Printable View
http://www.starcitygames.com/magic/m...me_Report.html
Editor's Blurb:
Enjoy!Quote:
Monday, November 16th - In today’s double header edition of So Many Insane Plays, Stephen Menendian examines the current Legacy metagame before turning his eye to the Vintage scene. With graphs, stats, and theories aplenty, Stephen brings us the Eternal lowdown as only he can!
In b4 complaints about Premium.
The Dredge data is really interesting. Thanks for the numbers; this seems to largely confirm my suspicions about the format, Dredge aside.
EDIT: I think the single biggest reason Countertop has caught up to Merfolk in that matchup is the advent of the Japanese Supreme Blue list. That and Progenitus are probably the two best versions right now.
EDIT 2: I think this data is a big argument in favor of splashing white in Merfolk for Absolute Law, Path/Swords, and maybe even Sygg. "Vial or lose" seems like a bad place to be in.
I'm thinking hard about going Premium now.
To hell with you! =P
"God, Premium blah blah..." :wink:Quote:
Originally Posted by Aggro_zombies
I found the Zoo vs. Goblins matchup interesting. 6-0/6-1? That's pretty ugly. Did these lists splash for black or green (or both?). Instigator in there? Any non-traditional cards in the board(s) for those Goblin decks?
Fantastic article. Thanks for putting in the work!
I look forward to more analysis like this. Its great to see actual match up statistics.
Is there a way to buy single articles? Or do I have to pay for all the standard and limited stuff I'll never read?
Articles like this one I'd definitely buy, but it's not worth a full subscription.
Really solid, really unprecedented in article-writing. I think you must have hand-compiled this; if you can get DCI Reporter files from the event, you might be able to make a program that can mine the data in it and do this automatically after you've assigned a deck value to each person. Or, more realistically, you could get someone else to do it : )
No, you have to buy a subscription.Quote:
Originally Posted by caiomarcos
I don't really care about Standard, but playing Limited is a great way to improve your game regardless of what constructed format you play, and discussion about it is definitely helpful. I've become a much better player since I started drafting regularly (even though I still suck at drafting).
So here's a question: assume that, as the article states and defends, Legacy is highly matchup-dependent. What does this imply for how we build decks?
Consider these proposals:
A. Sideboards should be very focused on using hate to remove your worst matchups. You can't afford to make your fifteen from 3 cards against each of 5 decks. Players should pack 6+ cards for their worst two matchups, and maybe some general utility in the remainder. 43land should play lots and lots of combo hate, and fight "fair" against CounterTop.
B. Players should ignore their bad matchups as unwinnable by any sideboarding means. Construct your sideboard to gain an edge in your winnable matchups: 43land should just assume it's never going to face Belcher, and pack its sideboard with tools to fight CounterTop decks.
1. Which of the above is true, if either?
2. Are these statements necessarily contradictory? (perhaps A is true for the best players, who can win their not-terrible-matchups through skill, and B is true for mediocre players)
3. Does the choice of strategy A vs. strategy B differ based on decktype? If so, how do you make that choice for each of the format's best decks?
Cue the long discussion on how 'hate' is bad term in MtG theory.Quote:
Originally Posted by MattH
Please don't. :cry:Quote:
Originally Posted by TorpidNinja
i really like the excel chart, wish there was more data to it, hopefully you can keep doing these analyses.
Depends, when you're facing extremes, like Lands vs Belcher, then you're a huge dog game 1 and even with hate you're a coin flip game 2 and possibly unfavorable game 3. The match ups where you want to commit "hate" are the match ups that are closer to 35/65 or 40/60 like Merfolk/w for 4 Swords to Plowshares and 4 Path to Exiles vs Zoo etc. or when "hate" radically turns the match up around with Tormod's Crypt vs Ichorid etc.Quote:
Originally Posted by MattH
It's also a question of whether or not you're utilizing your SB, decks like Tribal, Dredge and Storm can afford to dedicate their SB to one, two match ups because the synergy of the cards prevents extensive SBing.
It seems like the traditional answer to this has always been that it depends on the degree of how bad the bad matchups are. In some cases there's hope for winning a certain bad matchup post-board, in which case it's correct to build your board to do so in most cases. However, the corollary to this is that it also depends on how bad the deck's other weak matchups are and how much space you'd have to take away from those to address the truly horrendous matchups. At some point you run into diminishing returns on your focused matchup and the risk of losing the close matchups starts to outweigh that benefit. In scenarios like that (where one matchup is so awful it eats up almost all of your board space to fix it), it's probably best to just give up and focus on improving your game against a wider selection of decks.Quote:
Originally Posted by MattH
The other trick is anticipating the hate you'll face from the other guy's board and figuring out what that would do to your percentages. If it doesn't depress your win rate significantly, that matchup effectively needs no sideboard space. If it could make things closer, you'd have to add in general counter-hate cards. Some decks are vulnerable to hate cards that are widely accessible (like Ichorid against Relic, Crypt, or the new Trap) and can use a small number of counter-hate slots to answer a lot of post-board matchups.
I think in the end that it's impossible to make blanket statements like that in a format as deck-driven as Legacy. Some decks will only have a few matchups that actually need dedicated sideboard slots, so they can go overboard there. Others will have one or two awful matchups but a lot more close ones, and at that point it's probably better to go with the board plan that boosts your return against the widest swath of the field you can.
Does anybody know if Wizards/DCI makes available tournament reporter files (including decklists and match win/loss)?
You know, I don't think I've said it yet, so I'll say it now: Your articles are the only reason I even have premium (got it for the Vintage back then, but definitely not complaining about Legacy content). This one is a perfect example of why. Just a good bunch of data I wouldn't normally have access to, compiled from tournament results. Thanks for that. Starcitygames should thank you for at least my money.
Quote:
Originally Posted by sauce
I will, no worries. It's alot of work, but it's really rewarding, and I can see that alot of you appreciated the effort.
Anusien offered with a computer program, but, frankly, doing it by hand probably isn't much more complicated. I developed, through repetition, a pretty fast algorithm, and I can watch TV or something while I'm doing it.Quote:
Originally Posted by hi-val
It seems Countertop Goyf is still the best deck in the format upon looking at this analysis.
What makes you say that?Quote:
Originally Posted by FieryBalrog
I'd say it completely depends on your knowledge of the metagames, and how much a sideboard really helps in a matchup. If you know there will be a high percentage of combo decks and you plan on running 43-land no matter what, then you better at least try to pack a good portion of your sideboard to hate it. If you know there are 2 players or less out of 20 that run combo then you might as well just leave out sideboard all together against them and accept the loss. Similar situation for goblins.Quote:
Originally Posted by MattH
On the other hand, if you know that you will be facing a deck where hate is actually quite potent against them (I like to bring up Extirpate against Ichorid decks, even though I'm sure advanced ichorid players have a better shot at getting around it), then I would be perfectly happy with only running 4 sideboard cards against that matchup. The sideboard cards that have uses in almost every matchup (Krosan Grip for example), are definitely worth having, even if it doesn't bring your matchup against Enchantress to a 50-50. If you're scared of Elves or NO Progenitus, having Perish in the sideboard is going to help against random other matchups, like zoo.
So overall it depends on: How many of an archetype do you think will be there, How potent is a sideboard 4 or even more slots against them, and Does the sideboard you have against that deck help against other matchups?
I think it's a good idea to have some cards like K-grip or Extirpate that completely destroy certain decks, but can also be very good against a wide range of other decks.
Very cool article. More of this please.
This is a great Magic theory question, but I'm not sure the answers are so readily available. I think "know your meta" is a little difficult sometimes, particularly at larger events. I agree with whoever said there are degrees of "bad" when discussing bad matchups. I think one would have to determine the probability of needing a particular card and drawing it. I wouldn't fancy my chances of drawing an opening hand Leyline of the Void, for example, and wouldn't board that.Quote:
Originally Posted by MattH
Also, I think SB cards should always aid and not impede one's deck's innate formula/game plan (e.g. don't dilute aggro's speed). I know that seems like "duh", but I do believe that SBing (both choices pre-tournament and in actual matches) really is a tricky topic with so many variables and it remains one of the hardest and least understood/practiced parts of the game.
Anyway, back to the original question, I suppose if you were to use strategy A (SB for bad matchups) the key question that follows is how much do the SBed cards increase your win percentage? The same thing goes for B: how much do the cards meant to give you an advantage over 50/50 matchups really do give you that advantage? From those answers you should figure out your SB by using the cards that most greatly improve your win percentages.
How about neither or both? I think you should never just use a sb to shore up a bad matchup. Whether you do so or not has to be filtered through the first question of what you expect to show up?Quote:
Originally Posted by MattH
Suppose you are going to play in a 100 player tournament and you expect, justifiably for whatever reason, that at most 1 player out of the 100 will be playing your worst matchup, that requires like 6 sb slots. Is it worth it? Answer: no.
I always try to first figure out what I expect to face, and then fill out the SB accordingly.
Very nice article, it actually made me buy premium again... so good job to you Steve.
I'm looking forward to see how this analysis will change in the near future...
Agreed. In particular, I think the Japanese Supreme Blue list has a better Merfolk match-up than what Steve put in the article (I understand all CB-Goyf decks were lumped together for that analysis). Having 3 Firespouts MD and REBs post-side make this match-up hard for Merfolk IMO. In my tests/tournament experience, I haven't lost once to Merfolk with Enigma's list of Supreme Blue... I hope the future continues to be as bright.Quote:
Originally Posted by Aggro_zombies
Cough... cough...Quote:
I’ve measured Top 16 penetration
Awesome :) My SCG overlords will be pleased.
I will definitely be doing another matchup grid for the next SCG $5k, so your subscription will definitely be worth it.
you might google the similar analysis Paul Jordan has done for PTs if you are looking for comparison.
This was the final straw- I've bought premium now and reading all of the articles I've missed out on from the last month.
To jump on the answer bandwagon, I would say B always. To begin with, your deck choice should be appropriate for your local metagame- don't take Zoo to Friday Night Nauseam, or you deserve what you get. It doesn't make sense to me to start behind against more than half the field at any given local tournament and then try to compensate for an effective 0-1 deficit going into most of your matches. For a national event, the field is much wider, but as these articles have shown us, there is a definite metagame taking shape (go Legacy!).Quote:
Originally Posted by MattH
In either case, I think it's vitally important to make your 40-60% range matchups as strong as possible and just ignore the shitty ones. Let's say you have a roughly equal distribution of high (~60+%) matchups, medium (~40-60%) matchups and low (~<40%) matchups for any given deck. If you waste your sideboard trying to bring the low % matchups up, you needlessly risk losing your on-the-fence matches (say, Canadian Thresh vs CT-goyf) and still make no guarantees that you'll be winning your otherwise bad matchups (since G1 you're still probably going to lose and then your opponent gets 2 games where a bad hand, slight misplay or just the fact that your matchup still isn't great postboard can still screw you).
On the other hand, if you shore up your mediocre matches and just largely forget the bad ones, you now have a solid chance of winning against close to 2/3% of the field. Since with Swiss rounds the good known decks are more likely to rise to the top (i.e. the ones you probably sided for), you have a much better chance of making the finals even if you get unlucky and hit a bad matchup early on.
I think this is a large part of why Tempo Thresh is so successful. The decklist is largely unfocused misrange jank, but doesn't punt to anyone and has winnable matchups against nearly everyone. The sideboard shores up a lot of those, so postboard you're turning could-win into probably-will-win against a lot of the decks you're likely to see. Shoring up your bad matches tries to do too much against too wide a range of opponents and ends up screwing you over either way. Zoo is another quintessential example. All of the good players (such as the Hatfields) design their boards to make damn sure they're not losing to TThresh or GoyfSligh so that when they do come up against ANT, they can still recover and do well.
And now to the original reason why I wanted to reply-
I disagree with the author that metagame positioning is as big a factor as the article claims (even though I just spent several paragraphs defending it...). In fact, Cedric's path to the 8 suggests to me the exact opposite of what the following paragraph says: yes, Belcher is going to curbstomp RG Beats and slower obsolete combo decks, but he also faced some seriously tough matches and beat them. I see 3 matchups which are decidedly against him, (rounds 4, 7 & 8), two of which he won and one which was an ID. Even the unfavorable Stax match went to G3. Belcher's fairly straightforward, but it can easily lost G2 and 3 to a control deck with a full counterspell suite. It's not that tough to mull into force or daze or something that will stop the combo.
Yes, matchups are a big factor, but I think the conclusion the article makes is a fallacy. Player skill, I think, decides outcomes for decision-heavy decks (pure combo and control) almost entirely, let alone makes a dent. The grid is misleading because big events like the SCG5k are going to have a lot of opportunistic dilettantes who netdecked their (insert DTB here) list and skewed the list. If player skill wasn't a factor, really good players like the Hatfields, Chris Woltereck, David Gearheart, etc. would not consistently make the finals in most of the tournaments they enter. In fact, if player skill was so irrelevant, we would probably have entirely different names almost every time in large events since anyone can just rip a Tier 1 deck off the internet (which we dont; the same players do consistently make the T8/16, which suggests they're riding tight play rather than just a good metagame call).
Case in point, Woltereck's 43-land deck took 2nd in SCG Charlotte and (7th?) in SCG Philly. Their metagames were diametrically opposite- Charlotte put 3 Zoo decks in the T8, which is more than there were combo decks at the entire tournament. In Philly, combo was one of the most highly represented decks in the field. 43-land stomps most aggro and basically folds to combo. I don't know exactly what his path to the 8 was and he's definitely not the only example, but my point is that player skill and not matchup percentages is most often the deciding factor in how a player does. It would be wrong to take a single bad performance by a good player and use it as proof to demonstrate how much the metagame means because of how consistently good players actually do rise to the top.
Maybe I should be writing articles of my own if I want a soapbox?
This article is also free!
Ripped off, I was *fist in air* !