[Deck] Goblins

[Dice_Box]

It can be yes, but since this was a last minute choice (It was only meant to be a joke) I could only use what I had. I do not own Living Death so I used bidding. The last game I played, the first round I would have won with Death.

[GoboLord]

Hello fellow Chieftains,

over the course of the year I published the results of an analysis of an ever-growing dataset in three parts (reference: pt.1, pt.2, pt.3). The purpose of these analyses was to find out how single cards determine the performance of our deck (Vial Goblins). I found out that there were 3-4 different kinds of decklists (read more about that below) and that some cards affected the performance of some decklists more than those of others. One of my conclusions in the last article was that more data is needed to get a more complete picture of how and why some cards influence the performance of the respective decklists. It seems that, almost half a year later, you guys fed my dataset with a sufficient number of results to achieve exactly that.

“So, what is today’s article about?” you ask. Today I’m going to present you – for each of the 4 versions of the deck – how the number of copies of certain cards affect each decklist’s performance separately.

For starters here are the definition of the 4 decktypes.

(1.1) Mono R CLASSIC Goblins
You can think of the deck as the stereotype mono-colored Goblin decklist with some number of Mogg War Marshal, Rishadan Port and the like. Decklists with any dualland, Warren Instigator or Thalia in the maindeck were excluded here.
The set includes 143 decklists.

(1.2) CLASSIC Goblins with a splashcolor (most often green or white)
Same as above, only that decklists with duallands were included. Again: no Thalia, no WInstigator here.
N= 173

(2) Rw(g) THALIA Goblins
Basically every decklist with Thalia in the MD but without Warren Instigator.
N= 131

(3) WInstigator Goblins
Any decklist with Warren Instigator.
N= 92

Okay. Next comes a list of cards and I will explain which number of copies of the respective card is able to improve the performance of the decklist. i.e. the following list gives answer to the question “How many copies of [card] can improve the performance od [decklist]?”.

Rishadan Port
Mono R CLASSIC: decklists with 3 copies perform above average. Those with 4 copies perform below average.
Multicolor CLASSIC: lists with 2 and 3 copies perform above average, lists with 4 copies perform below average.
THALIA: lists with 4 copies < average.
WINSTIGATOR: 0 or 3 copies > average
Conclusion: It seems as if decklists with 4 copies of Rishadan Port tend to perform worse than those with 2-3 copies. However, it is a good idea to include more than 1 copy in every non-Winstigator decklist.

Goblin Piledriver
Mono R CLASSIC: 3 copies > average; 4 copies < average
Multicolor CLASSIC: 4 copies > average; 3 copies < average
THALIA: 0 copies > average; 3 copies < average
WINSTIGATOR: 4 copies > average; 3 copies < average
Conclusion
The optimal number of Piledrivers is dependent on the decklist.

Gempalm Incinerator
Mono R CLASSIC: 3 copies > average; 4 copies = average
Multicolor CLASSIC: 3 copies > average; 4 copies = average
THALIA: 4 copies > average; 3 copies < average
WINSTIGATOR: 2 and 3 copies = average
Conclusion
Decklists with 3 copies of Gempalm Incinerator perform better than those with 4 copies; and this is true for every type of list. For WInstigator lists it does not seem to matter much if they run 2 or 3 copies.

Mogg War Marshal
Mono R CLASSIC: 3 and 4 copies = average
Multicolor CLASSIC: 4 copies > average; 3 copies = average
THALIA: 4 copies > average; 3 copies < average
WINSTIGATOR: does not apply
Conclusion
Decklists with 4 copies of Mogg War Marshal perform above average; and this is true for every type of list. However, CLASSIC lists seem to be comfortable also with only 3 copies.

Goblin Chieftain
Mono R CLASSIC: 1 copy > average; 0 copies < average
Multicolor CLASSIC: 0 and 1 copies > average
THALIA: 3 copies > average; 0 copies = average
WINSTIGATOR: 3 copies > average; 4 copies < average
Conclusion
CLASSIC lists with 1 copy of Goblin Chieftain perform better than average; Thalia and WInstigator lists seem to perform at their best with 3 copies.

Tin Street Hooligan/Tuktuk Scrapper
Mono R CLASSIC: 0 Tuktuk Scrapper > average; 1 Tuktuk Scrapper < average
Multicolor CLASSIC: 1 Tin Street Hooligan > average; 0 Tin Street Hooligan < average
THALIA: 1 Tin Street Hooligan > average; 0 Tin Street Hooligan < average
WINSTIGATOR: 1 Tuktuk Scrapper > average; 0 Tuktuk Scrapper < average
Conclusion
Decklists with Tin Street Hooligan perform better than average. This is not true for Tuktuk Scrapper.

Skirk Prospector and Goblin Sharpshooter
Mono R CLASSIC: the most successful combination seems to be 1 copy of Skirk Prospector and 1 copy of Goblin Sharpshooter
Multicolor CLASSIC: the most successful combination is a 1-0 split
THALIA: 0-0 or 1-0
WINSTIGATOR: does not apply
Conclusion
While 1 copy Skirk Prospector seems to improve the performance of any decklist this is not generally true for Goblin Sharpshooter.

Krenko, Mob Boss and Siege-Gang Commander
Mono R CLASSIC: 1-2 or 1-1 split
Multicolor CLASSIC: 0-1 or 0-2
THALIA: 1-1 or 1-0
WINSTIGATOR: 0-2
Conclusion
While these results do not give us the possibility to tell which one of those is a better choice for the slots that they are competing for all-so-often, it seems to be clear that most successful lists run at least one copy of Siege-Gang Commander.

Other interesting facts
* lists with 4 copies of Thalia, Guardian of Thraben performed better than those with 3 copies
* lists with 4 copies of Warren Instigator performed better than those with 3 copies
* lists with 1 Karakas performed better than those with 0 copies



Some words at the end
Just like in past articles I want to point out that it is not possible to “calculate” the best decklist. Statistics like these give you information, which have no meaning without your interpretation. Speaking of interpretations: this time I am not going to give you my interpretation of the result, because I want to give you the chance to work with them for yourselves and discuss them with others in the forum.


I hope you enjoyed reading the article. Questions are, as always, welcome.

Merry Christmas,

-GL

[LeoCop 90]

Thanks for the article GoboLord ! It's a nice Christmas gift !

One thing that i noticed is that thalia lists perform better with 4 gempalm incinerator. This seem strange to me because playing thalia maindeck reduces your goblin count, so i'd be more inclined to play some tarfires instead of relying completely on gempalms. However, thalia lists perform better also with 4 mogg war marshals, so i think that they mitigate the problem providing 2 tokens for gempalm.

Another thing i would like to know is which number of lands is better to play (22 or 23). I assume that lists with 3 ports play 22 lands, and 23 lands list include the 4th port.

[Olaf Forkbeard]

Just out of curiosity do you have this database compiled in Excel or some other program? Secondly could you make it accessible to the public? I realize we could each go and pick out each report but why re-invent the wheel here. I'm hoping to have a look at it myself and see if I can draw my own conclusions. Crowd sourcing could help out here is the thought.

For instance. Multicolored lists that played 3 Gempalms peformed better than lists that played 4, but no information is displayed about if they ran any other removal. Instead of querying you, we could query the database.

[prateta]

-1 shattering spree +1 tuk tuk scrapper (spree is needed only against affinity/MUD, while the second scrapper is very good against blade decks.)
-1 rest in peace +1 grafdigger's cage (faster hate against reanimator and dredge, with the bonus of being a pain for elves. This is the change i would make also because i wouldn't side in all 4 rest in peace against goyfs decks, 3 are enough for me.)

Another card to consider is ethersworn canonist, you can do a mix between her and chalices, but in the end is just personal preference.

With this configuration you have a fighting chance against everything (i assume you are maindecking thalia). The worst matchup should be show and tell, but to improve that you should include some Red elemental blasts or Ashen Rider/Confusion in the ranks. I suppose the best plan is just to dodge it.

Hello, thanks very much for your advice!

so I'm in Vienna right now and I started scouting the meta. Firstly I wanted to attend one FNM draft and just ask people, then I found out that they publish all decklists + legacy meta analysis every 5 months on their website. Excellent! :-D now I can come with exactly right sideboard for this meta. That's the good news. The bad news is, they like sneak show and omnitell here in Austria. No way to dodge it, last 5 tournaments it was every time in top 4.

So I think I'll be bringing back that Stingscourger MD (I cut him off because czech meta ain't full of SnT decks and he was kinda useless most of the time). I also have this playset of REB, so I guess I'll be playing that too. Do you think 1 sting MD and 4 REB SB is enough hate versus Show & Tell decks? Or should I buy like 2 Confusion in the Ranks also?

As has been suggested by other posters in the past, I prefer to split my graveyard hate as 2 RiP & 2 Relic of Progenitus since a 2nd RiP in play in just insurance against the 1st getting destroyed, but RiP + Relic gives you one more potential draw.

If you'd like to be better prepared for S&T then how about -1 gy hate, -1 artifact hate, +2 Confusion in the Ranks. 3 Artifact hate from the board on top of your maindeck Tuktuk should still be sufficient for most matchups.

Good luck!

Thanks man, I'll definitely split the GY hate between RiP and RoP, 2:2. Seems like a good idea. Cantriping is always good. S&T will be present for sure, but so will be many blade decks (meta here is blade decks, canadian and S&T mostly), so I'll have to find some balance between blue hate, artifact and grave hate. Gonna be difficult, huh.


Currently I'm thinking about this sideboard.

Meta: Canadian, Jund, Omnitell, BURG, Merfolk, Sneak Show, Blade decks, Delver decks (with TNN now I guess). Show and Tell/Sneak Attack decks are significantly present.


3x Wear // Tear
2x Rest in Peace
2x Relic of Progenitus
3x Pyrokinesis
4x Red Elemental Blast
1x Chalice of the Void or Confusion in the Ranks


I decided to cut out the Chalices. They help against controls, canadian etc, but I want to be prepared for unfair decks instead. I guess I can beat control even w/o Chalice.

...amazing article-analysis...

Thank you very much for your effort. This made me thinking about my Thalia list a lot and I'll definitely make some changes in my list before my next tourney. I got to the point where you experiment and change your list so much, you get this feeling you've completely fucked it up. Your article will help me in "restarting" my list. Thanks again and merry christmas!

[GoboLord]

Just out of curiosity do you have this database compiled in Excel or some other program? Secondly could you make it accessible to the public? I realize we could each go and pick out each report but why re-invent the wheel here. I'm hoping to have a look at it myself and see if I can draw my own conclusions. Crowd sourcing could help out here is the thought.

For instance. Multicolored lists that played 3 Gempalms peformed better than lists that played 4, but no information is displayed about if they ran any other removal. Instead of querying you, we could query the database.

I have compiled the database in Excel and SPSS. However I don't want to make it "public" in the sense that everyone can download the file whenever he pleases. Rather I offer you to PM me your email adres and I will send you the file via email.

Besides, I have no problem at all with making analyses to answer your questions one-by-one.

@ LeoCop
I noticed that (Thalia + 4 Gempalm thing) and found it odd too. I think, as you said, Tarfire is the next-best replacement for Gempalm Incinerator in Rw lists. Then again, Thalia DOES reduce your Goblin count, but on the other hand it make Tarfire worse too, as it increases Tarfire's manacosts.

About the number of lands:

It appears that Mono R lists perform clearly better with 22 lands than with 23. For the multicolored and Thalia lists it does not seem to matter much for the decklists performance if you are playing with 22 or 23 lands.
The LAND #23 is likely to be Rishadan Port #4 indeed.

Of all lists with 22 lands...
* ...61 play with 3 Ports
* ...86 play with 4 Ports

Of all lists with 23 lands...
* ...36 play with 3 Ports
*...159 play with 4 Ports

[magicmerl]

@ LeoCop
I noticed that (Thalia + 4 Gempalm thing) and found it odd too. I think, as you said, Tarfire is the next-best replacement for Gempalm Incinerator in Rw lists. Then again, Thalia DOES reduce your Goblin count, but on the other hand it make Tarfire worse too, as it increases Tarfire's manacosts.

Speaking as a Thalia player, I think that it's also because Thalia builds are playing for the long game more than other builds are, and the card advantage is more useful.

One other point is that the 0-1 Karakas and 4-3 Port switch I don't think are separate things. You only have so much room for nonR lands, and the pressure to include a Karakas could be the cause of dropping the 4th port.

[woodjt5]

I played Legacy for the first time since GP DC last night. There was a local "Win-a-Force" event with 12 players, which I won . Unfortunately, I can't splash my new Force in goblins. Here's the list I played (the same 75 I took to DC):

4 Goblin Lackey
3 Goblin Piledriver
2 Mogg War Marshal
2 Stingscourger
4 Goblin Matron
4 Goblin Warchief
3 Goblin Chieftan
4 Goblin Ringleader
2 Krenko, Mobb Boss
1 Tuktuk Scrapper

3 Gempalm Incinerator
2 Tarfire

4 Aether Vial

12 Mountains
4 Cavern of souls
4 Wasteland
2 Rishadan Port

Sideboard:

4 Ashen Rider
2 Pyrokinesis
2 Tormod's Crypt
1 Grafdigger's Cage
1 Relic of Progenitus
1 Chalice of the Void
1 Pithing Needle
1 Goblin Sharpshooter
1 Legion Loyalist
1 Tuktuk Scrapper

I'm not going to post a full report, but here are my results:

2 matches for fun against Esper Stoneblade before the start of the event: (Win 2-1, Win 2-1)
Round 1: Win 2-1 over Dredge
Round 2: Win 2-1 over Miracles
Round 3: ID into top 4 with Esper Stoenblade (guy I was playtesting with beforehand)
Top 4: Win 2-1 over Miracles (same guy)
Finals: Win 2-0 over Esper Stoneblade (different guy)

There were a lot of blade decks so I kept the scrapper in the main. He was good. Chalice was excellent against Miracles. I guess Piledriver also slightly over-performed. He was good all night, and I never wished he was MWM. Other than that, everything felt like it should and nothing really under-performed. I continue to enjoy my 2-Krenko, 0-SGC, 7-Haste enablers build with only 2 ports because of the increased red mana requirement.

Happy Holidays guys!

[cronos]

Hi,

I've changed my sideboard to improve against combo/control, so I included three copies of Thorn of Amethyst. This is my list:

//Lands
10 Mountain
4 Cavern of Souls
4 Wasteland
4 Rishadan Port

//Artifacts
4 Æther Vial

//Instants
2 Tarfire

//Creatures
4 Goblin Lackey
4 Goblin Piledriver
4 Goblin Matron
4 Goblin Warchief
4 Goblin Ringleader
3 Mogg War Marshal
3 Gempalm Incinerator
1 Skirk Prospector
1 Stingscourger
1 Goblin Sharpshooter
1 Goblin Chieftain
1 Krenko, Mob Boss
1 Siege-Gang Commander

//Sideboard
SB: 3 Chalice of the Void
SB: 3 Relic of Progenitus
SB: 3 Thorn of Amethyst
SB: 3 Pyrokinesis
SB: 2 Tuktuk Scrapper
SB: 1 Boartusk Liege

Since yesterday, I've won 2-0 against BUG Control and 2-1 versus Canadian Threshold. I'll update the results in two - three weeks. Regards.

P.D.: I'm thinking about to play with one copy of Goblin Tinkerer and another of Tuktuk Scrapper, instead of two copies of the second one. It could be fine. What do you think?

[Chatto]

@ woodjt5: congratulations with your result!

Today I went to a GPT Paris here in my hometown. Didn't make extended notes, but my result was 3-1-1, losing in the top 8. Played against the following:

2-1 Imperial Painter (SB: +3 Wear//Tear, +2 Knesis)

First game and third game i raced my opponent, second game he does his thing.

0-2 TinFins (SB: +3 Thalia, +1 Relic, +2 RiP)

First game i am able to do some damage, but lack pressure. He goes off and kills me with Tendrills. Second game took a while. In his first turn he grabs almost his entire deck with Grisslebrand, except for six cards. He then Probes me and get his Tendrills... Sweet rips for him, not so much for me...

2-1 Tezzerator (SB: +3 Thalia, +3 Wear//Tear. Also Knesis, but that was just a mistake)

First game he makes a bunch of 5/5 artifact creatures, using the deck's namesake: Tezzeret, Agent of Bolas. He holds me of, while smashing my head in. Second game i waste his Ancient Tomb, after he doesn't get rid of my Lackey, which gives me Warchief. After that i land MWM, Piley, Matron into Piley, followed up with some other dudes and that was that. Third game he lands Plague, but Chieftain goes unanswered, followed by a Piley. He then lands a Ensnaring Bridge. I resolve a Ringleader, which i am able to blow up with a Scrapper. After that i am able to end the game quiet quickly.

2-0 Esper Stoneblade (SB: +3 Thalia, +3 Wear//Tear)

First game i get Krenko online with a hastelord, which is fast enough to race his Jitte. Second game I kill some dudes and a Batterskull, he whipes my board. He then drops two True-Named Nemesis, while I have Warchief and Incinerator (yes, i didn't cycle the guy). He then does something awkward: we are racing each other and i am about to lose if only he holds back on TNN. He manages to get me to one life, and i return the favor. He then whipes the board again with EE, saving my ass. I drop Matron, getting myself a Tarfire... I apologize to him for burning him to death, but hey...

i.d. into Top 8 (opponent played Elves, with NO-package, without Cradle. Played one match and I got my ass whooped turn three...)

1-2 Mud (with an red splash. SB: +3 Wear//Tear, +3 Thalia)

First game he mulls to five, easy win. Second game I am unable to kill his Goblin Welders. He then does tricks with Wurmcoil Engine and i die. Third game i have to mull to four and never recover.

Still won ten Theros-boosters, which was nice!

My list:

1 Siege Gang Commander
1 Krenko, Mob Boss
3 Goblin Piledriver
4 Goblin Warchief
4 Goblin Ringleader
4 Goblin Lackey
4 Goblin Matron
4 Aether Vial
1 Goblin Chieftain
1 Tuktuk Scrapper
1 Skirk Prospector
1 Goblin Sharpshooter
2 Tarfire
3 Mogg War Marshal
3 Gempalm Incinerator

1 Karakas
2 Arid Mesa
2 Scalding Tarn
2 Bloodstained Mire
2 Plateau
3 Mountain
3 Rishadan Port
4 Wasteland
4 Cavern of Souls

Sideboard (15)

1 Stingscourger
1 Relic of Progenitus
2 Rest in Peace
2 Pyrokinesis
3 Wear // Tear
3 Chalice of the Void
3 Thalia, Guardian of Thraben

Couple of notes:

- Won one game with Krenko, but it felt slow. I'm thinking of replacing it with an second SGC. I think SGC brings a direct threat, something Krenko lacks without hastelords,
- How to beat MUD? Any suggestions?
- Speaking of artifacts: Wear//Tear is awesome in general, I was thinking of using 4,
- TinFins: an autoloss or does someone have any suggestions?
- I'm thinking about replacing CotV. How about Ethersworn Canonist: keeping storm in place while having a beater as well,
- The lone Stingscourger in the SB will be replaced: didn't used the whole day long. Maybe this is because I didn't see any SnT, but even then i found it too slow. Perhaps replace it with the fourth W//T?

Any suggestions are welcome. Thanks for reading, cheers.

[cronos]

- Won one game with Krenko, but it felt slow. I'm thinking of replacing it with an second SGC. I think SGC brings a direct threat, something Krenko lacks without hastelords,

Sometimes, Krenko is quite better than SGC... If I were you I wouldn't replace it.

- How to beat MUD? Any suggestions?

Shattering Spree is probably the best card against MUD. By the way, the copy is not countered by CoTV of 1 because the copy is not cast (this is a common doubt).

- TinFins: an autoloss or does someone have any suggestions?

Mana denial + Thalia, Guardian of Thraben + Chalice of the Void + Relic of Progenitus / Rest in Peace should be enough to fight him. TinFins is a special Storm Combo deck... you can find this in my testing results, maybe it may be helpful:

vs. TNT: 2-1/2-1/1-2
vs. TES: 2-1/1-2/1-2
vs. DDFT: 2-0

It's a complicated pairing to win, but it isn't an autoloss. Regards.

[jrw1985]

https://www.youtube.com/watch?v=Nb8x9OWoDe8

[cronos]

Nice game ;). Cavern of Souls + Goblin Lackey + Krenko/SGC is "GG" usually.

[LeoCop 90]

Krenko gets better if you play more than 4 hastelords. Without haste, he grants inevitability unlike siege-gang-commander, but the risk of him being killed without doing anything is too high.

[Chatto]

Sometimes, Krenko is quite better than SGC... If I were you I wouldn't replace it.

I have played this split for a while, went back to 2 SGC, but now... I don't know... Thanks for the insight for those match-ups. I think i'm going to change my sideboard to the following:

1 Stingscourger
1 Tuktuk Scrapper
1 Relic of Progenitus
2 Rest in Peace
2 Pyrokinesis
2 Chalice of the Void
3 Wear // Tear
3 Thalia, Guardian of Thraben

Still like the three W//T, not sure I want to change one into Shattering Spree, going to think about it :-)

Krenko gets better if you play more than 4 hastelords. Without haste, he grants inevitability unlike siege-gang-commander, but the risk of him being killed without doing anything is too high.

This. Maybe playing Krenko means bringing the hastelord-count up to 6+... Right now I'm satisfied with my deck, maybe if I'm going hyper agressive Mono Red?

@ jrw1985: nice vid :-)) Krenko is still a bomb :-)

[apple713]

Nice game ;). Cavern of Souls + Goblin Lackey + Krenko/SGC is "GG" usually.

unless its G2, then pyroclasm hits, and then its GG for the goblin player...

[LeoCop 90]

Then you drop show and tell, we drop confusion in the ranks, steal your emrakul and roll over you.

[apple713]

Then you drop show and tell, we drop confusion in the ranks, steal your emrakul and roll over you.

i didn't say anything about S&T but i know there are other decks that play pyroclasm.

hell a ghostly prison stops the assault. Just don't overextend

[LeoCop 90]

Well, i can read that you play show and tell in legacy. My answer was rude just because i felt your comment was a bit arrogant... usually i don't feel the need to reply in threads of deck i don't play just to show how my x card beats them.

But, if your comment was meant to be constructive, then i apologise. Anyway, dropping krenko with lackey is not overextending... there is no reason not do it. And krenko survives to pyroclasm, while siege-gang dies.

[apple713]

Well, i can read that you play show and tell in legacy. My answer was rude just because i felt your comment was a bit arrogant... usually i don't feel the need to reply in threads of deck i don't play just to show how my x card beats them.

But, if your comment was meant to be constructive, then i apologise. Anyway, dropping krenko with lackey is not overextending... there is no reason not do it. And krenko survives to pyroclasm, while siege-gang dies.

great! it sounds like Krenko might be preferred then since pyroclasm is such a big problem.

Side note * I've been playing goblins since before vial was printed. Its my favorite tribe.