February 12, 2018 Banned and Restricted Announcement


Yes, I am not arguing that every form of aggregation/classification hides some of the truth, and reveals other portions of it. I am with you 100% on that.

As a corollary, I'm also with you that classifying every Vintage deck into: "Blue Control", "Workshops", "Dredge", "Storm", and "Fish" is a less than ideal situation, but I feel it is necessary to make any meaningful use of data to inform my deck building. What I am asking is what better classification is there that would provide an amount of aggregation? (I'll add without biasing towards Blue diversity like has been done with the currently used taxonomic system (i.e. broad stroking everything except for blue control decks))

Do you think it would be meaningful from a restricted/strategic discussion to classify decks as Control/Aggro/Combo? Essentially lumping Shops/Fish into one, and Dredge/Storm into one? (I gather from your dislike of me using "Blue" as a sweeping class means that you wouldn't like me just renaming it "Control")

Do you really think it would be useful to classify each deck into schools, given that there are so many?

last edited by vaughnbros

Well, Ryan and Matt have been classifying Vintage decks since January of 2016 on this forum, and rather effectively to my mind.

In that time, they've covered every single Vintage Power Nine Challenge since January of 2016, and done the same analysis for NYSEs, Waterburys, Vintage Championships, and even Eternal Extravaganzas.

In that time, their classification system, which borrows from and expands on classification I conducted going back to the mid-2000s, has evolved. What was one called "Gush" decks is not simply called "Xerox," which is just a synonym in my view for Comer School decks.

So, your question isn't really a hypothetical in the sense that we have ample (and I mean AMPLE) examples of data organization to look at and see if you think they are effective.

I mean, take a look at judge for yourself:

Yes, Matt and Ryan previously broke out the other archetypes into different categories that become too small of a sample to really useful from a statistical standpoint, but much more informative as to what was happening.

Now, we are seeing Matt present data that doesn't have any breakdowns (except for Blue decks) in a restricted list discussion seemingly as an effort to hit cards in Workshops and push blame away from cards like Misstep.

@smmenen I feel like you're so close ... Surely different sets of classification are better for different purposes, and you've laid out a comprehensive argument for why your classification system helps people think about the format strategically.

But the reason why people are using different classifications here isn't because they disagree it's a good fit for that purpose, it's because they have different purposes.

If your least favorite thing about vintage is a game where you get locked out under spheres, you don't really care if there's a nuanced interplay between Ensnaring-Bridge-Null-Rod-Shops and Thought-Knot-Shops, you just want fewer decks with Sphere of Resistance. If your favorite thing about vintage is playing Dark Ritual, you don't care that Inferno Oath vs Griselbrand Oath is more of a paradigm shift than Esper Mentor vs Jeskai Mentor, you just want fewer decks with Mental Misstep. In fact it's not even internally inconsistent to think of Hollow Dredge and Depths Dredge as different archetypes while building your deck, and the same archetype while in this thread.

The sticking point here isn't that people's models of the format don't line up, it's that people are modelling entirely different things. Debating over who's map is wrong when people are in different territories is the very essence of concealing through revealing.

If I think the worst thing about vintage is that Combo Control isn't good enough and @nedleeds thinks the worst thing about vintage is that Combo Control is too good, that's irreconcilable. One of can accept it, or we can both lose, but we can't both win.

Designing for humans is about knowing your audience, and not everyone can be in that audience. Whenever WotC changes the B&R list, that audience changes. Perhaps that's why people are so passionate about B&R. People aren't arguing over what makes the format "better", people are arguing over their right to be a vintage player.

last edited by Brass Man

@brass-man said in February 12, 2018 Banned and Restricted Announcement:

@smmenen I feel like you're so close ... Surely different sets of classification are better for different purposes,

That's almost exactly what I already said a few posts up. I wrote:

"So the question is which classification scheme is best for which purposes."

And I've already stated what those purposes are: 1) understanding the format for B&R list management, and 2) understanding the metagame to identify niches to exploit or attack for deck design purposes, among other purposes.

I have no problem with people using different taxonomies. I was simply objecting to an approach which sweeps Xerox and Landstill strategies together because they are "blue" for any of these purposes.

@vaughnbros said in February 12, 2018 Banned and Restricted Announcement:

Yes, Matt and Ryan previously broke out the other archetypes into different categories that become too small of a sample to really useful from a statistical standpoint, but much more informative as to what was happening.

Now, we are seeing Matt present data that doesn't have any breakdowns (except for Blue decks) in a restricted list discussion seemingly as an effort to hit cards in Workshops and push blame away from cards like Misstep.

@vaughnbros said in February 12, 2018 Banned and Restricted Announcement:

Yes, Matt and Ryan previously broke out the other archetypes into different categories that become too small of a sample to really useful from a statistical standpoint, but much more informative as to what was happening.

Now, we are seeing Matt present data that doesn't have any breakdowns (except for Blue decks) in a restricted list discussion seemingly as an effort to hit cards in Workshops and push blame away from cards like Misstep.

They haven't been entirely consistent over time. But neither was I. Compare:




I take the layered approach of using multiple simultaneous categories. I organize by tactic, archetype, and engine, at various times.

last edited by Smmenen

I know that's what you wrote a few posts up, I was agreeing with you as a rhetorical device 😛

Yes, Brass you make a good point that everyone's goals are different in the sense of where we'd like to see Vintage as a format. Some of just want to play a bunch of blue mirrors all day that is what makes Vintage, Vintage to them. Some of us like the deck/strategy/color diversity that we see in other formats, and would like to see more of it here. Some of us want restrictions/unrestrictions to see a particular deck excel and/or others fail. In these senses, everyone comes into the discussion with a different goal in mind.

Where I will disagree is that everyone does have two goals in mind:

  1. They want to have fun playing Vintage.
  2. They want to see Vintage be successful.
    I have a hard time believing anyone in one of these discussions doesn't have each other these larger goals in mind during the discussion.

Most people in this particular discussion also seem to have a similar idea in mind for improving those goals:

  1. Making a change to the restricted list.
    Although, I don't think this is necessarily true of most discussions.

So no matter how anyone of us slicing and dicing the data up, or classifying each deck, or however they are coming their conclusions, we all have a common goal and a common path to said goal.

Going back to the data (I'll scrap the "blue" classification) over the last six months, I see the top 22% of the metagame is shops (mostly ravager?) and 16.5% of the meta is "Xerox" (previously Mentor and previous to that Gush) still leading the format in meta % AND match win % despite numerous restrictions, and ample time for the format to "adjust". Both decks being at the top for a quite a while (you can argue however long you like, but I believe its been at least a couple years of regular data showing them at top). I think we can all agree that this is the root cause of our frustration? We have seen numerous restrictions yet, none of them to date have worked.

So literally the only part we all seem to have disagreements on are:

  1. What action to take on the restricted list?
  2. How did we come to this conclusion?

What's funny to me is that two people can have totally different views on 2. yet still reach the same conclusion in 1. At the same time, people can have extremely similar views on 2. yet still reach a different conclusion on 1. Now does 2 really matter? I mean in the end, the only actual action being taken is 1. What I really have to wonder is why does it matter if we are reaching the same conclusion?

So the following conclusions were:
Restrict something in Workshops (no consensus on what, and most people can't even give a single card).
Restrict Mental Misstep.
Unrestrict Flash.
Unrestrict Fastbond.

Now, I would literally do all of these things. As I started this discussion, I'm really sick and tired of the format having the same decks at top for what seems to be quite a while (again I guess arguable as to how long; In my viewpoint its been 8 years, in others maybe only 6 months). So lets shake everything up. Do it all. Unrestrict everything that won't benefit the big 2. Restrict a bunch more cards in the big 2. At least we get a different, fresher format that people can start to enjoy again.

last edited by vaughnbros

@vaughnbros So I start out completely agreeing with you, but we end up differing pretty fast.

everyone does have two goals in mind:

  1. They want to have fun playing Vintage.
  2. They want to see Vintage be successful.

Completely on board. I assume everyone here is is acting in good faith, and if there are a few bad actors? well? fuck'em. Most vintage players want other people to play vintage.

but then the very next line is

Most people in this particular discussion also seem to have a similar idea in mind for improving those goals:

  1. Making a change to the restricted list.

and that's where I'm lost.

Both decks being at the top for a quite a while ... I think we can all agree that this is the root cause of our frustration?

I'm not sure you realize how fundamentally people differ on this. I like Mentor mirrors. I think they're just great Magic. I think the gameplay decisions are deep and the deckbuilding decisions fascinate me. When Monastery Mentor got restricted and the deck got less popular, the average amount of fun I had per round of Magic went down. When I enjoy a matchup and the incidence of that matchup gets reduced because of a restriction, my frustration doesn't come from one deck being dominant for too long, it comes from other players campaigning for that restriction. I like narrow metagames, I can understand narrow metagames. A good deckbuilder is rewarded for making nuanced choices in narrow metagames. In a broad metagame they're subject to the DCI Reporter RNG. When those changes happen, I get frustrated and the game gets less fun for me, but I try to avoid talking about my personal preferences because I'm a good little spike and I feel that my job as a player is to win with the cards that are legal, not to change which cards are legal until I win.

I feel like campaigning for format changes as TMD admin is unfair, and my goal is to maximize fun for vintage as a whole, not for myself.

But what is Vintage as a whole? I get that some people don't like the format right now and I get that some of those people are in this thread. MOST people don't like vintage. MOST people don't like Magic at all. MOST people would rather play FreeCell. TMD is a niche within a niche within a niche. It's hard for me to reconcile the "everyone knows there's a problem" attitude in this thread with the fact that Leagues keep firing, and EW keeps trending upwards year after year. It's not lost on me that TMD is not growing at the same rate. The brand new archetypes popping up every few weeks on magic online? The people building those decks aren't posting them here. Here we just talk about how there's no room in the format for new archetypes.

So who decides what counts as Vintage being successful? I'm a dinosaur who thinks that vintage will never be as good as it was when shops was just okay and you could play 4 Brainstorm. If I'm having more fun, is that successful? If the MTGO pro who just dropped $3k on a set of Workshops gets to protect their investment, is that successful? If the FNM player who tried vintage and quit because he was steamrolled by Spheres give the format another chance after a Workshop restriction, is that successful? Who's spending more money? Who's hosting more tournaments? Who's writing more articles? Who's bringing more players into the format? Who's more fun to play games with? Who counts more and who counts less? It's an ugly question but everyone has an answer whether or not they're trying to.

I think if you ask the question "What's best for vintage" and it always just happens to be the thing that makes the game more fun for you, you might be missing some perspective.

The most clear example for me personally, I believe the following two things:

  • For some Vintage players, the presence of Mental Misstep makes the average game of Vintage Magic better. It makes games more skill testing, makes it easier to metagame against blue decks, and creates interesting deckbuilding decisions.
  • For some Vintage players, Mental Misstep is a genuinely frustrating card. Some people just really don't like playing against it, and they feel that it constricts their choices, and that makes Vintage less fun for them.

Where does that leave me? Who counts more? The % of the metagame made up of Blue decks as compared to White Eldrazi decks doesn't remotely matter. People who think Misstep is good for the game think it's good no matter how many people are running it. People who think it's bad for the game think it's bad no matter how many people are running it. I don't have a great idea of how many people who play Vintage even have an opinion on this.

At least we get a different, fresher format that people can start to enjoy again

At the expense, of course, of anyone who enjoys it now.


It sounds like you are in the class of individual that I described that wants endless Blue mirrors. But you also seem to want no changes to the restricted list? You wouldnt say want Mentor back? Or Brainstorm back? Or see Workshop be restricted?

last edited by vaughnbros

Pro and high level players choose their decks according to the meta and what they feel would be the best strategy. In theory this is the best way to do but in the real world they are just a very small % of vintage players. Most of us are just playing one kind of deck (or a very small number of them) because it is the kind of deck that suits our playing style and we have fun with it (another reason is card availability/price of course but that is not the point here).

From that point, the (theorically) best metagame would be the one where each 'playing style' can be incarnated by at least one deck (several would be best of course). As an example, i am just very bad at playing combo decks : whatever the deck or the format and how hard i try ... i always loose. It is just not my style. On the other hand, i know players that seem to be always playing combo decks whatever the incarnation or format (storm, belcher, elves, ....). I don't know what would be an adequate list of 'playing styles' but i presume you understand what i mean here.

I won't come back on what has been said on the classification of decks from a meta management point of view, but what i just said can put a different light on that problem. Note that i am advocating nothing nor judging here, i am only trying to put some light on the facts :

Let's have a look at the recent restrictions as an example : golem, chalice and thorns restrictions lead to the fact that prison shop does not exist any more and shop archetype is now only ravager builds. Let's see how people react to that :

  • For 'blue' players a shop deck is a shop deck and they don't care about the details. So they look at the numbers and see that shop archetype as a whole is still high so they scream "Nothing changed, we need more restriction and so on ..."
  • Aggro shop players are quite happy (i presume)
  • Control/prison shop players are frustrated. Unless they find another deck in the format that suits their style (and they can afford to play) they won"t have much fun. They could try ravager decks but it is a really different style (same difference as between 'Death & Taxes' and '4C control' deck in Legacy for example).

So basically the result of those restrictions is that the overall frustration of the vintage communauty has increased. Maybe they were a necessary evil, i don't know and i won't judge but this is a result that can explain some of those endless talks on that topic.

My conclusions are :

  • Some amount of categorization is necessary because it is not the same people who play decks that could look very similar from a quick glance. 'Playing style' could be a criteria (i don't know if that can be done practically or if it would lead to something different from the usual criterium).
  • Statistical are great but they are tools, the real stuff are people who are playing.
  • The 'perfect' metagame for me would be one where any new player who wants to try the format and looks at the available/competitive decks could always find one that will suit his playing style and have a reasonable chance of winning with it.
last edited by albarkhane

I really can't believe I have to keep doing this. Shops decks are incredibly homogeneous at present. Non-Ravager Shops was 2.6 % of Champs this year. It's about the same in the online metagame. When I looked it up for someone else before the Thorn restriction, it was like 88% Ravager Shops, 11% 'Control' Shops, and 1% Combo Shops, and that percentage has likely been skewed further towards Ravager online. It's not worth it to me to put in the extra work. You want to do this @vaughnbros, you put in the effort.

And so what if you lump all Force of Will decks together? You see a large metagame share and a near 50% win rate. What does that prove? Absolutely nothing other than that the card is played a lot. Which is pretty much what you get with our metagame breakdown. Or a casual glance at the format. Your argument is unchanged - it's still absurd. I'm done with this topic...


Sigh. Yes, its been gone through before so I dont understand why you cant take the criticism and adapt from it. I appreciate the data collection, but just because you collected doesnt give you the right the skew it to whatever message you want and then be completely dismissive of all complaints/conclusions that others make of it. By this aggregation method and what seems to be an ad hoc exclusion criteria, you have basically tailored your data to what seems like a previously decided upon conclusion. This type of analysis is a fairly extreme form of bias, and nearly the exact opposite of the purpose of data (to find an objective truth). You are doing a disservice to all data analyses and science in general with an analysis like this, and your complete disregard for your peers' opinions on the matter. If you didnt want you have a discussion about the data as its concerned towards the Restricted list, maybe dont make a post with that as the focus? I dont know that seems like a pretty easy way to avoid the conversation.

last edited by vaughnbros

@vaughnbros I can take criticism and adapt and have done so in the past, but what you are saying is without merit and literally no one has spoken in agreement with you. So blame me, or @Smmenen, or @Brass-Man, or whoever. Or examine your own position and show some introspection.

last edited by Guest

@chubbyrain I have re-adjusted my position within this argument already and tried to be open to viewing it in different ways. People have also posted in agreement and rec'ed my posts. Apparently not relevant to you though apparently.

@vaughnbros I think you're making some unfair assumptions about our classification, which to be fair could be attributed to not knowing how we make them. Sure, Matt and I are primarily blue players. However we both know that there are different types of Shops, Dredge, Eldrazi, and Thalia decks. We've often asked if it's worth splitting these archetypes into Aggro Shops vs. Stax, Pitch vs. Anti-Hate, etc. As far as archetypes are concerned, Stax and distinct subarchetypes of dredge have never been a large enough percentage of the metagame that we felt that separating out the 1 person playing Stax at a tournament has been worth it. Matt and I originally started collecting data to help prepare for tournaments ourselves. We really don't care if a single person had a 45% or 60% winrate.

When we have access to the decklists, we've used the "Tags" idea that we came up with about a year ago. It was to us a good compromise between breaking archetypes into such small percentages of the metagame that it becomes meaningless and capturing the nuances of different broad brush stroke categories. To be honest, when we don't have decklists it's a pain in the ass to do this sort of nuanced categorization. Matt and I have done almost all this data collection by ourselves, and watching replays for an additional hour so we can check if the shops deck ever casts a smokestack or the dredge deck has a FOW somewhere is often unpalatable, especially if we're trying to convince someone else to do the work because we wanted to have a Saturday off. I also think it says something about the differentiation of blue decks, and not Matt and I, that we can almost immediately tell the difference between Xerox and Oath but need to watch 3 rounds of a shops deck to tell if it eventually casts a non aggro shops card.

last edited by diophan

Alright, I'm going to take a deep breath here and exhale. Let's approach this again.

I reread every single one of your posts in this thread, Lance. The critique you appear to be making is that our analysis is focusing too much on individual cards rather than the strategic elements that go into them. Hypothetically, decks should be broken down into the schools of Magic: aggro, control, tempo, combo, etc. Lumping decks into an archetype like 'Shops' ignores the strategic distinction. Similarly, decks like Blue are typically strategically 'control, or 'tempo', or 'combo'. Correct me if I'm wrong, but I believe this to be the gist of you position. If so, I would say you are coming at this from a theory-based approach or an academic approach.

Where we differ is that we want our data collection to be useful. We want people to look at our breakdowns and do something with them. This can be adjusting their main deck or sideboard to contain more Dredge hate or more Shops hate. From that perspective, it frequently doesn't matter what the ultimate strategic goal of a deck is. Cards in Magic don't typically interact with strategies. There isn't a card in Magic that says "counter target aggro spell or destroy target aggro permanent." Cards in Magic interact on different axes: colors, permanent types, hand or graveyard, the stack. So what is useful is a breakdown based on these levels of interaction. While it might certainly be useful to include different cards against a Control Shops deck vs an Aggro Shops deck, it's a moot point when one strategic orientation is much less than the other.

Another way people use data is to make informed ban and restriction decisions. We know WotC does this although many players seem to make these decisions on a more emotional level. Sobeit. I don't believe we are tailoring our data to specifically favor or disfavor specific restrictions. We are simply trying to aid players in looking at the metagame in a way that lets them evaluate their opinions with regards to the metagame. For instance, people were clamoring for a Paradoxical Outcomes restriction immediately after Gush was restricted. While one could certainly argue that the card is non-interactive, it's a different question of how popular or unpopular a card is and how frequently decks with that card win. When you look at the data, decks with Paradoxical Outcome do not seem overly dominant when you look at metagame share or win percentage.

I think it's very common in science to do this. Start with the hypothesis: is there a correlation between smoking and cancer? Set up your cohorts and collect data on different groups, and you should end up with a study that answers your question, albeit with the usual statistical limitations, potential biases, and possible confounders. Is what's being done here dissimilar to a cohort study? Does it not make sense to look at Shops decks relative to the field to see if there are metagame or balance concerns with decks that play this card? Note: you could do this with Force of Will or Mental Misstep, but I'd argue that those cards are so prevalent that the statistics here simply do not matter. You either feel these cards belong in the format or you don't at this point. Whether they are 60% of the metagame or 70% really shouldn't affect anyone's position.

In conclusion, we are approaching the taxonomy of Vintage from two fundamentally different ways. You have a theoretical approach and we have an observational approach that is a bit of a mess based on trying to meet the different needs and perspectives of Vintage players (combined with limitations like our time and the available information). So if you can propose a question or purpose that we can change our classification scheme to provide insight into, we would be happy to do so. However, if you think that what we are doing is theoretically sloppy and inconsistent, and a more rigid approach would be better, we would be happy to share our data with you.

last edited by Guest


Thank you for taking the time to address my comments.

I think theory vs reality is a spot on difference between classification systems. Id argue that theory (that properly incorporates reality) is more important when discussing DCI action than simply discussing reality.

What Id suggest for the most objective theoretical restricted list discussion is to aggregate by individual card that is being considered for restriction (although I understand this might be difficult to collect in practice as Ryan brings up). So what Id suggest as an alternative, is to aggregate at the lowest level that is collected/practical and allow individuals to draw their own conclusions from there. I thought the initial reports were especially good because of the low level of aggregation. This provided a level of transparency which I think is lost when we start discussing Workshops as a whole (and not specifically Ravager Shops as a separate deck). Now, again, if thats not possible, Id look into strategical stratifications of aggro/control/combo. From a theoretical restricted list standpoint, these are the most important classifications, imo, though I think they are not perfect by any means and am open to other suggestions (inclusion of tempo: aka Control/aggro? Seems fine to me). An additional alternative is the old Manadrain Pillars classifications (Drains, Shops, Bazaars, Rods, Rituals). While these are pretty horribly outdated, they also allow better historical comparisons and are an established classification system.

In sum, I think there are a number of ways to slice and dice the data. However, I think there are certain ways of analysis that should be avoided. The first being aggregation on different levels (e.g. Shops for some decks, but Xerox/Paradox/ect. for others is a no no imo). Aggregation on different levels presents a classification bias. It would be similar to seeing an aggregation on "white" by "country of origin" while leaving other races aggregated simply by race. This presents a false level of "diversity" that wouldnt be present if we focused on a single level of aggregation. Id also avoid deleting data. In my general experience analyzing numerous data sources, deletion of data should only be done in extreme situations (basically where it is unavailable). Even messy data (like in this case of Champs occurring simulatniously) still has quite a bit of information that can be gleaned.

I'd add one more additional suggestion: Present the initial data with little to no conclusions. Then go further on an additional post on your opinions on how to interpret the data. This provides everyone with a much clearer objectivity of the data collection, and separates the data from the opinions of those collecting it.

last edited by vaughnbros


Present the initial data with little to no conclusions

They DO provide the data with no conclusions. Every time. There's a link to the raw data they collected in every post. You can do analysis there if you want.

The first being aggregation on different levels (e.g. Shops for some decks, but Xerox/Paradox/ect. for others is a no no imo). Aggregation on different levels presents a classification bias

This is begging the question. I think that aggro shops and prison shops (such that it exists) are more similar decks than Mentor and Paradox are. You don't. These aren't different levels of classification, they're entirely different classification frameworks. I split these cards this way because I use the same sideboard cards to beat aggro shops and prison shops, but I use different sideboard cards to beat Mentor and Oath. I'm not sure where a drive to split decks up by color would. Different preferences for ink pigments? To me that feels like splitting up decks alphabetically or by artist.

The fact is nothing in this thread or any other thread has made me think that the data analysis is a useful for us to discuss B&R policy. (I think it's VERY useful for deckbuilding) I have literally never seen someone change their mind on this site after being presented with data. I have seen time and time again people publicly state their opinion about what should change, then react to data being posted with what a suprise! this confirms my opinion! Even when two people had different opinions and look at the same data, they both come out the other side more sure they were correct in the first place. If the data just can't line up with what they're saying, they complain about the validity of the data. I don't want to gang up on @vaughnbros here, because tons of people do this.

All B&R analysis is post-rationalization

You're either having fun playing the format, or you're not. Sure there are degrees of enjoyment, but you either believe a change would improve your fun or it wouldn't. If you're not having fun, you find an explanation, whether it's through tournament data or b&r precedent or test results. You find that explanation to make yourself feel justified for not having fun, and try to make a case for a change with it. Maybe your analysis is riddled with errors or maybe it's flawless, but if someone opposes you and there argument is terrible, it doesn't matter. No amount of attacking that argument is going to convince them that they're not actually having fun playing vintage if they are, or that they are having fun if they're not. You just end up with 170 posts of people talking over each other. Note that by the time you get here you've dramatically weakened your position from a WotC perspective - all they cared about is whether or not you were having fun. Once you bring metagame penetration into it, they have their own data and their own analysis methodology (such that it is).

If I thought for a second that I could ban the subject from the site, I would, but at this time I don't believe it can be done.


Present*, not provide. They are two different things. "Go do your own analysis" is basically as dismissive as you could be.

I dont think no data is better than some data.

I dont agree that these discussions should be banned.

@vaughnbros of course you don't agree they should be banned. Nobody in this thread wants them banned. The people who them banned are all people who have quit TMD by now.

edit: And for what it's worth, I'm actually telling you not to do the analysis. Which, admittedly, may be even more dismissive.

last edited by Brass Man
  • 218
  • 72278