Hi AIRchrist, and welcome back. I'm no expert with Burn, but I've been playing the deck off and on for some time now. I do however have some ideas regarding your questions:
IMO it is not necessary to play Fetch, so going for all mountains is a valid option.Originally Posted by AIRchrist
SDT helps you set up your draws. If you are planning to play SDT, however, Fetch comes in handy to give you a fresh three cards to look at.
I have cut it entirely from my deck. I think it is a matter of preference.
Again, preference. I change my main from time to time. Sometimes there are two, and sometimes I split them between main and SB. I like how it can be the last bit of damage, takes down little critters, and cleans up your GY.
Big creatures are everywhere, and swarm decks are not always present. This is also meta-dependant: if you have lot of Grixis or Elves, well, a Volcanic Fallout is an option.
PS: Thunderous Wrath is bad indeed, and 'good in Modern' does not mean it is a valid card for Legacy perseI would trade it for a 4th Fireblast.
"Be it ever so crumbled, there's no place like home."
RGCL (GQ)
Youtube-playlist dedicated to RGCL
Fetchlands in mono colored builds are mainly used to fuel Grim Lavamancer and Barbarian Ring, it's tough to hit threshold without fetches. I think they're worth using if you have SDT (a card I favor) but otherwise you probably don't want them in mono colored builds.
I do really like SDT, my recent project I've written volumes on so far suggests that SDT isn't very good but I'm not accepting of those results yet. SDT makes Thunderous Wrath better, but Wrath isn't all that good in the first place. It requires a lot to go right and it's just 1 mana more than your other options. Unlike Modern we have more 4 damage spells than we can possibly play, Wrath is just 1 point more with significant downside.
I'm a fan of Volcanic Fallout, I use it when sweepers are good such as against Goblins or Delver, but I also like it against Miracles. It not only messes up Monks but it's uncounterable Burn. I play 2 in my SB.
I really appreciate the responses. Ok, alright, I'll drop the Thunderous Wrath.
SDT seems like it could either be really good or really bad, depending on the meta. I'll probably throw a couple in and give it a whirl, but what would be the acceptable number of SDT's, 1-3? I imagine that it would be terrible to draw multiples. To those of you who use SDT, do you feel that it slows you down at all? Also, I'm really wanting to bring in Flame Rift, as it's incredibly powerful and I feel like it plays around Leyline of Sanctity pretty well(even though not all decks side it in against burn). If I cut the TW, one Fireblast and don't run GLM, would running 3 Flame Rift's be a viable addition to the deck?
And lastly, what is your take on Molten Vortex as a 1-of MB? Does it hurt or help Burn?
By the way, I've really enjoyed reading the statistical breakdown that some of you have been updating the thread with. Great information.
I promise I'll stop asking so many questions! Thanks again.
The biggest problem with Flame Rift is that you can only deal so much damage to yourself. With all the stat stuff I've been running lately, I would like to add in a self damage component but making a better AI has taken priority. In general though I think that you can really only afford to run either Flame Rift or Eidolon. I would definitely max out on Fireblast before Flame Rift though. If you find yourself looking for more 4 damage spells, I really like Atarka's Command or if you want to stay in mono red, Exquisite Firecraft is pretty fantastic and Sulfuric Vortex is always solid too.I really appreciate the responses. Ok, alright, I'll drop the Thunderous Wrath
SDT seems like it could either be really good or really bad, depending on the meta. I'll probably throw a couple in and give it a whirl, but what would be the acceptable number of SDT's, 1-3? I imagine that it would be terrible to draw multiples. To those of you who use SDT, do you feel that it slows you down at all? Also, I'm really wanting to bring in Flame Rift, as it's incredibly powerful and I feel like it plays around Leyline of Sanctity pretty well(even though not all decks side it in against burn). If I cut the TW, one Fireblast and don't run GLM, would running 3 Flame Rift's be a viable addition to the deck?
And lastly, what is your take on Molten Vortex as a 1-of MB? Does it hurt or help Burn?
I promise I'll stop asking so many questions! Thanks again.
I would stay away from Molten Vortex. There's just better mana sinks you can be using like Grim Lavamancer, SDT, more 3 mana cards, Barbarian Ring, or if you're trying to stick to a budget even Hellspark Elemental. Also, do you really want to hold lands in your hand on the off chance you draw your 1 of if the game goes long? Perhaps it's a playstyle thing on my end, but I would much rather get those lands on the board so I have the resources in the event I need to make a long string of plays in one turn. Games typically only go long if the opponent has some way to lock us out like a Chalice on 1, Counterbalance/Top, Chill, and so on. In those situations I don't want to be sitting on a 1 mana card that costs me resources. I want to play out my cards, so that I can end the game in a flurry of spells where I can get several cards to resolve.
Glad to hear people like the information. I don't want to completely flood the thread with it though as I know it really only appeals to a subset of players so I'm taking a few days off from posting more, and with any luck getting a better more dynamic AI built during that time (a real challenge, but I've got another 6 weeks of summer to try and make it work).By the way, I've really enjoyed reading the statistical breakdown that some of you have been updating the thread with. Great information.
Since the fetchland questions were coming up, I started looking at them in my program. I tested 20 to 24 lands builds with 16, 12, and 8 fetchlands (and the remainder being fetchables) over 1000 games each on a list that I thought was optimal with 22 lands.
The optimal build was 21 lands with 8 fetches. The next best build however was 22 lands with 16 fetches, and then at 22 lands as the fetch count decreased the deck speed decreased as well. I don't really have any rule of thumb to suggest here, but it looks pretty clear to me that the fetch count can be tweaked to make an optimal land count deck. If for example the optimal point is above 21 lands but below 22 lands, 22 lands with lots of fetches gets you closer to that point. The tradeoff is in mulligans, I've started tracking this number quite a bit and as the mulligan count increases the ratio of fetch:fetchable that you want to see in your hand changes. Obviously this only applies to Burn because we have little to no color fixing to worry about, but lands are much better than fetches when you go to 6 and 5.
If I could think of some good representative lists this is actually a project I would like to do. It's becoming very clear to me that the value of each card changes as the hand size changes. Creatures hold less value at 7 than they do at 5 (I think most had already figured this out), but the value of Burn spells also change and it appears the same is true of lands.
This most practical application of this is that the ideal list contains a mix of everything, so that you always have a chance of drawing to the higher value cards. I also suspect that there's some tuning that can be done here on the play/draw when sideboarding such as taking out one creature on the draw and bringing in an extra spell or taking out a spell for a creature on the play.
Got curious about a bit more deck manipulation. Since I already have ways in my program to scry and to cantrip I combined them together to try out Serum Visions. Preordain would perhaps be a little more relevant to Legacy but Visions sees more play so that's the question I wanted to answer. Currently, Ponder and Brainstorm are beyond my simulators ability to play because they require a more detailed way to rate cards than I can currently calculate (though the method I'm using for AI, should it work, will eventually enable them)
21 land (no fetches), 4 eidolon, 4 swiftspear, 4 guide, 4 price, 3 fireblast, 8 bolt, 1 firecraft, 2 sulfuric vortex. The remaining 8 slots depending on deck were 4 probe/4 serum visions, 4 serum visions/4 lightning bolt, and 4 lightning bolt, 3 vexing devil, 1 land.
The 8 cantrip list took 5.25 turns to win, the 4 cantrip list took 5.18 turns to win, and the 0 cantrip list took 4.96 turns to win.
Not much else to say here other than this being another nail in the coffin of cantrips in burn. When I looked at the average hands for the turn 4 games (nothing won on 3) the best hands had no cantrips in them, which is particularly telling since the mulligan decision doesn't bias against cantrips in any way (it actually considers them on par with Lightning Bolt). This data is starting to reinforce my opinion that SDT is working properly in code, and that it's just not that strong since similar cards are all evaluating poorly too.
Built a new AI to play the games, it runs a little slower but it's more versatile and the games themselves are a bit more representative of the average. Not only does it play better but the creature bias is tempered a bit because the AI considers things like over extending and not playing a creature if it doesn't increase the clock. It's not perfect but it works pretty well.
Giving the new system for a spin I had it experiment with Fireblast a bit. While more Fireblasts did lower the clock and enable earlier wins, the average turn didn't go down by very much between 0 and 4. This either suggests the cards I put in, in place of Fireblast are equal to it in power (I added Flame Rift) or that the sample size was too small. I'm going to run it again tomorrow with 25,000 games vs 5,000 to rule out the latter.
When it comes to deck manipulation, Serum Visions was superior to Magma Jet, which was superior to Gitaxian Probe but they were all worse than aggressive cards. There's been enough testing with deck manipulation at this point that I'm pretty skeptical of all of it.
I meant to get self damage in tonight but other things took my attention. So hopefully, starting tomorrow self damage will be a category that gets tracked and a number can be put to how much life we're giving up in each game to various cards.
hi guys!!
it was really long time from my last post here
btw burn is my pet deck, i play it from 2011 ,
i changed many thing
i played with all possible splash
but now, i have problem
my match up against eldrzi is really bad!!
when he start with chalice at 1 and maybe a thron , i lose at 99%
when side in, i put bridge, smash etc and go better..
but main deck very very very hard,
i start think to cut 1 playset burn (for example lava spike) for start paly main, something defensive cards...
maybe is crazy because we lose damage to fire opponent, but play maindeck something against bad match up, can be interesting...
i m crazy??
but for example
play maindeck bridge, can be good against many match up...
against eldrazi
against sneak attack
if even darck depth..
what do you think?
Have you ever cast the card Price of Progress ? Try that.
Sent from my XT1080 using Tapatalk
Soldier Stompy
Price of Progress is what you want.
---------------------------------------------------
Ok, so I figured out something else I can do with this dataset that's perhaps more useful than measuring games. I figured out that I can actually use it to measure cards, this is something of an offshoot from my previous few posts where I mentioned being able to track the average opening hand in the best performing lists. Using a little bit of math (and a spreadsheet to make life easy), it's possible to create a value that scores a card based on how much it over/under performs. In my system these results range from -1 (over performs the most) to 1 (big under performer).
I'll be breaking a rule I've tried to stick to with posting these simulator results, and I'll be giving more numbers than usual this time, mainly because I'm not quite sure of a good way to graph them. I started running various tests looking for more optimal lists. Currently I'm on the following list:
7 Fetch
16 Land
1 Barbarian Ring
4 Eidolon
4 Swiftspear
4 Goblin Guide
4 Vexing Devil
1 Grim Lavamancer
2 Price of Progress
2 Fireblast
1 Atarka's Command
9 Lightning Bolt
4 Exquisite Firecraft
1 Sulfuric Vortex
This list has an average turn of 5.19 which is down a bit from previous decks. But one thing you'll notice is that it's a bit creature heavy. While the new AI does limit this somewhat (it stops over extension) it's still something of a symptom of goldfishing rather than having real opponents.
The chart for the turns is
http://imgur.com/yuizCUc
A bit over 70% of games end on turn 5 or sooner.
Using the T3 information (about 1% of games) I'm able to generate the following score for each card in the deck. A lower number means the card is over performing relative to other cards in the deck, and a higher number means it under performs. It's a relative rather than absolute ranking, and should always sum to the average of 0 (minus rounding)
-0.75 Lightning Bolt
-0.47 Price of Progress
-0.43 Land
-0.23 Goblin Guide
-0.18 Atarka's Command
0.02 Grim Lavamancer
0.02 Sulfuric Vortex
0.07 Vexing Devil
0.12 Barbarian Ring
0.23 Fireblast
0.32 Fetch
0.37 Monastery Swiftspear
0.47 Eidolon of the Great Revel
0.47 Exquisite Firecraft
The T4 information changes a bit, which should be no surprise. As games go longer the value of each card is going to change.
-0.17 Land
-0.14 Eidolon
-0.11 Price of Progress
-0.05 Atarka's Command
-0.03 Bolt
-0.02 Guide
-0.01 Exquisite Firecraft
0.00 Fetch
0.02 Swiftspear
0.03 Barbarian Ring
0.04 Sulfuric Vortex
0.08 Fireblast
0.09 Grim Lavamancer
0.29 Vexing Devil
Finally, the T5 information
-0.36 Land
-0.19 Eidolon
-0.04 Fireblast
-0.03 Firecraft
-0.02 Fetch
-0.02 Ring
-0.01 Atarka's Command
0.00 Rift
0.00 Sulfuric Vortex
0.05 Swiftspear
0.06 Lightning Bolt
0.09 Grim Lavamancer
0.18 Guide
0.29 Vexing
Now, to turn all of those numbers into the useful information I'm sure you're waiting for I'm weighing each turn win by how common it is.
-0.05 Land
-0.04 Eidolon
-0.04 Price
-0.02 Atarka's
-0.01 Swiftspear
0.00 Bolt
0.00 Fetch
0.00 Firecraft
0.01 Barbarian Ring
0.01 Vortex
0.03 Fireblast
0.03 Lavamancer
0.09 Guide
0.09 Vexing
So we can see here that most cards are actually performing pretty close to the desired average. In a perfect world every card here would be near zero and that would indicate that the numbers are right on every card in the deck. We're not that far off though. I find this to be a pretty damn good case against Atarka's Command which is a card I've supported. It's just barely overperforming as a 1 of, but 2 gives the results that it's too many copies. Considering what this suggest the optimal fetch count is I see no way that the green splash is viable. Vexing Devil also looks to be slightly over rated at 4 copies. Removing one copy should increase the value of Guide and Lavamancer since they won't be competing as a T1 play. In exchange for that cut I'll add an additional price which is over performing and suggests it needs more copies. I'm hesitant to go up another land here and I think increasing the mana curve slightly will accomplish the same goal of balancing out the lands.
There's also a possibility that a 5th Eidolon in the form of a Pyrostatic Pillar could be a worthwhile addition over the Lavamancer. But, Pillar isn't in my simulator (yet) so that remains an unknown.
The biggest surprise to me here is that Exquisite Firecraft out performs Fireblast so consistently. With them both being 4 damage, I really thought Firecraft would be the weaker of the two.
What I'll probably do is run this simulation again on the draw (this was on the play) and figure out what should be done from there. And then make the suggested change and try again.
---------------------
On the draw results are in. -1 Vexing Devil for +1 Land looks like the highest recommended change. It also suggests -1 Lavamancer for +1 Bolt.
Just for fun, because I'm curious in how far apart the formulaic spreadsheet approach is from the simulation approach. I put this deck into my spreadsheet. The results are actually rather impressive. It comes in over budget on damage and under budget on mana, though it averages 1 turn slower than my sheet calculates. I think that's a point in favor of this simulation approach being somewhat representative of reality. Though I should stress that both are tuned for a goldfish. No promises as to when, but eventually there will be an opponent or two to choose from and at that point I expect to see cards like Searing Blaze (which is effectively just Incinerate here) and Grim Lavamancer increase in value.
Last edited by Brael; 07-19-2016 at 12:35 AM.
You are doing all this math and typing but show me the results that back up your numbers.
Sent from my XT1080 using Tapatalk
Soldier Stompy
In response to your latest analytical results, the Vexing seems to be a generally "not great" card for burn, but if these numbers are to be believed, it might have a place in creature heavy builds, in limited quantities. Does your simulator take into account life loss from fetches? Just curious. I'm honestly really interested in the Fireblast to Firecraft usage. While Fireblast has always been the card of choice(and I don't dispute its strengths), I'm wondering if some mix of the two would be better than 4 of Fireblast. In my latest real world testing, I've noticed that I've had better luck with Firecraft against the vast majority of my meta. I see a lot of Merfolk, Grixis Delver/Control, some Miracles and way too much Eldrazi. There have been a lot of games where I felt crippled by having 4 Fireblasts MB, but welcomed seeing a Firecraft from T3 on. I understand the power of FB, but it too often gets countered(feelsbadman) or stuck in my hand. Hitting a Thoughtnot or weaker Goyf without sacrificing lands and/or getting countered just feels like a safer bet in my current meta. I'm obviously not basing this off of hard numbers(I'll leave that to you), but I've noticed an uptick in my wins. If that means anything...
You wouldn't like my results, I only play small local tournaments, I don't go to anything large. And at that I only play Legacy about once every 2 weeks, and it's split between two decks that I like (Burn and Nic Fit). Even if I had some real results though, we're mostly talking about making changes that adjust the decks win turn by fractions of a turn. Some of those games you'll lose and of the ones you win, any typical tournament has you seeing between 10 and 50 games. Most of this stuff is really only apparent if you're tracking it, and even then over many thousands of games.
Bottom line, any results would be too small a sample size to say one way or the other. However, in my defense I'll say this: Frank Karsten came up with a near identical mana curve here, my spreadsheet suggests the build is good, and my simulator says the same. I probably don't have things perfect yet, but that's 3 independent sources that are all coming to the same conclusions through three different methods.
As far as being wordy, I can't help it, I have a real problem with being succinct, particularly when I'm talking about things I enjoy.
Life loss from fetches, PoP, Flame Rift, Eidolon, Barbarian Ring, and so on is not currently implemented. I've been saying for a week now that I would add it (maybe it's been two weeks?) but other little things always seem to pop up such as getting a new AI to run or in the case of the past 2 nights working the kinks out of the card by card analysis.
I too am surprised by the Firecraft numbers, the next time I get a chance to play Burn in paper it's something I'm certainly going to experiment with. Right now I have 4 MB but this program seems to suggest wanting them MB. Interestingly, 4 looks to be the right number so there's no Flame Javelins waiting in the wings as a 5th Firecraft. The land counts it suggests though I'm not at all shocked at (well, the fetch count a little), the 22-23 range is in line with every other test I've done on other decks that suggest a land count higher than conventional wisdom is typically a good idea. I am surprised at the Barbarian Ring count though. I really thought it would suggest more of them.
On the subject of Vexing Devil, it has actually scored quite well, it just doesn't score well as a 4 of. It could be an instance of my code being unfair to Vexing Devil or it could just be a proper reflection of the cards power. It doesn't get cast very often, and when it does it doesn't do very much. In my code on turn 3 or later, I assume it hits the board and then dies so it's a low priority to cast, on turns 1 and 2 though it hits the board. Problem is, Guide/Swiftspear are higher priority to cast because they have haste. So Devil is ideal on turn 1 and ok on 2, but bad beyond that. Devil will get more accurate representation once there's an opposing deck rather than a goldfish. But for now my program likes it most as a 2-3 of.
Edit: Played with some things a bit more. This deck scores near 0's across the board which is an indication that the numbers are all right.
7 Fetch
16 Mountain
2 Barbarian Ring
4 Eidolon
4 Swiftspear
4 Guide
1 Lavamancer
4 Price
2 Fireblast
11 Bolt
4 Firecraft
1 Vortex
Obviously, it needs some interaction, but goldfishes aren't very good at finding interaction. Short of an interesting question for this to generate numbers on I probably won't do too much more with it for the next few days/weeks as I'm going to be making some big changes to it for an actual opponent. I'm not quite sure how long that will take but it will probably keep me busy for a while. After that I'll almost certainly have to figure out some optimizations as well because an opponent means double the calculations, which means half the runtime and I'll want to get some of that time back.
Last edited by Brael; 07-19-2016 at 05:46 PM.
10 days with no post in this thread, time to fix that. I've had a few RL setbacks lately like my good computer getting stolen in a home invasion, and lots of travel time. But, I had the software backed up online and I'm still working on it. Though a larger project I had running that would take 3 weeks of dedicated running in order to test a huge range of decks has fallen apart since I don't have the computer to do it anymore.
Currently the next version of this simulator has an opponent that's mostly capable of interacting, but as I add more cards I keep having to reimplement large blocks of code to account for how they work. Right now I would say I have the counterspells working 100%, attacking/blocking 80%, and other spells 90%. There's just the odd bug that needs tracked down and fixed. Other than that, I just have to alter the database design so that it can record when the Burn player wins vs loses the game.
Once that's done I'll start posting a few more lists against a dummy deck that can interact and counterattack (I ended up going with a modified Delver deck based on what cards are easier than others to implement).
Terrible to hear, hopefully they can't get in the PC and find your personal data
Naa, they got all that stuff... quite an ordeal to get straightened out. My PC security was non existent on my desktop because I never considered someone would break in and steal it.
Anyways, it's taken me a bit but I think I finally have some working AI to make two decks interact. About every 2 days I've written and then rewritten it until I finally settled on something that's a good mix of simplicity (important for debugging) and solid strategy. I can't even begin to go into how much reading I did between articles on basic/advanced aspects of combat, to computer algorithms, I even looked at 4 different AI's others have already made trying to make something reasonable at combat.
In the end, my AI is pretty simple but I think it has a solid strategy behind it. Where most AI's look for reasons to include cards in an attack/block, which leads to millions of permutations that have to be compared I took the opposite approach. Everything that can attack, attacks. Everything that can block, blocks. Once a card has been identified as a potential attacker/blocker I look for reasons to exclude it. Mine is pretty simple but I think it has a solid strategy behind it. At a minimum I think it's good enough to beat novice players, and the beauty of my approach is that it only requires a quantity of games, not quality. 10,000 games worth of data generated by Finkel playing Budde is as good as 1,000,000 games of two brand new players, so simply generating more data makes up for any deficiencies in the AI's skill.
I was able to speed things up a bit too, it takes about 2 hours to run 100,000 games where previously that would take 5 hours.
The opposing player used is a generic Delver deck. For reasons I won't explain (PM me if you're really curious) certain cards are more difficult to program than others, so that limits my card pool a bit. Basically, on action triggers are still beyond my programs capabilities which excludes Young Pyromancer, in depth card evaluation is too time consuming (in particular shuffling away bad cards) which excludes Ponder and Brainstorm, and alternate costs are close to impossible to do in an elegant fashion which excludes Force of Will. This causes the test deck to be some weird hybrid that falls between Legacy and Modern legality. It also doesn't really impact testing because counters, removal, and blockers are still counters, removal, and blockers even if they exist at a lower quality, and really it's the interactions as they come up that are being measured. The "final" UR Delver list being used is
8 fetches
10 fetchable land
4 Monastery Swiftspear
4 Stormchaser Mage
4 Delver of Secrets
4 Flame Rift
2 Fireblast
6 Lightning Bolt
4 Gitaxian Probe
4 Preordain
4 Serum Visions
1 Exquisite Firecraft
1 Cryptic Command
3 Counterspell
2 Force Spike
After looking at the data, and filtering the results to only the games Burn won (it won ~67% of games btw), using the previous list I gave which was the optimal performing goldfish list we get the following results, looking only at T3/T4/T5 wins (9339 games over the initial dataset):
0.37 Fetch
-0.62 Land
-0.01 Ring
-0.13 Eidolon
0.15 Swiftspear
0.15 Goblin Guide
-0.06 Rift
0.01 Fireblast
0.05 Bolt
0.01 Firecraft
0.07 Lavamancer
-0.01 Vortex
So from this we can say that the big under performer was fetchlands, the card the deck most wanted was lands (note that it's already running 25, and this says that even more are still optimal), another Eidolon would be fantastic (already running 4, but maybe a Pyrostatic Pillar?), and the 1 drop hasted creatures are a little over represented.
So I think that for the next test I'm going to try -1 Fetch, -1 Swiftspear, -1 Lavamancer, +2 Land, +1 Eidolon (for now it will just be 5 Eidolons, I'll downgrade to a more accurate Pyrostatic Pillar in the future, the next time I add new cards to the pool).
As an aside, for a bit of trivia, I expect Eidolon to consistently overperform in the future. In fact, my original test deck for the opposition was 40 lands, 20 Eidolons and it won 80% of games despite the mulligan code almost always forcing it to 4 card hands.
@ Brael: out of curiosity, could you also build software to test certain configurations vs Miracles? Or better, test hypothetical decks which whoop Miracles, and at the same time have a decent game vs the rest of the field? Just out of curiosity, not to derail this thread.
"Be it ever so crumbled, there's no place like home."
RGCL (GQ)
Youtube-playlist dedicated to RGCL
Maybe eventually. Miracles is a very different sort of deck though. I chose Delver as the test deck because it does a little bit of everything. Attacking, blocking, removal, burn, counters, and it's aggressive enough that it uses a fairly similar AI all while having cards that are either good/bad regardless of context.
Miracles is a very different sort of deck. It involves lots of Top (which I've since broken in my program), alternate costs (Snapcaster/FoW), and at any given time the power level of a card is very contextual. It's a bit beyond my abilities right now and it wasn't really even registering as a long term goal. My very long term goal is to make a generalized AI, hook it up to a group of draft bots, and see if it can solve a limited format before a PT so I'm primarily looking at decks that like to attack and block.
For what it's worth though, I think Burn has a pretty reasonable Miracles matchup. Especially if you apply one of the big takeaways from this program which is that Exquisite Firecraft is 4 of MB material.
Also, for what it's worth since I know people do like having real world results to back up theory. I played my current list (posted below) which is very heavily based on what this program said was goldfish optimal, for 12 hours straight the other day against a couple willing test subjects with them using a mix of Shardless BUG, BUG/RUG Delver, Nic Fit, S&T, and Miracles. I didn't keep precise numbers, but I played each matchup for about 2 hours split 1 hour preboard/1 hour post board. The deck performed very well. The biggest thing I noticed was that Barbarian Ring only turned on once the entire time, but the time it was on, it won me a game that no other card could have won.
Additionally, in the past couple weeks I've played 2 small local events, 3 rounders and went 2-1 and 3-0. I'm either on a hot streak or the build is matching the theory behind it.
Here's what I played
25 Land
16 Mountain
8 Fetch
1 Barbarian Ring
4 Goblin Guide
4 Monastery Swiftspear
4 Eidolon of the Great Revel
1 Grim Lavamancer
4 Lightning Bolt
4 Chain Lightning
4 Price of Progress
3 Searing Blaze
4 Exquisite Firecraft
1 Sulfuric Vortex
2 Fireblast
Sideboard
3 Ensnaring Bridge
2 Volcanic Fallout
2 Red Elemental Blast
1 Flame Rift
2 Faerie Macabre
2 Sulfuric Vortex
1 Fireblast
2 Skullcrack
The thing that boggles my mind about your list is your high land count. How are you not flooding out ~30 percent of the time? There is no library manipulation in Burn, barring half-measures like Magma Jet, so you get what the deck gives you. I played Burn for about six months straight when I started playing Legacy eight years ago, and the optimal number I found for my deck was 19.5, so I ran 20. Looking at your mana curve, it's almost the same as the one I ran. My curve may have even been a hair higher.
I've done a bit more testing with additional decks, that was the optimal amount in a goldfish, but in actual gameplay the fastest deck was at 24 lands (I mostly post these results as I find them, so sometimes after I write something I find new, more reliable information that contradicts it... though I usually go back and edit in these instances).
I've been thinking about your question for a while though, for a month actually, because I've seen the high land count thing before. My best explanation is that curving out is simply worth having more land. Note that I'm including 3 drops in the deck. Having more land represents more resources, this is basically a turn 5 deck (it has some hands that win on 3 and 4, but the highest number is at 5, at least with my software playing... it plays slower than a real person does). With a low land curve, you're expecting your mana to develop along the lines of 1-2-2-3-3 but with more land you can expect it to go 1-2-3-4-4. Over 5 turns that's a difference of 11 vs 14 mana to spend, or 27% more resources. In practice it works out to a bit more than 27% because 4 land turns on Fireblast #2 (which you'll also see I was finding diminishing returns on). Getting 27% more mana means that until you're out of cards you're also playing your hand out 27% faster. In order to do so you're only giving up 8% of your deck.
Having more mana, means that until you're out of cards, you're also getting damage out faster. In addition to that it means your hand is larger because you're taking fewer mulligans.
This isn't something I've seen exclusive to Burn either. This simulator is my third attempt at such a project (the first two played other decks), and in every case they all came to the same conclusion, more lands than traditional wisdom dictates is better. I've since applied this to a fourth deck with Infect and it also came to the same results. I'm not 100% sure on the reasoning for it, but I've run literally over 100 million games of Magic now across multiple decks, with different logic sets and they all come to the same conclusion.
Though Burn remains my only deck where I've actually used an opponent rather than a goldfish (and the opponents are still pretty new at that). Interestingly, adding the opponent reduced the optimal land count slightly I think, but it also could have been because I cut the fetches entirely (~12 fetches seem to be worth -1 land).
As far as not flooding out goes, I think it's mostly just a matter of perception. When people play an extra land, say 21 over 20 the times they flood stick out in their head, but in reality they're only seeing an additional extra land in 1 in 60 cards and many of those flooding instances still would have happened with 20 or 21. Humans are very bad at simply observing statistics in action, it's really only in measuring this stuff over many thousands of games (and even my results are cutting the number of repetitions needed short) that a true pattern can be seen.
To give an example of this, a couple hours ago I ran a set of 1000 games at 23 land and a set of 1000 games at 22 land (1 Ring the rest Mountains). The 23 land deck had an average turn of 4.162 but the 22 land deck had an average turn of 4.172. From a handful of games even if you're very observant, you simply don't have the sample size to definitively say that 1 fewer land is leading to your deck being slowed down by 1/100th of a turn.
There are other considerations as well, it's something I haven't gotten into much yet with Burn since adding an opponent, but often times cutting a land for a silver bullet can be very beneficial, because that 1/60th chance of drawing a super relevant card outweighs the 1/100 chance of ending the game a turn sooner. This is something that most frequently comes up with modal cards, but because I'm playing with a limited card pool I don't have any of those... yet. It's in this area that I think the best arguments for cutting lands come into play, and the sort of reason I'm a fan of Atarka's Command... but RG is wholly inferior to mono red. But even with non modal cards it's important to keep in mind for 1 drops, and a large part of the logic behind why I like to play a 1 of Grim Lavamancer.
I have a few ideas in mind for why I'm getting the results I am, which I'm trying to rule out or confirm 1 by 1. I've tried to keep the topic of my posts on the subject on the results themselves instead of the whys behind them though, because I don't want to spam the thread with theory more than I already am and the math/cs process underlying what I'm doing isn't really the sort of thing people are reading the thread for.
There are currently 1 users browsing this thread. (0 members and 1 guests)
Reply With Quote