not +2. +1.Quote:
Originally Posted by Dihensoeur
I don't really like 4 GSZ since they tend to get stuck in my hand.
It is an older argument I had with Ben a while back, so it is of no concern really.
Regards,
Matt
Printable View
not +2. +1.Quote:
Originally Posted by Dihensoeur
I don't really like 4 GSZ since they tend to get stuck in my hand.
It is an older argument I had with Ben a while back, so it is of no concern really.
Regards,
Matt
Guys, when you're playing with Leyline of Sancticity in the sideboard, do you want two G/W duals, or do you just put 'em in, and mulligan until you get one, and treat the rest as dead cards?
You want at least one land. Faeries and sprawls and possibly moxes can take care of the second white.Quote:
Originally Posted by Muffinman
Oh yeah, that's true. They're never really dead.Quote:
Originally Posted by waytowinatwar
What about the chrome moxes by the way. Are they worth featuring? I'm guessing they speed the deck up somewhat, but I feel like they could also screw up some hands. Would you say they're worth it?
Yes, in moderation. A few posts ago, i did some math about unkeepable hands, and the number of chrome moxes in your deck contributes to this number. However they support the other forms of acceleration in the deck, are amazing when going off, and t1 argothian or cmc2 sideboard hate is no joke. Maybe 1 is right (gainsay ran 1), I run 2 and have had some success with this. I know ben favors 3.Quote:
Originally Posted by Muffinman
Although I’m delighted that activity in this thread is increasing I’m hoping that this post doesn’t get buried since I think it is very important.
So there are as of yet no detailed, well-grounded rules about how to mulligan with this deck. This is the norm rather than the exception for magic decks, and there are several good reasons for this; mulligan decisions are matchup-dependent, and which matchups that are relevant change over time, and there might be differences in these decisions depending on if you are on the play or draw and on whether it is a sideboard game or not. So there is considerable branching, and at least some instability, and these two factors both make theorizing, and rule-formulation less straight-forward.
As a small step towards formulating a set of heuristics for how to mulligan with this deck, I took upon myself to try to create a data set that could underlie the formulation of heuristics for how to mulligan when goldfishing (in order to improve your chances of winning as fast as possible). This is a lot of work in and of itself, but I’m hoping that it could eventually lead to formulating detailed, well-grounded rules about how to mulligan in particular matchups.
Towards the end of this post I make some suggestions about how to carry on this (very preliminary) investigation. Gathering this kind of data is quite time consuming so I'm hoping that some of you who contribute to this thread can help me out. It is important that we progress in a rigorous way here, but I’ll leave aside the details for now (send me a message and I can go through them with you)
Anyway, so what I've done is go through 30 goldfishes with this deck with a 7 card opener (i.e. a normal game on the play) and 25 goldfishes with this deck with six cards (i.e. a game on the play after a single mulligan). I wrote down each opener together with the number of turns it took to win with that hand, where I took winning to mean "achieve the cycle that allows us to bounce all of our opponents permanents" (actually winning typically takes exactly two more turns). (I don’t know if there is a way to share excel files here but send me a message and I can email it to you if you are interested, I’ll just go over some generalities here)
With this kind of data one can do a number of interesting things, e.g. calculate how fast the deck is on average, and test various rules for selecting better hands than others. What I’m focusing on here is when to mulligan a seven-card hand for a six-card hand. Below I will assume that the data is representative of the general case. This builds on a few assumptions, among them that I goldfish the deck fairly optimally. In generating the data I haven’t taken any mulligans, just played out the hands that were dealt me, with the exception of no-landers (/no moxen). I excluded these hands on the assumption that it is always right to mulligan these, and that including them in the data set would only introduce uninteresting noise. I have taken note of how many non-landers(/no moxen) I was dealt (2 for the seven card hand 3 for the six card hand). The probabilities below, are thus propabilities given that you are not given a non-lander(/no moxen). The probabilities reported below correspond to the proportion of hands that lead to a winning on a certain turn.
The seven-card hand:
With a seven card hand you have a
3% chance of winning on turn 3,
27% chance of winning on turn 4,
37% chance to win on turn 5,
17% chance to win on turn 6,
7% chance to win on turn 7
10% chance to win on turn 8 or later
(numbers are rounded up for readability so percentages don't sum to a 100).
So if you take zero mulligans (and we ignore the quite infrequent no-landers) you are more likely than not to win on turn 5 or earlier. A little less than 1 games out of 3 you will win on or before t4, but about 1 games out of three your game will take 6 turns or longer. Your average win turn is 5.4 (but this number is somewhat inflated by isolated games taking improbable amounts of time, e.g. one game took 12 turns, where I hit no enchantress effects at all until turn 11). The median win turn is 5. It is thus fair to say that the deck is turn 4-5 deck (or even a turn 4-6 deck). (Incidently, this means that the estimate in the first post of this thread ('this is a turn 3-4 combo deck') is too optimistic).
The six -card hand:
With a six-card hand you have
0% chance of winning on turn 3
8% chance of winning on turn 4,
24% chance to win on turn 5,
24% chance to win on turn 6, a
28% chance to win on turn 7 and a
16% chance to win on turn 8 or later.
This is quite a bit worse than the seven-card hand. The majority of your games will now be won on turns 5-7 rather than on turns 4-5. The average win turn is 6.6. (This number is, again, influenced by a few games taking very long, e.g. one went to 11 turns and another to 14). The median win turn is now 6.
So given that this data is representative of the general case, which seven-card hands should be mulliganed? (Remember: we only care about goldfishing at this point, so the advice below should be taken with a grain of salt, see the tell tale example towards the end of the post).
One reasonable way to answer this question is to interpret in terms of which seven-card hands that has a lower average win turn than the average six card hand (and then maintain that these hands should be mulliganed). This is not the only way to interpret the question (and I’d be very interested in hearing alternative suggestions) but it is a reasonable one.
If the mulligan-question is interpreted in this way, and one is taking the data at face value, it turns out that one should be VERY conservative with mulliganing (since the average win-turns for seven and six card hands are so far apart).
Should for instance, all seven card 1-landers (/or 1 mox no land) be mulliganed? No. The average win turn for these hands is actually 5.7 (compared with the average of 6.6 for a six-card hand).
What about hands with only moxes or only one land and no acceleration(=land aura)? Well this happened very infrequently (4 out 30 games) but the average win turn for these is 6.5. So no, strictly speaking.
What about hands without acceleration (=no moxes, no land auras)? No. These win on average on turn 5.7.
What about hands without enchantress effects? No, these win on average on turn 6.2.
So basically, if the data is representative of the general case, it looks like the loss of a card that ensues when one mulligans is likely to be more important than the particular configuration of your seven cards! Although this is unexpected, it could be explained by the fact that the deck contains an unusually high amount of cards doing double duty as several things (so faeries ramp, but also cycle, gsz gets you enchantresses and thus extra draw triggers but also gives you enchantments (via witness) that trigger the draw triggers etc.).
There are of course hands of seven cards that are clearly better than other hands, for instance, if you have two enchantress effects, two mana sources (land or mox) and one additional piece of acceleration (mox or land aura) your average win turn is 4.5 and you are in fact about 67% likely to win, at the latest, on turn 4.
There is a lot to write about this, and a lot of holes to be filled in, but I think this is a good start. Let me know if you want to help out with this and we can discuss some details so we can ensure compatability. Replicating the data I obtained would be nice, but also getting data on 5 and 8 card hands (for when you are on the draw and haven’t mulliganed)
To illustrate the difference between the gold-fish case and the tournament case I repeat a decision I described above. I went 2-2 in a tournament last week, defeating spiral tide and miracles. I lost a game, fair and square against painter grindstone, a quite horrible (I think) thankfully fringe matchup for the deck. The final match was against a very aggressive zoo variant with Burning-Tree emissaries. I won the first game, and lost the second. The third game probably came down to an incorrect mulligan decision.
I mulliganed five lands, an elephant grass, and an utopia sprawl for two lands, argothian, presence, witness and prim. I got completely slaughtered. I played out my seven-card hand against my oppenent and won. Grass is so crucial here, and mulligan decisions against zoo for instance should clearly be influenced by whether Grass is in your opener or not. Grass is just an enchantress trigger like any other in the Goldfish. So even though the kind of analysis I offer above is important, it should be taken with a grain of salt.
So if I do understand you well enough (assuming also we took the same class in probability mathematics), we would need to establish a base.Quote:
Originally Posted by waytowinatwar
Or if we start collecting the data seriously and start to evaluate the data seriously better yet a core.
I assume we will start with your decklist?
*starts thinking aloud* Three weeks ahead I should have enough time to cobble up a small Prog to simulate the
tableaus you would require. Until then will have to do with what we have. My spare time is currently very restricted due to a high intensity period @work but I should be able to give you 30 - 60 minutes per day.
Send me a spreadsheet or description of a spreadsheet and I will start running those 60 games starting Friday.
Regards,
Matt
It's an interesting idea to try to find mulliganing heuristics, but as you said, the norm for any deck in the history of Magic on whether or not to mulligan has always been "it depends..." The only thing I would offer is that 30 and 25 games is close, but probably not enough to get a good representative sample; I say this having minored in statistics in college, where I think they said the rule-of-thumb number was 30 or so samples. But usually when I draw conclusions for Magic decks, I tend to take a much larger sample, simply because a Magic deck is such a complex idea and you're often asking numerous questions that interact with each other. When I was trying to find good numbers for Solidarity, for instance, I didn't feel comfortable saying anything meaningful until I'd played 100 games, simply because the potential to create outliers is very high. So I would say, it's a good start. I also don't think that speed is the best measure, though I guess you have to choose something.
As for my heuristic, I've found that, as long as I'm not in a match where I have to have an answer immediately, I'll keep anything within reason. Usually I'll have to look at the hand and see if it actually does something, but as long as it has lands and spells, it should be enough to get there. There really aren't too many cards that I don't want to see in my opener, other than maybe multiple Cloud of Faeries, but because the deck is going to buy me lots of time, there should be plenty of time to draw out. It's somewhat rare that I will keep a hand with no enchantress effects, but then it's also pretty rare that I get an opener without any. Usually the limiting factor is actually lands.
Are you guys familiar with Sensei's Divining Top and its interaction with Words of Wind? Together they can bounce your opponent's board. Works similarly to Abundant Growth, but can be abused with Sanctum mana.
I've really liked having 3 Chrome Mox while running this deck. They only work out poorly when you're getting flooded anyway. Those games won't be saved by having lands over the Chrome Moxen.
What's up with the random Mindbreak Trap in some lists? Random mulligans are bad if you aren't seeing like 25% Storm/Belcher.
Made Top 8 of a 33 person tournament at Monster Den in Minneapolis yesterday.
4 Misty Rainforest
2 Verdant Catacombs
4 Forest
1 Island
1 Tropical Island
1 Savannah
3 Chrome Mox
1 Karakas
1 Serra's Sanctum
1 Gaea's Cradle
1 Dryad Arbor
4 Argothian Enchantress
3 Cloud of Faeries
1 Eternal Witness
4 Wild Growth
4 Utopia Sprawl
4 Seal of Removal
4 Elephant Grass
2 Abundant Growth
1 Seal of Primordium
4 Enchantress's Presence
2 Words of Wind
4 Green Sun's Zenith
1 Emrakul, the Aeons Torn
2 Sensei's Divining Top
Sideboard:
4 Leyline of Sanctity
3 Carpet of Flowers
3 Gilded Drake
2 Chill
1 Tormod's Crypt
1 Gaddock Teeg
1 City of Solitude
1 Harmonic Sliver
Matches:
R1 against Chad with monoblack deck from 1998
Games went fairly fast since he didn't have much action.
2-0
R2 against Tom with Lands
Unfortunate mulls in these games. Almost killed him with his Bobs in game 3, but came up 2 life short.
1-2
R3 against Derrick with 4-color Delver
I locked him out at 7 and 8 life. He didn't get amazing draws and I didn't need to use my Seal of Removals much.
2-0
R4 against Eddie with Reanimator
G1 I played Karakas early. Instead of just trying for card advantage with Griselbrand, he went for Tidespout. I eventually bounced his board. G2 he got a turn 2 Griselbrand and just attacked until I died, basically. G3 He got Iona on Blue out with a Pithing Needle on Seal of Removal and then another on Words of Wind. I found Seal of Primordium and went off.
2-1
Top 8 against Matthew with Sneak and Show
I feel like this matchup is hopeless. He 2-0's me fast.
0-2
I haven't actually playtested the matchup against sneak and show, but in my eyes, it isn't really that bad. Storm looks much worse. And easy answer to emrakul is seal of removal, on top of that, you can also board in your own emrakul (If you're playing with one) into your main, to sometimes put it in off of their show and tell.
Also, I have to ask you about the Serra's Sanctum. Is it really worth it? I haven't tried featuring one, but I feel like most of your spells are only green/blue mana, and mostly not much that you can use the white mana for. I own a sanctum, but I'm not sure how I feel about playing with it in the deck.
Your list is a little different, so maybe you can utilize it better.
I know you talked about this earlier with the really large entry where you mentioned how often 7-card hands and 6-card hands win however early. I was just wondering how you guys would do with some certain hands. My list is very similar to the "standard" build here in the thread, so you work on the supposition that the deck is like that of the first post.
(Assuming that we're in game 1, and don't know what our opponent is playing.)
Would you mulligan a hand like this:
Forest
Misty Rainforest
Misty Rainforest
Elephant Grass
Green Sun's Zenith
Cloud of Faeries
Cloud of Faeries
I'm just wondering how I should treat a hand like this. Of course, a hand without ramp/a way to get an enchantress effect should be mulliganed, but that hand has either. Just not both. Either you ramp with GSZ, but then you won't have anything relevant to play unless you topdeck it, or you're just really slow. I'm guessing it's a mull? 6 Randoms should be better no?
That hand looks terrible in most match ups. Mulligan to anything has a better chance at winning.
I just shuffled this up for funsies and am playing time warp to get broken with some eternal witness recursion. It is super fun and seems to be a step in the right direction.
Hi there,
I found this thread a long time ago and read it carefully many times. I thank waytowinwar, benthetenor and Dihenosaur for the work and effort in this thread.
I am a legacy noob and have this deck sleeved up since about four years and finally managed to bring it to a small tournament this weekend. It was a GP trial for Paris. I used the standard deck list (with Ravnica Duals and Spike Feeder instead of Kitchen Finks). My sideboard lacks leylines, but It turned out didnt need them.
My first match was UW Miracles:
In both games he got down his Sensei's Top and Counterbalance after I had 2 or more enchantress in play. I just cast stuff, some sticks and he never countered words of wind.
Second match UW Miracles( ? probably)
Writing down a sideboard at 3am in the morning didnt work out to well and I got a loss. In the game he got down a t2 rest in peace. He followed up with an energy field. I try to play very fast, knowing that helm of obedience will kill me. He counters Words of Wind, Seal of Primordium and Words of Wind again. I end up having enough mana to cast emrakul (which is not in my 75). Left without answers i scoop. I still think this is an easier matchup and should be won.
Third match Death and Taxes:
This was the most frustrating match (also I love stax and tax stuff). In the first game I mulligan to 5. In both games he gets down a Thalia t2. With Thalia on the board he develops way faster than me. I feel this match up is more like a race: If he gets to much time, I am dead. First game I had 3 grass out with 5,3 and 1 upkeep. He had an Ethersworn Canonist, so I could not just bounce and recast them. Once he reached 6 mana I was dead.
The second game he plays very slow after dropping Thalia. Just using his two ports to tap down my enchanted lands, which I still use to pay for grass to survive for some extra rounds. I had to spread my sprawls among three different lands. I set some of them to white to fake some sideboard answers (he fell for my bluff and kept tapping the white instead of the blue sprawls). I don't know how to win this matchup and my sideboard also doesnt provide any answers.
fourth match RUG Delver:
He drops an early nimble mongoose and immediately starts throwing burn at me. Luckily he wastelands himself (I didnt drop any non-basics) to reach threshold. Being forced to block with my enchantress, I manage to bounce his board one turn after. I keep him at zero permanents, while staying at 3 life and frantically looking for my spike feeder. I find it, go infinite and follow his hint to put the +1/+1 markers on my fairies, which can swing without summing sickness for lethal.
The second game I cast a t2 enchantress, which gets dazed ( I actually wanted his daze: gives me a timewalk and I could cast a zenith for 3 next turn). However I get flooded and he gets down a Tarmogryf, which ends game 2 for him.
The third game he plays slower (nimble mongoose beatdown without threshold). He tries to counter my words of wind. After 3 force of will I bounce my witness with seal and cast words of wind again. One turn after I go infinite and win.
As I said before, I am a legacy noob. However I feel that the countertop-lock is not to bad, if we manage to get down enough stuff before-hand. I didnt encounter any discard/storm decks (which I heard are bad for us). How to you play against Death and Taxes? I felt there was very little I could do at all. Elephant grass kept me alive for quite a while, but is expensive. I thought about propaganda, but probably would just delay it. Engineered plague (set for humans) would shut down half of his guys. However, if the game goes as the second match, the death and taxes player would never let me get a black mana in my main phase. He was siting on his wastelands in both games, so adding a bayous would not help much.
Also I wonder how you get enough time to write down your hand and mulligan decisions. I almost always end up being the last active table.
I was placed 9 out of 15 people, missing the top 8 due to opponent score.
Countertop is definitely not very bad, since we have a lot of answers to it, depending on when you draw them (Seal of Primordium, GSZ for Harmonic Sliver, Krosan Grip) and if you get a few Enchantress effects down, you can just start casting cards into it and still maintaining or gaining cards, since the draw triggers on the casting of enchantments, not on the resolving. If you ever resolve a Words of Wind with a few Enchantresses, you will win eventually.
In the fourth match, I don't like running Argothian Enchantress into Daze. This deck is better than pretty much any other deck at waiting, making basic land drops and playing around Daze and Spell Pierce (particularly if you can land a Carpet of Flowers, but it's not necessary), so there's no reason to just run into it. If I were you, and he had no pressure, I would have just made my land drop and passed the turn. Once you resolve an Argothian Enchantress, that game is pretty much over, and if you get to the second one (with GSZ) it's only a matter of time.
As for Death and Taxes, I haven't played against it but I would imagine it's a pretty good matchup for us if you understand how they're going to attack you. You don't have to worry about running into soft permission so you can just jam your Enchantress effects (and they can't kill your Argothian Enchantress), and once you start drawing cards and making basic land drops they will be hard-pressed to lock you out, and for a creature deck they're on the slower end of the scale.
All in all, I'd say just keep practicing and I think these matchups will get better with experience.
im trying to get into this archetype. is there a good bunch of videos somewhere that i could watch to learn the inner workings of this deck? it seems pretty skill intensive. thats good and all but it would be helpful to watch someone play it to know whats up.basically sensei top + words of wind= 2: each opponent bounces a permanent. seems decent.Quote:
Originally Posted by waSP
is there a reason sensei's top is not featured in some UG lists? it seems like this could have potential.
There are some videos in the youtube channel of mtgvideo. At least some of them (if not all) should be played by waytowinwarQuote:
Originally Posted by sun tzu
http://www.youtube.com/user/mtgvideo...ry=enchantress
First of all there are only two words of wind in the main deck. so it is not that likely that we will happen to have words of wind and top in our opener.Quote:
is there a reason sensei's top is not featured in some UG lists? it seems like this could have potential.
If we would, we could play words of wind t2 and start bouncing t3 (which is good). However if we dont get down words of wind in t2, then we wont be faster than playing without top. I my experience with the usual build you can bounce a good amount of permanents/all permanents by turn 4 to 5.
The deck has a high density of 1-mana enchantments of which most even cost less/generate mana (seal by bouncing fairies, wild growth, sprawl, carpet of flowers). Thus playing a 1-drop and having some enchantress in play, it is very likely to draw another 1-drop. This allows us to bounce, develop board presence and draw cards( compared to just bounce, if using senseis top). Bouncing our own permanents can be an adavantage: bounce lands to tap again, bounce fairies for mana, bounce 1-mana enchantments to be recast using the mana they provided before. The point i want to make here: Bouncing with 1-mana enchantments is clearly better than using top if some enchantress are present.
The problem with top is: If you have your enchantress setup and are rushing through your deck to find words of wind to finish the game, top increases the chances to not draw an enchantment. Then you might be stuck with lands/artifacts on your hand, having to pass the turn and give your opponent the time to finish the game.
The thing with the mindbreak trap mainboard is the following: After clearing the board and swining with faeries, your opponent might still play land and 1-mana spells on his turn (in particular burn spells). Mindbreak trap (with eternal withness recursion) was there to counter that. I don't know if there is any other 1 mana card that frightens us (one can play around pithing needle), but kitchen finks also helps against burn and is tutorable with zenith in aggro matchups; so has more utility.
the reason i think it may have potential is not just in the 2:target player bounces a permanent 'combo', but in its abilities. for example, when you have words of wind in play and someone is trying to target it, you can instant-speed draw a card to replace effect winds back to your hand to save it.Quote:
Originally Posted by yell
the look at the top 3 effect is also very relevant when you are trying to build your early game with very little mana investment.
dont get me wrong, i wouldnt ever use more than 1 or 2 tops, but i think 1 might have potential, so i'll test it and let you guys know how good it seems when i draw it early and use it to set up my turns, or draw it late and use it to ideally save a words of wind at a crucial moment.
has anyone tried ancestral knowledge as a 1 of, or more? whats the verdict on this card?
I don't see there being a whole lot of situations where you're going to need the instant-speed draw effect to save a Words of Wind off of the top. The enchantments that the opponent should be targeting if they don't want to just lose are the Enchantress' Presence. Eternal Witness + Seal of Removal gives you effectively 8 Words of Wind if they are really trying to kill it with Abrupt Decay or whatever you have in mind, not to mention that you get to play with Mindbreak Trap to protect it. But most importantly, just because of the way that the deck works out most of the time, you don't put the Words of Wind into play until you're comboing off and bouncing their entire board, so they will have a very, very small window to destroy Words of Wind in the first place.Quote:
Originally Posted by sun tzu
The main reason why there's no real room for either Sensei's Divining Top or Ancestral Knowledge is due to the fact that you just don't need card draw or card selection; this deck already has the best card draw engine in the format in the form of 12 Enchantresses and the rest of the deck being enchantments. If it weren't so soft to non-Show and Tell based combo, this deck would just crush the entire field.
are there any other cards that are good against land-hate cards like rishadan port, wasteland, and their ilk? obviously wasteland isnt AS much of an issue if you are playing basics like you should be, but rishadan port specifically reallly hurts. i found sheltering prayers, which could work if you expect a bunch of hate that it deals with, but it seems kinda narrow to be honest. ideas anyone? suppression field could also help i guess.