Man I'm surprised a mod hasn't found this yet. Keep the discussion off of bans and ****. Go to the actual ban discussion thread for that. This is for GP discussion.
Edit:
Speak of the devil...
Ah, the perils of being lenient. How does the phrase go? "This is why we can't have nice things"?
One interesting thing about this GP that has not yet been mentioned (or at least, has gone undermentioned): Attendance. This was a huge event with 1461 players, a definite high point for Modern events. At first, I thought it was because a lot of Canadians came down to join. But according to the GP Detroit page, only 185 were in attendance. 1263 were Americans. Huge turnout that was unexpected by me, the community, Wizards, and the organizers. Anyone have thoughts on this?
just name it UR tempo and combine it into one. its the same deck. its like WUR control including restoration angels and snapcasters or just snapcasters.
Incidentally, the differences between UWR Midrange and UWR Control are also often hard to detect. The big indicator there, at least for me, is Geist. If I see a Geist, it's not counted as UWR Control. But as with Delver/Faeries, that distinction can often be lost when you don't have access to full lists. And even then, it's hard to tell when a deck crosses from Control to Midrange.
I do like the idea of combining both into a URx Tempo primer, consolidating discussion in both threads. Do you think that would muddle the conversation too much? Or is there enough overlap that it would make sense and even add voices to your deck development?
As part of the upcoming forum reorganization, some decks are getting shifted between subforums. Delver is going down to Established and Faeries is going down to Creation. But as we all know, the distinction between these two decks is very nuanced and can be difficult to tell from just looking over datasets. Some lists might get labeled as "Delver" when they are actually closer to "Faeries"; the biggest distinction between the two appears to be the inclusion of either Scion of Oona and/or Mistbind Clique (The latter very rarely on MTGO).
I am looking for your help in trying to figure out what the fairest and most accurate way is to handle these threads. Should they be combined into one? Should they remain distinct? What do you guys think?
As part of the upcoming forum reorganization, some decks are getting shifted between subforums. Delver is going down to Established and Faeries is going down to Creation. But as we all know, the distinction between these two decks is very nuanced and can be difficult to tell from just looking over datasets. Some lists might get labeled as "Delver" when they are actually closer to "Faeries"; the biggest distinction between the two appears to be the inclusion of either Scion of Oona and/or Mistbind Clique (The latter very rarely on MTGO).
I am looking for your help in trying to figure out what the fairest and most accurate way is to handle these threads. Should they be combined into one? Should they remain distinct? What do you guys think?
Established Tournament Decks with Results
A deck is considered "Established" if it does not meet the requirements of "Proven" but still fulfills ONE or more of the following criteria:
Has finished 4-0/3-1 at 12+ dailies in the past 3 months (above average finish count)
Has 1+ Grand Prix/Worlds/Pro Tour Top 16 appearance in the last 6 months
Makes up 1.75%+ of the Day 2 metagame at all GPs in the last 6 months (i.e. any deck with above average day 2 prevalence for all day 2 decks at Modern GPs)
Has placed in the Top 8 of 5+ paper event with 100+ players in the past 6 months (above average finish count)
Has placed in the Top 8 of 3+ MTGO Premier Events in the past 3 months
Under those definitions, which are much more inclusive, here is what the Established would look like. As with the previous section, I give the criteria that these decks fulfill under the Established definition. So for example, Hatebears fulfills criteria 1 and 5 but nothing else, so it gets a (1,5)
NEW ESTABLISHED
UR Delver (1,2,3,5)
Bogles (1)
GW Hatebears (1,5)
Wx Tokens (2,3)
Soul Sisters (1)
Gruul Zoo (1,2,3,4)
Mono U Tron (1)
UR Storm (1,2)
Living End (2)
Naya Zoo (2)
Domain Zoo (2,4)
Modern Merfolk (1)
Griselbrand (1,2)
Junk (1,2,3,4,5)
Death and Taxes (1)
Eternal Command (1)
Infect (4)
Finally, here are the current Established decks that would get moved down to Deck Creation:
DEMOTED TO CREATION
U(x) Faeries
4CC Gifts
Dredgevine
Assault Loam
Combo Elves
BUG Midrange
Ritual Gifts
Azorius Midrange
Goblins
Bant
Modern Boros
UW Tron
Once in Creation, at least two of those definitely have the qualifications to get stickied (Elves and UW Tron). Will have to look over the rest.
Surely BG Rock comes under category 2 also. You have it down under categories 3 and 4 but not 2.
Fixed! Lots of data entry leads to typographic mistakes!
I'm almost done with the Established criteria but am currently looking over MTGO Premiers and how best to incorporate them. It will probably be based on above average finishes, as with the other Established criteria.
You guys are getting renamed soon because the title of the thread doesn't exactly fit the naming conventions of this deck. What do you all think your new name should be?
If you guys agree that RUG Scapeshift is accurate and fine, that's alright too. Just let me know. Renaming will probably happen at the end of the week and will stay that way for the foreseeable future.
You guys are getting renamed soon because the title of the thread doesn't exactly fit the naming conventions of this deck. What do you all think your new name should be?
Based on some of the numbers izzetmage posted, along with his observations and my own, here are some revisions to the criteria for Proven. Numbers were tweaked to reflect reclassifications of decks on both MTGO and on Day 2 of GPs. For example, all the Twin variants just got combined, the different Pods just got classified as Kiki or Melira, etc. I also reclassified a ton of decks on TCDecks; those guys are both inconsistent and outright wrong in naming decks (e.g. Tron has about 4 different names in the last 6 months alone).
Proven Competitive Decks Representing the Current Metagame
A deck is considered "Proven" if it fulfills TWO or more of the following criteria:
Makes up 3%+ of the MTGO metagame over the last 3 months (i.e. has prevalence greater than one standard deviation over average MTGO deck prevalence)
Has 1+ Grand Prix/Pro Tour Top 8 appearance in the last 6 months
Has 2+ Grand Prix/Pro Tour Top 16 appearances in the last 6 months
Makes up 3.75%+ of the Day 2 metagame at all GPs in the last 6 months (i.e. has prevalence greater than one standard deviation over average day 2 prevalence at GPs)
Makes up 4.50%+ of the Paper Top 8 metagame for events with 100+ players (i.e. has prevalence greater than one standard deviation over average deck prevalence at large paper events)
Under this new criteria, here are the decks that would fit into Proven, followed by a list of decks that do not meet the new Proven criteria. I have also given the numbered criteria that the decks meet, just for the sake of transparency. For example, Affinity fulfills all the criteria, so it gets a (1,2,3,4,5) next to it.
NEW PROVEN DECKS
Affinity (1,2,3,4,5)
UWR Control (1,2)
Melira Pod (1,2,3,4)
RG Tron (1,4)
Twin (1,2,3,4,5)
BG Rock (2,3,4)
Scapeshift (1,2,3,4,5)
Burn (1,5)
Jund (1,2,3,4,5)
Kiki Pod (3,5)
UWR Midrange (3,5)
@Izzetmage:
If we change the criteria to be 2+ Top 8 appearances, then both Kiki and UWR Midrange are out. That feels wrong to me as it does to you. I am personally comfortable keeping it at 1+ Top 8 appearances because all the other qualifications effectively control for the "lightning strikes once" phenomenon. Because decks need to meet TWO criteria, it's probably okay.
And then here are the decks which are moved out of Proven back down to Established, along with the Proven criteria that they filled (if any).
DEMOTED FROM PROVEN TO ESTABLISHED
UR Delver (1)
Bogles
GW Hatebears
Wx Tokens
Soul Sisters
Gruul Zoo
Mono U Tron
UR Storm (1)
Living End (2)
Naya Zoo (2)
How does that look for PROVEN? I am still looking over Established criteria to figure out the most logical way to incorporate Modern MTGO Premier data (top 8 lists) into the definition, as well as how that might affect other criteria. Also, the new Proven threads are going to get renamed to fit the standard deck naming conventions. So "RUG Scapeshift" is just going to be "Scapeshift". "SplinterKiki-MiteExarch" is going to become "Twin Combo" or "Kiki Twin".
I'm sure the data is there to support these decks remaining in "Established" but I seems like this is one of those cases where the spirit of the law is being violated if not the letter of it.
Assault Loam for example, I just...wow. Now please keep in mind I play on MTGO exclusively where games are extremely competitive 24/7. For this reason anything marginal is really not seen very often if at all.
The other problem I have is that while some of these decks like Faeries and Reanimator probably belong on the list, the deck lists can vary widely. There are a ton of awful "reanimator" lists out there for every one that is marginally playable. Same with Faeries and Junk and a few others.
It seems misleading, but I guess it doesn't really matter. Creation is still the best forum for the development of decks. The new "Established" list is essentially a graveyard forum for decks that are too "good" to be moved to archives, but not interesting enough to warrant much discussion or thread activity of any kind.
Dredgevine got 3rd at a 481 player tournament in Italy back in June. Under the fairly inclusive Established criteria, that keeps it in Established.
Assault Loam is getting moved to Creation
Reanimator is being renamed "Griselbrand Reanimator" or "Griselcannon" or "Glass Cannon Griselbrand" or something similar.
Junk is staying as Junk. They can internally decide what constitutes Junk within the thread and report/redirect posts that clearly don't belong.
The Faeries and Delver distinction is tricky. I would be happy to talk to posters that frequent those threads about a way to distinguish between them in a meaningful way. It's especially problematic given that they are both going to be in Established.
For the tenth plus time. Established has some threads that receive a ton of traffic (D&T, Merfolk, 4C Gifts, Faeries, Junk, etc.). These threads receive essentially equal traffic, or greater in the case of D&T/Merfolk/Faeries, than the stickied threads in Creation. Other threads don't get any traffic, but in many cases, that is because they aren't competitive (Bant, Boros, BUG, Azorius, Ritual Gifts, etc.). This shows that there is nothing inherently wrong with the subforum, as you keep insisting. It just means there is a problem with the decks in that subforum. And also for the tenth plus time, is there anyone else in this forum that wants to see something different done with Established? And if so, what?
I entered everything in by hand so there might be some errors.
GP Portland Day 2 was not provided.
I'm wondering if 3.5%+ (above average prevalence) is too high of a bar for Established. Maybe it should be lower? Perhaps one standard deviation BELOW average? Also, your average prevalence appears to be different than the one I calculated (yours is 1.8% by my calculations). Will recheck my math later and see where that 3.5% came from, but if I'm wrong, I'd be happy to switch to the 1.8%+ as the standard.
This is known as "self-selection bias" and makes any argument of "this thread is full of people who hate the banned list!" null and void. Note that I don't have a horse in this argument, but there's simply no way that you can point at a place where people post without being compelled and claim that it looks anything like a correct statistical distribution.
Everyone, yourself included, knows that this gives the "squeaky wheels" an inordinate voice. Everyone also knows that voluntary posting is much more likely to be done when someone has a problem than when someone thinks everything is great.
There are no relevant statistics in a forum. Even in a forum poll, however fun it is to post them. Please don't try to find any here.
I'm going to have to agree with Barandis here regarding the sample of Magic players on our forum. We have no reason to believe that it is representative. It certainly isn't random, and it is very difficult to identify the mechanisms that lead some players to go to forums over others. Moreover, even amongst those users that go to a forum, we only have a very small number that post in this thread.
In this current iteration of the banlist thread, there have been a grand total of 255 unique, unduplicated posters. On average, each poster has made 12 contributions to the thread, but the population's standard deviation is pretty large at 29. So we can already tell that this thread is representing a vocal minority. This interpretation is supported by the general distribution of posts across the thread, with just 15 users making up 50% of posts in this thread. Stated another way, 6% of the unduplicated posters in this thread make up 50% of the posts. Going up to 75% of the posts brings us to 17% of the posters, which is still pretty darn disproportionate.
No matter how you shake it, I can't put my statistician credentials behind using this thread as a representative sample of Magic players, or even Modern players. Everyone has selected into this thread, and unless we account for that selection, we aren't going to be able to make any conclusions.
A much more reasonable interpretation of the thread is this: Vocal Modern players with a history of online Magic activity tend to disagree with the banlist (and maybe Modern's direction as a format). That's much more accurate but still based on an extremely biased sample.
I'm just curious how Infect made it into Established. I am a fan of the deck, heck I was one of the early creators and testers of the Modern version of the deck here on MTGS, but it has not jumped out to me from results in a while. That said, you have data that I do not so I trust that it meets the criteria, I am just wondering how.
Infect got in because it has 4+ Daily finishes at 4-0/3-1 in the last few months (only 4, I think), and because it had a number of Top 8 finishes at 50+ player events in the last 6 months. Most of those were back in March with a few over the summer, but they technically fall under the 6 year perspective.
The MTGS Modern Forums is a hub for discussion, debate, and decklists on the best decks in the Modern format. The "Proven" subforum reflects the Tier 1 decks of the metagame. The subforum is updated every ban quarter.
Criteria for what makes a deck "Proven" were decided upon by staff and everyday forum users in August and September of 2013. It has also received small updates since then. Data collected for this quarter's update is presented in the MTGS Modern Metagame Analysis: 7/2014 - 9/2014 Google Doc. This sheet compiles all Modern event data for your convenience, and gives an accurate depiction of both the online and paper metagames. Many of the criteria for Proven are dynamic (based on averages/standard deviations for the population), so the exact cutoffs are always changing to reflect the current metagame. The formulas, however, remain the same. Criteria for the 9/22/2014 update are given below:
Proven - 9/22/2014 Competitive Decks Representing the Current Metagame
Proven decks are updated every ban cycle. A Proven thread must have a primer updated since the last ban cycle. A deck is considered "Proven" if it fulfills THREE or more of the following criteria:
Makes up 3.25%+ of the MTGO metagame (reported 4-0/3-1 and T16 decks at Dalies/Premiers) over the last 3 months
(i.e. has prevalence greater than one standard deviation over average MTGO deck prevalence)
Has 1+ Grand Prix/Pro Tour Top 8 appearance in the last 6 months
Has 3+ Grand Prix/Pro Tour Top 16 appearances in the last 6 months
Makes up 3.5%+ of the Day 2 metagame at all GPs in the last 6 months/since the most recent significant ban update
(i.e. has prevalence greater than one standard deviation over average day 2 prevalence at GPs)
Makes up 3.25%+ of the Paper Top 8 metagame for events, held in the last 6 months/since the most recent significant ban update, with 40+ players
(i.e. has prevalence greater than one standard deviation over average deck prevalence at above average attendance paper events)
Under these criteria, here are the decks that go into the Proven subforum for the 9/22/2014 update. Again, data is taken from the spreadsheet here. Decks are also accompanied by the criteria that they fulfill on the list, just for the sake of transparency. For example, Affinity fulfills all the criteria, so it gets a (1,2,3,4,5) next to it. RG Tron only fulfills two, so it receives a (1,5).
If anyone has any questions, comments, or suggestions, feel free to post in the Modern metagame discussion thread, or to message me or one of the other moderators.
Update again! More moves and more reorganization coming! This is basically the final stage, so input is critical to making sure that the process reflects the desires of the community and advances the mission of our site. Check out the Official Reorganization Thread for details on the upcoming changes to Proven and Established!
Ah, the perils of being lenient. How does the phrase go? "This is why we can't have nice things"?
One interesting thing about this GP that has not yet been mentioned (or at least, has gone undermentioned): Attendance. This was a huge event with 1461 players, a definite high point for Modern events. At first, I thought it was because a lot of Canadians came down to join. But according to the GP Detroit page, only 185 were in attendance. 1263 were Americans. Huge turnout that was unexpected by me, the community, Wizards, and the organizers. Anyone have thoughts on this?
Incidentally, the differences between UWR Midrange and UWR Control are also often hard to detect. The big indicator there, at least for me, is Geist. If I see a Geist, it's not counted as UWR Control. But as with Delver/Faeries, that distinction can often be lost when you don't have access to full lists. And even then, it's hard to tell when a deck crosses from Control to Midrange.
I do like the idea of combining both into a URx Tempo primer, consolidating discussion in both threads. Do you think that would muddle the conversation too much? Or is there enough overlap that it would make sense and even add voices to your deck development?
I am looking for your help in trying to figure out what the fairest and most accurate way is to handle these threads. Should they be combined into one? Should they remain distinct? What do you guys think?
I am looking for your help in trying to figure out what the fairest and most accurate way is to handle these threads. Should they be combined into one? Should they remain distinct? What do you guys think?
Established
Tournament Decks with Results
A deck is considered "Established" if it does not meet the requirements of "Proven" but still fulfills ONE or more of the following criteria:
Under those definitions, which are much more inclusive, here is what the Established would look like. As with the previous section, I give the criteria that these decks fulfill under the Established definition. So for example, Hatebears fulfills criteria 1 and 5 but nothing else, so it gets a (1,5)
NEW ESTABLISHED
UR Delver (1,2,3,5)
Bogles (1)
GW Hatebears (1,5)
Wx Tokens (2,3)
Soul Sisters (1)
Gruul Zoo (1,2,3,4)
Mono U Tron (1)
UR Storm (1,2)
Living End (2)
Naya Zoo (2)
Domain Zoo (2,4)
Modern Merfolk (1)
Griselbrand (1,2)
Junk (1,2,3,4,5)
Death and Taxes (1)
Eternal Command (1)
Infect (4)
Finally, here are the current Established decks that would get moved down to Deck Creation:
DEMOTED TO CREATION
U(x) Faeries
4CC Gifts
Dredgevine
Assault Loam
Combo Elves
BUG Midrange
Ritual Gifts
Azorius Midrange
Goblins
Bant
Modern Boros
UW Tron
Once in Creation, at least two of those definitely have the qualifications to get stickied (Elves and UW Tron). Will have to look over the rest.
Fixed! Lots of data entry leads to typographic mistakes!
I'm almost done with the Established criteria but am currently looking over MTGO Premiers and how best to incorporate them. It will probably be based on above average finishes, as with the other Established criteria.
Ideas:
1. Scapeshift Valakut
2. Scapeshift
3. Scapeshift Combo
4. _________
If you guys agree that RUG Scapeshift is accurate and fine, that's alright too. Just let me know. Renaming will probably happen at the end of the week and will stay that way for the foreseeable future.
Ideas:
1. Kiki Twin
2. Twin Combo
3. Splinter Twin
4. _________
Fire away! Renaming will probably happen at the end of the week and will stay that way for the foreseeable future.
Proven
Competitive Decks Representing the Current Metagame
A deck is considered "Proven" if it fulfills TWO or more of the following criteria:
Under this new criteria, here are the decks that would fit into Proven, followed by a list of decks that do not meet the new Proven criteria. I have also given the numbered criteria that the decks meet, just for the sake of transparency. For example, Affinity fulfills all the criteria, so it gets a (1,2,3,4,5) next to it.
NEW PROVEN DECKS
Affinity (1,2,3,4,5)
UWR Control (1,2)
Melira Pod (1,2,3,4)
RG Tron (1,4)
Twin (1,2,3,4,5)
BG Rock (2,3,4)
Scapeshift (1,2,3,4,5)
Burn (1,5)
Jund (1,2,3,4,5)
Kiki Pod (3,5)
UWR Midrange (3,5)
@Izzetmage:
If we change the criteria to be 2+ Top 8 appearances, then both Kiki and UWR Midrange are out. That feels wrong to me as it does to you. I am personally comfortable keeping it at 1+ Top 8 appearances because all the other qualifications effectively control for the "lightning strikes once" phenomenon. Because decks need to meet TWO criteria, it's probably okay.
And then here are the decks which are moved out of Proven back down to Established, along with the Proven criteria that they filled (if any).
DEMOTED FROM PROVEN TO ESTABLISHED
UR Delver (1)
Bogles
GW Hatebears
Wx Tokens
Soul Sisters
Gruul Zoo
Mono U Tron
UR Storm (1)
Living End (2)
Naya Zoo (2)
How does that look for PROVEN? I am still looking over Established criteria to figure out the most logical way to incorporate Modern MTGO Premier data (top 8 lists) into the definition, as well as how that might affect other criteria. Also, the new Proven threads are going to get renamed to fit the standard deck naming conventions. So "RUG Scapeshift" is just going to be "Scapeshift". "SplinterKiki-MiteExarch" is going to become "Twin Combo" or "Kiki Twin".
I'm wondering if 3.5%+ (above average prevalence) is too high of a bar for Established. Maybe it should be lower? Perhaps one standard deviation BELOW average? Also, your average prevalence appears to be different than the one I calculated (yours is 1.8% by my calculations). Will recheck my math later and see where that 3.5% came from, but if I'm wrong, I'd be happy to switch to the 1.8%+ as the standard.
I'm going to have to agree with Barandis here regarding the sample of Magic players on our forum. We have no reason to believe that it is representative. It certainly isn't random, and it is very difficult to identify the mechanisms that lead some players to go to forums over others. Moreover, even amongst those users that go to a forum, we only have a very small number that post in this thread.
In this current iteration of the banlist thread, there have been a grand total of 255 unique, unduplicated posters. On average, each poster has made 12 contributions to the thread, but the population's standard deviation is pretty large at 29. So we can already tell that this thread is representing a vocal minority. This interpretation is supported by the general distribution of posts across the thread, with just 15 users making up 50% of posts in this thread. Stated another way, 6% of the unduplicated posters in this thread make up 50% of the posts. Going up to 75% of the posts brings us to 17% of the posters, which is still pretty darn disproportionate.
No matter how you shake it, I can't put my statistician credentials behind using this thread as a representative sample of Magic players, or even Modern players. Everyone has selected into this thread, and unless we account for that selection, we aren't going to be able to make any conclusions.
A much more reasonable interpretation of the thread is this: Vocal Modern players with a history of online Magic activity tend to disagree with the banlist (and maybe Modern's direction as a format). That's much more accurate but still based on an extremely biased sample.
Infect got in because it has 4+ Daily finishes at 4-0/3-1 in the last few months (only 4, I think), and because it had a number of Top 8 finishes at 50+ player events in the last 6 months. Most of those were back in March with a few over the summer, but they technically fall under the 6 year perspective.
The MTGS Modern Forums is a hub for discussion, debate, and decklists on the best decks in the Modern format. The "Proven" subforum reflects the Tier 1 decks of the metagame. The subforum is updated every ban quarter.
Criteria for what makes a deck "Proven" were decided upon by staff and everyday forum users in August and September of 2013. It has also received small updates since then. Data collected for this quarter's update is presented in the MTGS Modern Metagame Analysis: 7/2014 - 9/2014 Google Doc. This sheet compiles all Modern event data for your convenience, and gives an accurate depiction of both the online and paper metagames. Many of the criteria for Proven are dynamic (based on averages/standard deviations for the population), so the exact cutoffs are always changing to reflect the current metagame. The formulas, however, remain the same. Criteria for the 9/22/2014 update are given below:
Proven - 9/22/2014
Competitive Decks Representing the Current Metagame
Proven decks are updated every ban cycle. A Proven thread must have a primer updated since the last ban cycle. A deck is considered "Proven" if it fulfills THREE or more of the following criteria:
- Makes up 3.25%+ of the MTGO metagame (reported 4-0/3-1 and T16 decks at Dalies/Premiers) over the last 3 months
(i.e. has prevalence greater than one standard deviation over average MTGO deck prevalence)
- Has 1+ Grand Prix/Pro Tour Top 8 appearance in the last 6 months
- Has 3+ Grand Prix/Pro Tour Top 16 appearances in the last 6 months
- Makes up 3.5%+ of the Day 2 metagame at all GPs in the last 6 months/since the most recent significant ban update
(i.e. has prevalence greater than one standard deviation over average day 2 prevalence at GPs)
- Makes up 3.25%+ of the Paper Top 8 metagame for events, held in the last 6 months/since the most recent significant ban update, with 40+ players
(i.e. has prevalence greater than one standard deviation over average deck prevalence at above average attendance paper events)
Under these criteria, here are the decks that go into the Proven subforum for the 9/22/2014 update. Again, data is taken from the spreadsheet here. Decks are also accompanied by the criteria that they fulfill on the list, just for the sake of transparency. For example, Affinity fulfills all the criteria, so it gets a (1,2,3,4,5) next to it. RG Tron only fulfills two, so it receives a (1,5).PROVEN DECKS: 9/22/2014
- Affinity (1,2,3,4,5)
- UR Twin (1,2,3,4,5)
- Melira Pod (1,2,3,4,5)
- BG Rock (1,2,3,4,5)
- Jund (2,3,4,5)
- Scapeshift (1,2,3,4,5)
- RG Tron (1,2,5)
- UWR Control (2,3,4)
- RUG Twin (2,3,4,5)
- Burn (1,2,4,5)
If anyone has any questions, comments, or suggestions, feel free to post in the Modern metagame discussion thread, or to message me or one of the other moderators.