about what we should play. No cards are omitted b/c of personal dislike, only b/c they do not meet requirements
- For deckbuilding purposes, Dryad Arbor is to be considered strictly as a creature and not a land
To get things started, I'd like some requirements for...
- Number of cards that can interact with the opponent
- Number of combo pieces
- Number of card advantage/selection cards
- A mana curve
- The manabase
To start, some obvious constraints:
- No card that we want to actually cast may cost more than 6 mana (although we might want to tighten this constraint, since we're not on a regular Nic Fit build)
- No more than one card we want to cast may cost 6 mana
- No more than two cards we want to cast may cost 5 or more mana (this means a max of 2 5 and/or 6-drops we intend to cast)
- No more than seven cards we want to cast may cost 4 or more mana
- The manabase must consist of at least 22 lands
- The manabase must contain at least 14 initial green sources
- The manabase must contain at least 14 initial black sources
- The manabase must contain at least 13 initial white sources
I think applying restrictions to mana costs before determining how we're going to set up the mana base is going about things in the wrong order. We might end up in the same place anyway, but I don't think this is the starting point we should be going for here. I'd aim for something like this instead:Quote:
Originally Posted by Echelon
- Deck should be capable of resolving a Rector trigger if undisrupted by turn 4 at the latest 80% of the time
- Deck should be capable of resolving a Rector trigger if undisrupted by turn 3 at the latest 50% of the time
- Deck should be capable of resolving a Rector trigger against a single piece of disruption by turn 4 at the latest 50% of the time
- Single rector trigger should end the game against at least 40% of the metagame
- Single rector trigger should put us heavily advantaged against at least 80% of the metagame
- Two rector triggers should end the game against at least 80% of the metagame
- Three rector triggers should end the game against 99% of the metagame
- Deck should should be able to stabilize / disrupt until turn 4 against each meta deck 60% of the time
Mana restrictions can be extrapolated from needing to have a Rector and sacrifice outlet on board by turn 3-4.
Interaction requirements can be split up by deck. Against Burn our minimum disruption to reach turn 4 is probably something like one blocker and either two removal/discard effects or a Collective Brutality. Against Storm or S&T it's probably two discard effects. Elves is probably a spot removal or discard effect and either a sweeper or two more removal spells.
I read interaction/disruption w/ specific roles. I like it. It probably suffices to make a distinction between creature spot removal, creature sweepers and hand disruption. The target of those should be "number cast before the start of your fourth turn", so for hypergeometric distribution we're working with n in 9 cards (assuming a card drawn on T2 and T3) out of a 60/61 card pool.Quote:
Originally Posted by Navsi
Edit: And since those cards need to be cast on 3 mana or less, this gives us the following constraint:
- Cards considered interaction/disruption of any sort must cost 3 or less mana
- 2/3rd of cards considered interaction/disruption of any sort much cost 2 or less mana
We could expand on this by defining X chance to have cast at least 1 on turn 2 & Y chance to have cast at least 2 on turn 3.
I guess we can make the assumption we'll always have the mana needed to cast something on T2 and T3. That way we can pick a percentage we're comfortable with for the various turns and find the corresponding numbers via a hypergeometric distribution table.
All right, so I've done some maths. Currently I've just been looking at Rector triggers, since that's the core of the deck and the most complicated part.
I've been quite loose with these numbers. Actual odds are lower.
P(Enchant) - odds of getting a rector trigger
P(Enchant) = P(Rector in hand) x P(4 mana available) x P(White mana available) x P(Sac outlet)
---
For the odds of a Rector in hand, I've just looked at odds of naturally drawing one, or Leaping into one.
P(Rector in hand) = P(draw Rector) OR P(Leap hit)
The odds of drawing Rector, if we run 4, are 52% (10 cards) or 48% (9 cards).
P(Leap hit) = (Rectors / Total creatures) x P(draw Leap) x P(Green mana available) x P(Fodder available)
Fodder is anything we can feed to Leap. I was looking at Souls and Explorer and Zenith, which already puts us at 86%. Adding fetchlands for Arbor will increase it more, but I cba trying to do that maths.
We can't Explorer for green mana because we need it first so it's linear - 92% if we have 14 sources and 9 cards drawn.
If we run 9 creatures and 4 Leaps:
P(Leap hit) = (4/9) x 48% x 92% x 85% = 17%
P(Rector in hand) = 48% OR 17% = 56%
This is the biggest roadblock / lowest pecentage of the four. Only 56% odds of actually having a rector to sacrifice, combined with any interaction the opponent has, is something of a problem. More filtering or tutoring would help a lot.
---
P(Sac outlet) = P(Tower) OR P(Leap) OR P(Therapy)
If we have Leap or Tower we need an additional mana from elsewhere to compensate for the effect's mana cost / use of a land drop. However, if we do have spare mana, other removal spells could also do the job.
If we run two Towers, four Therapy and four Leap:
P(Tower) = 28%
P(Therapy) = 48%
P(Leap) = 48%
P(Sac Outlet) = 81%
---
P(White mana available) = P(white land) OR P(explorer trigger)
P(Explorer trigger) = P(Explorer or Zenith) x P(Sac Outlet) x P(Green mana)
If we run 4 Explorer and 4 Zenith:
P(Explorer trigger) = (48% OR 48%) x 81% x 92% = 54%
P(White mana available) = 92% OR 54% = 96%
---
P(4 mana) - this is the awkward bit. If we have only 3 turns to make land drops in:
P(4 mana) = P(3 lands including Tower) OR P(2 lands and Explorer trigger) OR P(3 lands and Zenith)
P(3 lands including Tower) = P(2 non tower lands) x P(Tower) x P(Fodder creature)
If we run 23 lands and 2 Towers:
P(3 lands including Tower) = 89% x 27% x 86% = 21%
P(2 lands and Explorer trigger) = P(2+ lands) x P(Explorer trigger)
P(2 lands and Explorer trigger) = 92% x 54% = 50%
P(3 lands and Zenith) = P(3+ lands) x P(Zenith) x P(Green mana) x P(No Arbor)
P(3 lands and Zenith) = 74% x 48% x 92% x 85% = 28%
P(4 mana) = 50% OR 21% OR 28% = 72%
-
If we can make a 4th land drop, we can add on the odds of just hitting 4 lands, which brings us up to 83%.
---
P(Enchant) = P(Rector in hand) x P(4 mana available) x P(White mana available) x P(Sac outlet)
P(Enchant) = 56% x 72% x 96% x 81% = 32%
If we want to increase these numbers, we want either more acceleration, or more ways of finding Rector. Preferably the latter. Note that even if we got P(Rector in hand) all the way up to 100%, we'd still only have 65% odds of getting a trigger on turn four. Improving our ramp is the other place to be looking.
I might run some numbers on whether Eladamri's Call helps our numbers much.
Awesome post! I'll update the model in the morning and continue development from there.
So I brought Nyx Fit to a side event at GP Jersey with this list:
MB
2 Abrupt Decay
4 Academy Rector
2 Bayou
4 Cabal Therapy
2 Collective Brutality
1 Cruel Reality
1 Curse of Death's Hold
1 Diabolic Intent
1 Dovescape
1 Dryad Arbor
3 Evolutionary Leap
4 Forest
2 Garruk Relentless
3 Green Sun's Zenith
3 Lingering Souls
1 Overwhelming Splendor
3 Pernicious Deed
2 Phyrexian Tower
2 Plains
1 Savannah
1 Scrubland
2 Swamp
2 Sylvan Library
4 Verdant Catacombs
4 Veteran Explorer
4 Windswept Heath
SB
1 Bitterblossom
1 Choke
2 Ground Seal
3 Leyline of Sanctity
3 Leyline of the Void
1 Pithing Needle
2 Thoughtseize
2 Toxic Deluge
Games were:
Grixis Delver: 0-2
Mull to 5 game 1. mull to 6 game 2, opponent told me he just kept ripping every answer he had, games went quick.
Nic Fit Walkers: 1-1-1
Game one got out Cruel Reality, Overwhelming Splendor, and Curse of death's hold. He had out a Nichol Bolas God Pharaoh. Kept ripping walkers to save the bolas from cruel reality, he got to ultimate it, and I scooped for time.
Game two got out an early cruel reality, he only had Nicol Bolas out. Then a dovescape followed by a GSZ X=7 sealed the deal.
At this point time was called and we decided playing 5 turns was useless so we took the draw
Miracles: 2-0
Game one, was able to get out an early cruel reality which sealed it vs a jace and not much else.
Game two, Garruk was an allstar here. Kept making 2/2 wolves. Sooner or later I landed a dovescape and was able to win a bird war.
BUG Delver: 2-1
Game one: Was able to therapy his hand to nothing, taking a FOW, and then saccing an explorer to get two goyfs on the second half. He played a Tomb Stalker, I was able to get out a cruel reality off of a rector game was over.
Game two: Mulled to a 6, he overwhelmed me with creatures, forced a deed, couldnt find a rector to splendor
Game three: We got to a board state when I had an explorer, 2 spirits, and a deed on the board. He had a library and two delvers. I have him down to two, and am waiting for him to tap so he doesn't have BG open because I played him earlier in the day and he blew me out with golgari charm giving his guys regeneration. At this point he extracted my rectors and my lingering souls. I get an opportunity to pop my deed and clear the board. Next turn I rip a land, fetch, and jam a reality. He forces it, and when we go to make note of life totals we realize he killed himself.
All in all I think the main deck is pretty good right now, it has enough "need to answer" things that it can make it through a lot of counter spells, and once a leap is in play it pretty much just flows. I think the sideboard could definitely use some work though.
If you guys have any questions let me know, sorry for the fuzzy report, we ended at 10, after being there since 9 my brain was super fuzzy at the time. I had a ton of fun playing the deck and definitely think it can hang with the big boys in the format.
Impressive work here !
Some things to consider : crop rotation count as sac outlet and interaction, some split of green sun zenith and other tutor (in order to get rector in hand) like intent or eldamri's call.
Keep going :smile:
@Navsi: To get to the consistency we want, we can define the number of ramp/tutor (i.e. Rector copies) as requirements and figure out how to fill those in later.
For now I'm thinking of the following (since we pretty much must have one of each in order to hit our goal):
- The list must contain at least 12 "copies" of Academy Rector (this puts us on just over 90% chance of having it on turn 4)
- The list must contain at least 12 sac outlets
To hit each of our land drops the first three turns (assuming we'll always want a ramp card in there somewhere), hypergeometric distribution suggests we run 23 lands (although 22 lands yields only a .56 percent less chance of hitting them. I'm comfortable with that).
- The list must contain at least 22 lands
And since at least one of them must be green:
- The list must contain at least 14 initial green manasources
Ideally we do want a ramp card in there somewhere and I think the latest it can hit the field is around turn 3, so:
- The list must contain at least 13 ramp cards (putting us on just over 90% chance of having it on turn 3)
The next thing we need to figure out is how many spot removal/sweepers/hand disruption this list needs in order to survive long enough. This'll give us the requirements for those.
Edit: This might bleed into sideboard requirements a bit. There's no way in hell we can jam all that in the main 60/61.
Edit 2: We also have to start considering card velocity.
Note that if we run additional filtering options (like Library, Guile or similar) we can increase our odds of drawing Rectors without needing tutors or wishes. Getting to see ~12 cards rather than 10 (7 hand + 3 drawn + 2 Library trigger) helps our odds.Quote:
Originally Posted by Echelon
If we run 4 library effects we have 52% odds of getting one in the opening hand. If we draw one, we get to see two additional cards, up to 12 - but one of those 12 cards is 'taken', so functionally we have 11 'available' cards when determining the odds of seeing any single other card from the deck. Painful Truths lets us see three new cards rather than two.
Worldly Tutor is a source of Rectors, but if we draw a copy we functionally have one less card, so it reduces our opportunities to draw everything else. If we draw (and cast) 0.5 worldly tutors per game, the number of cards we see is reduced by 0.5 effectively.
This stuff is giving me a headache, lol.
Edit: It does show that adding (cheap) card selection add a lot of consistency to the deck, in every regard. Which is logical, b/c that's why every blue deck starts w/ 4 Brainstorm & 4 Ponder. Funnily enough cutting card selection/advantage is one of the first things most non-blue players/lists do "b/c it's underwhelming" while in fact it's the glue that holds the deck together.
Basically, if we want reasonable odds of a rector trigger we need sources of Rectors and sources of sac outlets.
Possible rector tutors:
Leap
Diabolic Intent
Living Wish
Eladamri's Call
Recruiter of the Guard
Worldly Tutor
Possible sac outlets:
Therapy
Tower
Leap
Diabolic Intent
Living Wish (for Tower or Rusalka or whatever)
Starved Rusalka (via Zenith)
Therapy is definitely in. Towers are Ramp + Sac Outlet which is great.
Leap and Wish are both Rector + Sac Outlet in one card. Wish gets us Rector faster whereas Leap can do both at the same time and has other synergies. I think we are probably on one or the other of these, since moving a Rector to the SB for Wish makes our Leaps less likely to hit.
Rusalka via Zenith is probably worth it, especially if we're on the Wish plan, since it gives us a lot more sacrifice outlets. Therapy + Tower + Wishes + Zeniths + Rusalka might get us there reasonably well. Intent is also potentially an option, since it's another Sac Outlet + Rector source.
For reference:
3 Rector
4 Explorer
4 Zenith
4 Therapy
4 Wish
2 Souls
2 Intent
1 Rusalka
1 Arbor
3 Painful Truths
25 lands (14/14/14, 2 Tower)
By turn 4 (10 cards drawn naturally, 10.8 cards seen on average):
Puts us at:
P(Rector): 82%
P(4 mana): 94%
P(White mana): 98%
P(Sac Outlet): 96%
P(Enchant): 74%
I don't think we are going to get it much higher than that. Current build:
4 Rector
4 Explorer
4 Leap
4 Souls
4 Therapy
4 Zenith
1 Rusalka
1 Arbor
2 Guile/Library
1 Intent
23 lands (14/14/14, 2 Tower)
Stats:
P(Rector): 63%
P(4 Mana): 91%
P(White mana): 98%
P(Sac Outlet): 94%
P(Enchant): 53%
Note that the Leap build has significantly more inevitability because if your first Leap 'misses', you can keep using it to dig more. I don't think the build above is necessarily any better - it would probably crumble to disruption, since very few of the cards are X for 1s like Leap and Souls are. The advantage of the Leap build is that it still has reasonable odds of getting a fast enchantment out, but if it doesn't the deck still actually does something.
Edit:
Card selection doesn't actually help all that much. Cutting the 2 Libraries or Guiles from the standard Nyx build and replacing them with blank cards reduces P(Enchant) by approximately 1.5%. Basically each Library/Guile we add improves our odds by about 0.75%. In comparison, each copy of Eladamri's Call we add increases our odds by around 4% - Intent gives 5%, at the expense of vulnerability to disruption. Our odds of getting a turn 3 or 4 Rector trigger are actually a bit higher if we replaced our Mirri's Guiles with third and fourth copies of Phyrexian Tower. Obviously it does other things later on, but in terms of early effect it improves our odds of everything slightly, but has nowhere near the impact of most of the other combo pieces.Quote:
Originally Posted by Echelon
The framework so far does leave room for interaction/disruption though (somewhat surprisingly).
I think we do need to mind the manacosts of the cards we add, regardless of impact for goal achievement percentages. That way we aren't so soft to disruption and make the plan a bit more flexible. If we have to do all the things on T2, there are limitations to what we can and cannot do (which also hampers our ability to achieve our goal on a certain turn). Card velocity matters, even if that means we have to run less powerful/impactful cards.
I'm investigating wish options too.
I think leap+wish can still be coexistent. You can swap the "missing" rector by a witch for exemple if you're concerned by the leap's doing.
Edit: a big + is that we're not dead on a surgical...
Or should you put a Witch in the SB since it's basically a worse Rector?
The Sideboard option probably shouldn't be the Witch, since 50% of your rector/witch triggers would be that card, and not finding dovescape is awkward (and Wish is already mana intensive, so 4 -> 5 is an issue).Quote:
Originally Posted by Echelon
True, as is the point on Surgical.