That is already 8 cards you have listed assuming the "bunch of equipment" is one other equipment. I count 12 damage as the most you can do in this scenario.
So why 7 lands? It's not the best number for any goal I can think of, and I think that having 8 lands gives you the best chance of having exactly 1 land in your hand, which seems ideal.
OK, I think you might want to do something like this, if this were actually feasible (which it is clearly not):
Round 1: Run every possible deck against every other possible deck (1 game). The AI for round 1 plays in a vacuum (that is, it doesn't make assumptions about what cards would be in a certain deck given the cards that it already knows). Take the top 10% of decks from this round and go to round 2.
Round 2: Run the top 10% of decks against each other (again, 1 game), using the same AI as we used in Round 1. Again, take the top 10% of decks and go to round 3.
Round 3: Same as above rounds.
Round 4: Same as above rounds.
Continue until round 7 or so. We now have a semi-reasonable number of decks (I'm too lazy to calculate, but whatever). All of the decks in this round will probably be able to cast most of their cards. Now, instead of playing 1 game, we play best 2 out of 3, and take the top 10%.
Round 8: Still best 2 out of 3, taking the top 10%
Round 9: Same as above.
Continue until about round 20-ish (again, dependent on the numbers which I don't want to run). Now, we should actually have decent decks. We now change our AI. This AI learns from previous games. If any of you have heard of Tom Murphy's work with a learning program (which can be found here if you want to learn more), we're doing something like that, albeit much more complex. Our program will run about 10 rounds best 2 out of 3 without making cuts, which will teach it which cards work well and which ones don't.
We continue in this manner until only a few decks remain (350-ish is ideal). Then, our AI runs each deck against each other deck, and calculates win percentages until it can develop a metagame, where the prevalence of a deck is (deck's win percentage)/(all win percentages). Now, every possible combination of these decks is played against each other (best 2 out of 3 played 1000 times), and each set of matches contributes a certain "weight" to a deck's win percentage. After all matchups are run, the prevalence is recalculated, and we continue until our metagame is stable. Now, the deck with the highest win percentage in our simulated metagame is the best deck.
Of course, this is entirely unrealistic, and probably isn't the best way to approach the problem either. However, we WILL get decent results with this approach, which is good.
So I just got MTGO (or MODO, whatever you like to call it), and I decided to get UW Heroic as my first deck, because it's super cheap. So, I decided to play Dance of the Skywise as a 1-of just for testing. I drew it twice in 3 matches, and it won the game both times I drew it. Could this actually be a playable card, or am I just having weird luck?
That on-the-draw Win% vs. Abzan is very interesting. This data may do a lot more than just decide the Muddle the Mixture vs. Serum Visions debate if this trend holds
It seems like people are debating whether or not to play 1x Blood Artist. As a 1-of, it's not going to be much good. However, I don't see why playing 3-4 would be bad. In multiples, they make boardwipes much less attractive, any non-Path to Exile spot removal is bad, and blocking often becomes terrible. It also gives you a viable alternate win-con by saccing a ton of stuff to Viscera Seer/Cartel Aristocrat(/Bloodthrone Vampire if you want to play that). Has this been tested?
Well, I have some old playtest data from playing 100 games on the draw. It's using an outdated list which I think was goldfishing s bit slower than my current list, although I haven't actually done 100 tests for my current list, but here it is:
Turn 1: 2
Turn 2: 27
Turn 3: 33
Turn 4: 23
Turn 5: 9
Turn 6: 5
Turn 7: 1
This data is certainly not perfect, but it's at least close to accurate.
Maybe I'm reading this wrong, but does this mean you never fizzled once in all those games? I'm currently estimating fizzle rates when you try to go for turn 2 wins, and it is definitely a nonzero percentage. How are fizzles accounted for in this data?
I fizzled a lot, actually (mostly due to bad sequencing while I still didn't know the deck), but I wasn't completely unable to combo. Games like the T7 win were due to me going for T1-T2 kills and fizzling.
Well, I have some old playtest data from playing 100 games on the draw. It's using an outdated list which I think was goldfishing s bit slower than my current list, although I haven't actually done 100 tests for my current list, but here it is:
Turn 1: 2
Turn 2: 27
Turn 3: 33
Turn 4: 23
Turn 5: 9
Turn 6: 5
Turn 7: 1
This data is certainly not perfect, but it's at least close to accurate.
This is a hypothetical situation that my brother and I were discussing.
So, let's say I control a Puresteel Paladin and I play a Spidersilk Net or some other equipment. Somehow I don't realize that my library has no cards, and I reach my hand down to draw, only to find that there's nothing there. Have I indicated that I wish to use Puresteel Paladins ability to draw, thus losing the game (is it even possible to lose the game this way)? We decided that it was possible to lose this way, but the player would need to verbally announce that they were choosing to attempt to draw. Is this correct?
Round 1: Budget Homebrew BR Minotaurs
Not much to say here. I was careful with my spoils and this was pretty easy. G1 I played Spoils of the Vault at the end of his 3rd turn to win on T4, G3 I had two natural Paladins by T3. 2-0, 1-0
Round 2: ?
Round 2 was a homebrew that I can't really remember. G1 was a T3 win by natural Paladin. G2 I was at 5 life facing down 10-ish power after fizzling T3. I had 7 equipments on the board, and a hand full of lands+Grapeshot. I cracked an EOT Flooded Strand, drew Sigil of Distinction into Retract, and won. 2-0, 2-0.
Round 3: Rg Goblins
Goblins with a green splash (I'm not sure what the splash was for other than Destructive Revelry out of the board). He won the roll, kept on a mull to 6, played a Bloodstained Mire and passed. I played a Gemstone Mine and passed. T2 he played a Legion Loyalist, pinged for 1, and passed. I played a Plains and passed. T3 he played a Dragon Fodder, swung for 1, and passed, missing his land drop. I comboed T3 to win. G2 I brought in 2x Silence and 3x Leyline of Anticipation. I was on the fence on Pithing Needle for Krenko, but I decided against it. He brought 2x Blood Moon and a ton of artifact hate, presumably for Opals. I draw a hand with a likely T2 win through natural Paladin, pretend to think about it for quite a while, and eventually announce that I will keep. He plays a T1 Wooded Foothills and passes the turn. I play a Gemstone Mine and pass, while he cracks for a Stomping Ground. T2 he plays a Goblin Guide, hitting me for 2 and revealing a Plains. He then plays a Bloodstained Mire (I think) and passes. I play the Plains and pass. He finally taps out for a Goblin Chieftain to hit me for 5 and passes. I play my Puresteel Paladin and combo to win. 2-0, 3-0
Game note: My acting was not particularly good, but I don't think he expected it out of a teenager, so it worked fine. Baiting red opponents into tapping out is very important with this deck.
All in all, I'm pretty happy with the deck, but I think my particular list needs some work. Here is a list of notes for improvement I remember:
- Gemstone Caverns should be replaced. It's fun to get those T1 wins, but it just comes up at the worst times and isn't good for the deck.
- The sideboard should be really varied. I couldn't sideboard as many cards as I had wanted to because I needed to avoid diluting the combo too much. I want to play 3x Silence, 2x Angel's Grace, 2x Wear // Tear, and a bunch of 1-ofs. That way, I'll be able to consistently combo while having some hate cards in every G2.
- Noxious Revival is probably very good here. My copies didn't come for some weird reason, but I will be playing 2-3 as soon as they do. However, I don't think 4 is the right number.
Also, I'd like to know what the best way to beat Chalice of the Void is, as I can't seem to find anything that really helps. Suggestions?
Yes. Before steps or phases end, each player must pass priority in succession, starting with the active player. So you get priority before end step, kill the thing, and then everyone gets priority again.
Oops I forgot a card because I'm dumb. There should be a Puresteel Paladin after the Kite Shields.
Pithing Needle
Isamaru, Hound of Konda
Round 1: Run every possible deck against every other possible deck (1 game). The AI for round 1 plays in a vacuum (that is, it doesn't make assumptions about what cards would be in a certain deck given the cards that it already knows). Take the top 10% of decks from this round and go to round 2.
Round 2: Run the top 10% of decks against each other (again, 1 game), using the same AI as we used in Round 1. Again, take the top 10% of decks and go to round 3.
Round 3: Same as above rounds.
Round 4: Same as above rounds.
Continue until round 7 or so. We now have a semi-reasonable number of decks (I'm too lazy to calculate, but whatever). All of the decks in this round will probably be able to cast most of their cards. Now, instead of playing 1 game, we play best 2 out of 3, and take the top 10%.
Round 8: Still best 2 out of 3, taking the top 10%
Round 9: Same as above.
Continue until about round 20-ish (again, dependent on the numbers which I don't want to run). Now, we should actually have decent decks. We now change our AI. This AI learns from previous games. If any of you have heard of Tom Murphy's work with a learning program (which can be found here if you want to learn more), we're doing something like that, albeit much more complex. Our program will run about 10 rounds best 2 out of 3 without making cuts, which will teach it which cards work well and which ones don't.
We continue in this manner until only a few decks remain (350-ish is ideal). Then, our AI runs each deck against each other deck, and calculates win percentages until it can develop a metagame, where the prevalence of a deck is (deck's win percentage)/(all win percentages). Now, every possible combination of these decks is played against each other (best 2 out of 3 played 1000 times), and each set of matches contributes a certain "weight" to a deck's win percentage. After all matchups are run, the prevalence is recalculated, and we continue until our metagame is stable. Now, the deck with the highest win percentage in our simulated metagame is the best deck.
Of course, this is entirely unrealistic, and probably isn't the best way to approach the problem either. However, we WILL get decent results with this approach, which is good.
So I just got MTGO (or MODO, whatever you like to call it), and I decided to get UW Heroic as my first deck, because it's super cheap. So, I decided to play Dance of the Skywise as a 1-of just for testing. I drew it twice in 3 matches, and it won the game both times I drew it. Could this actually be a playable card, or am I just having weird luck?
I fizzled a lot, actually (mostly due to bad sequencing while I still didn't know the deck), but I wasn't completely unable to combo. Games like the T7 win were due to me going for T1-T2 kills and fizzling.
Turn 1: 2
Turn 2: 27
Turn 3: 33
Turn 4: 23
Turn 5: 9
Turn 6: 5
Turn 7: 1
This data is certainly not perfect, but it's at least close to accurate.
So, let's say I control a Puresteel Paladin and I play a Spidersilk Net or some other equipment. Somehow I don't realize that my library has no cards, and I reach my hand down to draw, only to find that there's nothing there. Have I indicated that I wish to use Puresteel Paladins ability to draw, thus losing the game (is it even possible to lose the game this way)? We decided that it was possible to lose this way, but the player would need to verbally announce that they were choosing to attempt to draw. Is this correct?
Thanks.
List:
4 City of Brass
1 Flooded Strand
1 Gemstone Caverns
4 Gemstone Mine
1 Godless Shrine
1 Hallowed Fountain
1 Plains
1 Polluted Delta
1 Seachrome Coast
Enchantment
2 Myth Realized
Artifact
4 Accorder's Shield
4 Bone Saw
1 Golem-Skin Gauntlets
4 Kite Shield
4 Mox Opal
4 Paradise Mantle
4 Sigil of Distinction
4 Spidersilk Net
2 Grapeshot
Creature
4 Puresteel Paladin
Instant
4 Retract
4 Spoils of the Vault
3 Leyline of Anticipation
1 O-Naginata
3 Pithing Needle
2 Silence
2 Storm Entity
1 Deflecting Palm
3 Wear // Tear
Round 1: Budget Homebrew BR Minotaurs
Not much to say here. I was careful with my spoils and this was pretty easy. G1 I played Spoils of the Vault at the end of his 3rd turn to win on T4, G3 I had two natural Paladins by T3. 2-0, 1-0
Round 2: ?
Round 2 was a homebrew that I can't really remember. G1 was a T3 win by natural Paladin. G2 I was at 5 life facing down 10-ish power after fizzling T3. I had 7 equipments on the board, and a hand full of lands+Grapeshot. I cracked an EOT Flooded Strand, drew Sigil of Distinction into Retract, and won. 2-0, 2-0.
Round 3: Rg Goblins
Goblins with a green splash (I'm not sure what the splash was for other than Destructive Revelry out of the board). He won the roll, kept on a mull to 6, played a Bloodstained Mire and passed. I played a Gemstone Mine and passed. T2 he played a Legion Loyalist, pinged for 1, and passed. I played a Plains and passed. T3 he played a Dragon Fodder, swung for 1, and passed, missing his land drop. I comboed T3 to win. G2 I brought in 2x Silence and 3x Leyline of Anticipation. I was on the fence on Pithing Needle for Krenko, but I decided against it. He brought 2x Blood Moon and a ton of artifact hate, presumably for Opals. I draw a hand with a likely T2 win through natural Paladin, pretend to think about it for quite a while, and eventually announce that I will keep. He plays a T1 Wooded Foothills and passes the turn. I play a Gemstone Mine and pass, while he cracks for a Stomping Ground. T2 he plays a Goblin Guide, hitting me for 2 and revealing a Plains. He then plays a Bloodstained Mire (I think) and passes. I play the Plains and pass. He finally taps out for a Goblin Chieftain to hit me for 5 and passes. I play my Puresteel Paladin and combo to win. 2-0, 3-0
Game note: My acting was not particularly good, but I don't think he expected it out of a teenager, so it worked fine. Baiting red opponents into tapping out is very important with this deck.
Round 4: UW Gifts Tron
I'm not sure how this MU goes, but I think it's favorable for us. However, this being the deck that it is, there is a lot of variance. The following match is a perfect example of why not to play this deck.
Game 1 I mull to 4 before deciding it's not worth it to go further, keeping 2 equipments, 1 Retract, and 1 Seachrome Coast. I actually had an almost keepable 7, so I'm really regretting not taking that. I can't draw into Paladin (the variance hurts), and he hardcasts an Elesh Norn, Grand Cenobite. I use a Retract to kill it with a Grapeshot, but he Gifts Ungivens for an Iona, Shield of Emeria naming red, and quickly wins.
Game 2 I bring in 2x Silence, 2x Storm Entitys, and 3x Wear // Tears. I saw this guy's board when he played Storm in Round 3, and he brought in 13 cards, so I'm assuming that he has Leyline of Sanctity (turns out he doesn't, but oh well) and Rule of Law, and I know that he has Mindbreak Trap and Chalice of the Void. I bring out Gemstone Caverns, all my alternate win-cons, a Spoils of the Vault, and some equipments. I try to combo T3, but I fizzle, and my Storm Entity at 7/7 gets Mindbreak Trapped. However, I draw into a Retract to win.
Game 3 I bring out a single Wear // Tear for a O-Naginata, and bring my Storm Entitys back out for Grapeshots. However, I mull to 3 to get nothing, and he's able to cast Gifts Ungiven at the end of my T4 for Unburial Rites+Iona, Shield of Emeria. I have a Gemstone Mine, a Mox Opal, and a Gemstone Caverns (no luck counter) on the battlefield, with 2 equipments and a Silence in my hand. I need a Puresteel Paladin on top plus a bunch of other stuff, but it's worth a shot. I Silence on his upkeep, he responds by Disenchanting my Mox Opal, and I lose. The top of my deck was a Spidersilk Net. 1-2, 3-1.
All in all, I'm pretty happy with the deck, but I think my particular list needs some work. Here is a list of notes for improvement I remember:
- Gemstone Caverns should be replaced. It's fun to get those T1 wins, but it just comes up at the worst times and isn't good for the deck.
- The sideboard should be really varied. I couldn't sideboard as many cards as I had wanted to because I needed to avoid diluting the combo too much. I want to play 3x Silence, 2x Angel's Grace, 2x Wear // Tear, and a bunch of 1-ofs. That way, I'll be able to consistently combo while having some hate cards in every G2.
- Noxious Revival is probably very good here. My copies didn't come for some weird reason, but I will be playing 2-3 as soon as they do. However, I don't think 4 is the right number.
Also, I'd like to know what the best way to beat Chalice of the Void is, as I can't seem to find anything that really helps. Suggestions?