Too Much Information- Worcester + Indy

[Tammit67]

http://www.starcitygames.com/magic/legacy/24388-Too-Much-Information-Legacy-In-Worcester-and-Indianapolis.html

I am horrified to see sneak and show do so poorly against the field. It seems those that flocked to the deck can't play it very well. Edit: In fact, combo players seem to be non existent and terrible at SCG events, with the exceptions of Joey Manner/Ari Lax.

As always, thanks to Jesse and Alix Hatfield!

[DragoFireheart]

How is someone terrible at Sneak and Show?

[Mirrislegend]

How is someone terrible at Sneak and Show?

I'm not sure. However, I have witnessed the difference between an average/not terrible Sneak and Show player vs a good/experienced/excellent Sneak and Show player. The difference is fairly significant, like many decks.

Also, I had a good chuckle cuz I crushed the quoted RUG list in the last round. It was pathetic :)

[rxavage]

How is someone terrible at Sneak and Show?

I was wondering how he inferred this from that article as well. While not being very difficult to play SneakShow isn't as straight forward as playing burn, especially when it no longer has the surprise factor it did a month ago.

[Einherjer]

How can BUG have a bad Maverick-MU? The MUs is one of the best, normally. What did they do?!:eek::eek::eek:

[CorpT]

How can BUG have a bad Maverick-MU? The MUs is one of the best, normally. What did they do?!:eek::eek::eek:

I think there are too many variations on BUG to lump them all together. It's not a well defined list, so there are a lot of ways to make it wrong or bad against Maverick.

What I find interesting is the very good Stoneblade results that came on the heels of Levin talking about how bad Stoneblade was for weeks. I played Stoneblade at the Invitational and did fairly well in Legacy: 5-2-1.

[Piceli89]

I find these type of articles based exclusively on statistical data collection to be pretty inaccurate and non-representative of the actual state of Legacy (matchups and dominance), since they're based on tournaments results which imply a lot of poor pilots and/or luck- this is especially true for SCG opens, where I see lots of misplays and punts on camera every single time.

[Tammit67]

I was wondering how he inferred this from that article as well. While not being very difficult to play SneakShow isn't as straight forward as playing burn, especially when it no longer has the surprise factor it did a month ago.

Sneak and Show is a huge favorite preboard to RUG. All sneak has to do is wait until pierce is a non factor, and sneak will often have 2 pieces of protection by then.

The deck runs 5+ sol lands, has basics and a ton of counters. It is perfectly set up to beat soft disruption so long as you don't run headlong into everything. It should not be going 50% to RUG. It should NOT be going 46% historically. Therefore, people are not sequencing lands/spells correctly as a whole.

This isn't based on personal experience, although my testing with RUG has only made me sad about this matchup. This is mainly from Joey Manner/James higgenbottom who have been playing the deck non-stop for this reason to a ton of success at SCG invitational/Indy Opens and Jupiter Invitational/Qualifiers

[wcm8]

How can BUG have a bad Maverick-MU? The MUs is one of the best, normally. What did they do?!:eek::eek::eek:

BUG Thresh/Tempo: just a worse version of RUG tempo
Team America: seems well positioned, but not very popular right now, and can potentially struggle vs. Maverick
BUG Control (4 Snap/4 Goyf): good deck, decently positioned, should not struggle against Maverick as long as the SB isn't complete garbage
Full-On BUG Control (2-3 Snap, no other creatures): poorly positioned in my opinion, too many answers and not enough threats, seems like it loses to Maverick on occasion due to the one threat slipping through

Lumping these decks altogether seems short-sighted, because even though there is a fair amount of overlap in the card choices, each iteration has different strategic goals, weaknesses and strengths. Saying 'BUG did poorly' doesn't really say much of value.

[catmint]

The Data shows that all the "Griselbrand against the format" nonsense of SCG writers and the whining in the community was just based on a hype (happened last time with hive-mind).

As I wrote before often enough: Sneak/Show has a high power but low consistency cause you have to mull a lot and can draw a bunch of dudes and altough the decision tree for the deck is not as complex as in some other decks it makes of course a big difference if you are a good pilot or just someone getting a deck together to play in a big tournament.

[rooneg]

I'm kind of curious what sort of munging the data had before it ended up in that spreadsheet. The numbers on the Worcester chart add up to 282 players. There were actually 308 legacy players at that tournament, and I know of several archetypes that were present that didn't make their list (I played Dream Halls, as did at least one other guy and that's not listed, and I played against an Armageddon Stax player and there's nothing like that on their list, plus you can't forget the one guy playing Battle of Wits!). It's one thing if you say "top 128 players" or something, but when you've fit almost the entire field and the lines at the bottom of the chart are for Hive Mind and U/B delver at 1 player each it sort of implies a thoroughness that doesn't seem to actually be present.

[Chikenbok]

BUG Thresh/Tempo: just a worse version of RUG tempo
Team America: seems well positioned, but not very popular right now, and can potentially struggle vs. Maverick
BUG Control (4 Snap/4 Goyf): good deck, decently positioned, should not struggle against Maverick as long as the SB isn't complete garbage
Full-On BUG Control (2-3 Snap, no other creatures): poorly positioned in my opinion, too many answers and not enough threats, seems like it loses to Maverick on occasion due to the one threat slipping through

Lumping these decks altogether seems short-sighted, because even though there is a fair amount of overlap in the card choices, each iteration has different strategic goals, weaknesses and strengths. Saying 'BUG did poorly' doesn't really say much of value.

Playing snapcaster was the problem in the first place (need more cliques)

I still don't understand how these BUG pilots did so poorly against maverick with loam/deed/perish/virtues ruin/etc. -- Seems like bad pilots are bad pilots.

[Mad Zur]

BUG Thresh/Tempo: just a worse version of RUG tempo
Team America: seems well positioned, but not very popular right now, and can potentially struggle vs. Maverick
BUG Control (4 Snap/4 Goyf): good deck, decently positioned, should not struggle against Maverick as long as the SB isn't complete garbage
Full-On BUG Control (2-3 Snap, no other creatures): poorly positioned in my opinion, too many answers and not enough threats, seems like it loses to Maverick on occasion due to the one threat slipping through

Lumping these decks altogether seems short-sighted, because even though there is a fair amount of overlap in the card choices, each iteration has different strategic goals, weaknesses and strengths. Saying 'BUG did poorly' doesn't really say much of value.
Unfortunately, I don't actually see the decklists, and I'm not the one who does the classification, so I can't definitively say what is and isn't included under "BUG." But in the context of what people have been playing lately, and given the fact that those doing the classification didn't feel the need to differentiate between multiple builds, I think it's safe to assume that most if not all of the BUG in these tournaments was BUG Control, not Team America or anything with Tarmogoyfs. The only "BUG" to make it into SCG's deck database was clearly BUG Control, and that's probably a reasonable example of what people were doing. We played Team America at the GP -- if it looked like that was the kind of BUG that was doing poorly in the Opens, we wouldn't have.

I'm kind of curious what sort of munging the data had before it ended up in that spreadsheet. The numbers on the Worcester chart add up to 282 players. There were actually 308 legacy players at that tournament, and I know of several archetypes that were present that didn't make their list (I played Dream Halls, as did at least one other guy and that's not listed, and I played against an Armageddon Stax player and there's nothing like that on their list, plus you can't forget the one guy playing Battle of Wits!). It's one thing if you say "top 128 players" or something, but when you've fit almost the entire field and the lines at the bottom of the chart are for Hive Mind and U/B delver at 1 player each it sort of implies a thoroughness that doesn't seem to actually be present.
Well, it looks like the link to the actual spreadsheet for Worcester is broken (it mistakenly links to the previous article). I'll see what I can do about that. The spreadsheet should include the full data. The tables in the article, as you've noticed, don't list all decks. Traditionally, they contained all decks that were at least 1% of the field. That would be three people in Worcester, so no Dream Halls (two), Stax (one), or Battle of Wits (one). At some point I started including decks that were in at least 1% of the field in one of the tournaments examined in the article, so that if a deck was in one table, it would be in the others (unless there were zero in a particular event). Based on several comments like this, I'm now of the opinion that this is confusing and suboptimal. I'd love to hear suggestions -- more decks, or fewer? The idea was that things can get too cluttered if we list all the one-ofs, and they don't provide much data anyway.

Thanks for the comments!

[rooneg]

At some point I started including decks that were in at least 1% of the field in one of the tournaments examined in the article, so that if a deck was in one table, it would be in the others (unless there were zero in a particular event). Based on several comments like this, I'm now of the opinion that this is confusing and suboptimal. I'd love to hear suggestions -- more decks, or fewer? The idea was that things can get too cluttered if we list all the one-ofs, and they don't provide much data anyway.

Thanks for the comments!

I don't so much care if you're leaning towards more decks or fewer (in the end it's not particularly relevant if there was one guy playing some fringe archetype), I just think you should state what goes into the chart more clearly. If it's only archetypes that made up 1% of the field then say that. If it's that plus anything that was on one of the other charts then say that. If it's absolutely everything that the classifier could deduce a meaningful archetype for then say that.

[FieryBalrog]

Unfortunately, I don't actually see the decklists, and I'm not the one who does the classification, so I can't definitively say what is and isn't included under "BUG." But in the context of what people have been playing lately, and given the fact that those doing the classification didn't feel the need to differentiate between multiple builds, I think it's safe to assume that most if not all of the BUG in these tournaments was BUG Control, not Team America or anything with Tarmogoyfs. The only "BUG" to make it into SCG's deck database was clearly BUG Control, and that's probably a reasonable example of what people were doing. We played Team America at the GP -- if it looked like that was the kind of BUG that was doing poorly in the Opens, we wouldn't have.

Well, it looks like the link to the actual spreadsheet for Worcester is broken (it mistakenly links to the previous article). I'll see what I can do about that. The spreadsheet should include the full data. The tables in the article, as you've noticed, don't list all decks. Traditionally, they contained all decks that were at least 1% of the field. That would be three people in Worcester, so no Dream Halls (two), Stax (one), or Battle of Wits (one). At some point I started including decks that were in at least 1% of the field in one of the tournaments examined in the article, so that if a deck was in one table, it would be in the others (unless there were zero in a particular event). Based on several comments like this, I'm now of the opinion that this is confusing and suboptimal. I'd love to hear suggestions -- more decks, or fewer? The idea was that things can get too cluttered if we list all the one-ofs, and they don't provide much data anyway.

Thanks for the comments!
If you're only cutting off a handful of decks at each tourney, why not just include them for completeness sake? Stopping at 95% completeness just feels weird.

[Koby]

If you're only cutting off a handful of decks at each tourney, why not just include them for completeness sake? Stopping at 95% completeness just feels weird.

The odd-ball and singleton decks lose meaning from a statistical point of view. The whole idea with this analysis is to show macro-view metagame trends.

[Gui]

I really like TMI articles, but I find it hard to make the connection between it and Dredge, with so low win %, beating tourneys every here and there.

I can only assume that the players are bad, way more than the deck itself.

[Julian23]

I really like TMI articles, but I find it hard to make the connection between it and Dredge, with so low win %, beating tourneys every here and there.

I can only assume that the players are bad, way more than the deck itself.

It's actually a pretty common theme among Dredge players to overestimate the deck.

[Koby]

I really like TMI articles, but I find it hard to make the connection between it and Dredge, with so low win %, beating tourneys every here and there.

I can only assume that the players are bad, way more than the deck itself.

Consider also this: Reanimator was being pegged as the big-bad Boogeyman around the time these tournaments were going on. This pushes people to play more than a normal amount of graveyard hate. This in turn, destroys Dredge's matchups.

Ergo, Dredge has been underperforming.

[Aggro_zombies]

I really like TMI articles, but I find it hard to make the connection between it and Dredge, with so low win %, beating tourneys every here and there.

I can only assume that the players are bad, way more than the deck itself.
Just to play devil's advocate here: can a deck really be "good" if no one can play it correctly? Decks like Dredge or Doomsday may be objectively powerful on paper, but if no one can get them across the finish line then they can't really be that broken, can they?

But yeah, I agree with Koby that there might have been more hate around. Most dredge players I know can handle game one but walk right into Crypts and Surgicals games two and three because they don't know how to plan around hate cards.

[Tammit67]

Just to play devil's advocate here: can a deck really be "good" if no one can play it correctly? Decks like Dredge or Doomsday may be objectively powerful on paper, but if no one can get them across the finish line then they can't really be that broken, can they?


Yes they can. There is a difference between perceived truth and objective truth. I for one believe in objective truth

The deck isn't at fault. The pilot has the tools but cannot use them properly.

Penicillin is still amazing even if I don't know what the hell to do with it.

[FieryBalrog]

Yes they can. There is a difference between perceived truth and objective truth. I for one believe in objective truth

The deck isn't at fault. The pilot has the tools but cannot use them properly.

Penicillin is still amazing even if I don't know what the hell to do with it.

The problem with this is you can apply this line of reasoning to many different underperforming decks.

[TheInfamousBearAssassin]

Just to play devil's advocate here: can a deck really be "good" if no one can play it correctly?

Obviously not. By definition, if no one can make the deck perform then it doesn't matter if the deck might be theoretically good.

The question of relevance is if someone can perform well with the deck.

If they can't then suggestions that someone might be able to are mere speculation.

[Amon Amarth]

The problem with this is you can apply this line of reasoning to many different underperforming decks.

I don't see that as a problem. Some decks just don't do as well because they are more difficult to pilot than other decks.

[Gui]

The problem with this is you can apply this line of reasoning to many different underperforming decks.

I guess you can apply this logic to a lot of currently underperforming decks, but this doesn't mean that it's a flawed logic. Maybe these decks are indeed underperforming due to players lack of skill.

Proof of this is that, despite dredge being at 30% winning rate on this TMI, it still made it into the DTB this month once again.

Dredge was once know to be an autopilot deck, but against hate you can see how hard it is to play it. Not every avarage joe can do it properly.

[joemauer]

Just to play devil's advocate here: can a deck really be "good" if no one can play it correctly? Decks like Dredge or Doomsday may be objectively powerful on paper, but if no one can get them across the finish line then they can't really be that broken, can they?


Well these decks have shown up high at big tourneys. Doomsday at the BOM, top four and top sixteen, and Dredge at the GP Indy, top four.

So the decks have proven themselves and recently too. They underperform for a few reasons.

Firstly, difficult decks to play. Kids see Dredge going off and think it would be awesome to do that not realizing how difficult it is to play. Everyone is well aware of how tough Doomsday is and not enough people give it a shot because it is too difficult.

Secondly, both decks can get hated out pretty easy. Griselbrand decks are bad for doomsday decks. Since Reanimator is popular again people are packing extra graveyard hate. Dredge decks can only fight so much hate games 2/3 with or without a competent player. Sometimes the opponent just topdecks the tormod's crypt at the wrong time.

Thirdly, these are combo decks. They both can poop on the pilot. This is important in big tourneys where one or two poopings can knock you out of top eight/sixteen.