To clarify, Abzan only refers to the midrange version of the deck and excludes something like Abzan/Devoted Company, correct?
I believe those are in the "GWx Company" category.
I'm more curious how Grixis shadow is in tier 3; more so than that, every individual share of the meta data point for it is higher than the average share of the meta so there is clearly some amount of time this has to be going back to make that happen. I'm not sure how useful the percentages 8 months ago are to the current meta game but perhaps I am missing the point of this tiering?
A global tier list identifies not only how strong a deck is but also how strong a deck has been. Going back 8 months is very useful when exist decks that have been tier 1 for the whole 8 months! If you are interested in last week ladder you have the link to the weekly data which are gathered and provided by Ashockfan.
I was referring to the 6/3 update where it is listed as untiered as seen in the screenshot bellow. The 6/17 list has it as tier 3 as seen in the screenshot bellow. The first post has it currently as tier 1. So maybe clean up your data some so it doesn't show conflicting information. UBx shadow has been tier for longer than both of those updates and while it is reflected in the main post, it isn't supported by your own data as seen in the screenshots bellow. Whether this is an oversight, improper data entry, or a flaw in your system, it needs to be corrected if you are going to be setting the tier lists for the entire site.
Forgive me if i've missed something obvious, but I really can't get my head around the ladder idea with points. Specifically:
If points are converted into a percentage of the total points obtainable in that category, how is it that a deck can ever achieve 100%? Surely this means that one deck received all the points obtainable that week, and no other decks got anything?
Honestly I'm quite perplexed that I find it this confusing because I teach physics so I'm no stranger to maths haha. What have I missed?
Private Mod Note
():
Rollback Post to RevisionRollBack
Modern: G Tron, Vannifar, Jund, Druid/Vizier combo, Humans, Eldrazi Stompy (Serum Powder), Amulet, Grishoalbrand, Breach Titan, Turns, Eternal Command, As Foretold Living End, Elves, Cheerios, RUG Scapeshift
Forgive me if i've missed something obvious, but I really can't get my head around the ladder idea with points. Specifically:
If points are converted into a percentage of the total points obtainable in that category, how is it that a deck can ever achieve 100%? Surely this means that one deck received all the points obtainable that week, and no other decks got anything?
Honestly I'm quite perplexed that I find it this confusing because I teach physics so I'm no stranger to maths haha. What have I missed?
Just a refreshing then
lim (a+n)/(b+n) = 1 as n grows.
numerator = number of weeks you have been tier 1
denominator = number of weeks total
If you reach tier 1 every week you add 1 to both numerator and denominator, and you get the limit.
Forgive me if i've missed something obvious, but I really can't get my head around the ladder idea with points. Specifically:
If points are converted into a percentage of the total points obtainable in that category, how is it that a deck can ever achieve 100%? Surely this means that one deck received all the points obtainable that week, and no other decks got anything?
Honestly I'm quite perplexed that I find it this confusing because I teach physics so I'm no stranger to maths haha. What have I missed?
Just a refreshing then
lim (a+n)/(b+n) = 1 as n grows.
numerator = number of weeks you have been tier 1
denominator = number of weeks total
If you reach tier 1 every week you add 1 to both numerator and denominator, and you get the limit.
Yeah, makes sense. Thanks! My confusion was thinking that the "total available" applied to multiple decks and so the percentage was a percentage control of the metagame. My bad, so thanks for clearing it up.
For what it's worth though, I don't think it's worded very well in the description?
Private Mod Note
():
Rollback Post to RevisionRollBack
Modern: G Tron, Vannifar, Jund, Druid/Vizier combo, Humans, Eldrazi Stompy (Serum Powder), Amulet, Grishoalbrand, Breach Titan, Turns, Eternal Command, As Foretold Living End, Elves, Cheerios, RUG Scapeshift
Sorry I'm a little late on the updates. Heres the shifts:
Azban Company, BW Eldrazi, Pinned in Developing Competitive for being tier 3 this month.
I also Pinned the other tier 3 stuff thats been tier 3 for a while to seperate them from the life time members of Developing Competitive
Why not just add a tier 3 sub-forum? Seems like a simpler/cleaner solution to the problem, and it's a reasonable distinction in such a deep and diverse format.
Sorry I'm a little late on the updates. Heres the shifts:
Azban Company, BW Eldrazi, Pinned in Developing Competitive for being tier 3 this month.
I also Pinned the other tier 3 stuff thats been tier 3 for a while to seperate them from the life time members of Developing Competitive
Why not just add a tier 3 sub-forum? Seems like a simpler/cleaner solution to the problem, and it's a reasonable distinction in such a deep and diverse format.
Sorta like when you give a mouse a cookie situation. Thats why we have deck constructed and developing competitive in the first place. For all intensive purposes developing competitive is tier 3.
Sorry I'm a little late on the updates. Heres the shifts:
Azban Company, BW Eldrazi, Pinned in Developing Competitive for being tier 3 this month.
I also Pinned the other tier 3 stuff thats been tier 3 for a while to seperate them from the life time members of Developing Competitive
Why not just add a tier 3 sub-forum? Seems like a simpler/cleaner solution to the problem, and it's a reasonable distinction in such a deep and diverse format.
Sorta like when you give a mouse a cookie situation. Thats why we have deck constructed and developing competitive in the first place. For all intensive purposes developing competitive is tier 3.
Would it be confusing to call it tier 3? Seems like a fairly easy change to make.
Of course you may then ask the question "what's the point. It's just nomenclature" and you'd be right to ask. I just think tier 3 sounds better is all. More fitting with the rest of the forum sections in modern. Nothing particularly deep or whatever
Private Mod Note
():
Rollback Post to RevisionRollBack
Modern: G Tron, Vannifar, Jund, Druid/Vizier combo, Humans, Eldrazi Stompy (Serum Powder), Amulet, Grishoalbrand, Breach Titan, Turns, Eternal Command, As Foretold Living End, Elves, Cheerios, RUG Scapeshift
Well mostly because it would then cause "why is ascension storm or 8 rack on there when they are not tier 3." So wed have to make a tier 4.
Just keeps going.
The pin is a visable elegant way to just put a line in the sand and move on. Its not a big deal. anyone who cares THAT much about what are tier 3 and 4 decks can just use the ranking threads, which are literally pinned to the front of the sub.
MTGsal is a communication hub for deck building, not the go to place for rankings.
All time ladder construction: each week every deck above 3.50% meta share gets a tier1-point, every other deck above 1.50% gets a tier2-point, every other deck above 0.50% gets a tier3-point. The following ratios are then calculated for each deck: tier1points/total number of weeks, (tier1points+tier2points)/total number of weeks, (tier1points+tier2points+tier3points)/total number of weeks. These ratios tell how many weeks, over the total number of weeks, a deck has been tier1, at least tier2 and at least tier3.
All time tier1 decks are those decks with a tier1 points ratio greater or equal than 0.75 (=75%). That means those decks have been tier 1 at least 75% of the weeks. E.g. in 10 weeks a deck has been tier 1 for 8 weeks and tier 2 for 2 weeks, so it has a tier 1 percentage equal to 80%.
All time tier2 decks are, among the others, those decks with a tier1 + tier2 points ratio greater or equal than 0.66 (=66%). That means those decks have been at least tier 2 at least 66% of the weeks. E.g. in 10 weeks a deck has been tier 1 for 3 weeks and tier 2 for 4 weeks, so it has a tier 2 percentage equal to 70%.
All time tier3 decks are, among the others, those decks with a tier1 + tier2 + tier3 points ratio greater or equal than 0.66 (=66%). That means those decks have been at least tier 3 at least 66% of the weeks. E.g. in 10 weeks a deck has been tier 1 for 1 week, tier 2 for 2 weeks and tier 3 for 6 weeks, so it has a tier 3 percentage equal to 90%.
How does the new changes to the released list effect tier lists? since they wont print repeats anymore, we cant really get a "sample" now right?
Unless there is a change in policy for release of data, the actively-chosen 5-0 lists no longer represent even a random sampling of the format (which can normalize over time). So the only thing we have to go on moving forward are a handful of paper lists every few weeks and months. Until they give us MTGO data, the days of meta tracking are effectively over.
Right. I hought that might the case. Perhaps paper is the only good way to track it now.
Which is truly unfortunate. A handful of lists every couple weeks or months is too few data points too far apart to be anywhere near as accurate and relevant as MTGO data.
Would it be a stretch to try to get larger LGS staff/TOs to try and voluntarily submit data about Top 8s? I know Modern Monday/FNMs are not even remotely as important as other bigger events, but collectively tracking smaller events with 20-30+ people nationally/internationally, coupled with other paper tournaments, would probably be a better representation. Basically, with a large enough sample size and limiting the sample to larger small tournaments coupled with a lower statistical weight, you could help eliminate meta bias and still increase the overall data pool.
Unless there is a change in policy for release of data, the actively-chosen 5-0 lists no longer represent even a random sampling of the format (which can normalize over time). So the only thing we have to go on moving forward are a handful of paper lists every few weeks and months. Until they give us MTGO data, the days of meta tracking are effectively over.
All time ladder construction: each week every deck above 3.50% meta share gets a tier1-point, every other deck above 1.50% gets a tier2-point, every other deck above 0.50% gets a tier3-point. The following ratios are then calculated for each deck: tier1points/total number of weeks, (tier1points+tier2points)/total number of weeks, (tier1points+tier2points+tier3points)/total number of weeks. These ratios tell how many weeks, over the total number of weeks, a deck has been tier1, at least tier2 and at least tier3.
All time tier1 decks are those decks with a tier1 points ratio greater or equal than 0.75 (=75%). That means those decks have been tier 1 at least 75% of the weeks. E.g. in 10 weeks a deck has been tier 1 for 8 weeks and tier 2 for 2 weeks, so it has a tier 1 percentage equal to 80%.
All time tier2 decks are, among the others, those decks with a tier1 + tier2 points ratio greater or equal than 0.66 (=66%). That means those decks have been at least tier 2 at least 66% of the weeks. E.g. in 10 weeks a deck has been tier 1 for 3 weeks and tier 2 for 4 weeks, so it has a tier 2 percentage equal to 70%.
All time tier3 decks are, among the others, those decks with a tier1 + tier2 + tier3 points ratio greater or equal than 0.66 (=66%). That means those decks have been at least tier 3 at least 66% of the weeks. E.g. in 10 weeks a deck has been tier 1 for 1 week, tier 2 for 2 weeks and tier 3 for 6 weeks, so it has a tier 3 percentage equal to 90%.
Private Mod Note
():
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
I believe those are in the "GWx Company" category.
I'm more curious how Grixis shadow is in tier 3; more so than that, every individual share of the meta data point for it is higher than the average share of the meta so there is clearly some amount of time this has to be going back to make that happen. I'm not sure how useful the percentages 8 months ago are to the current meta game but perhaps I am missing the point of this tiering?
Also This link from the 1st post is dead.
If points are converted into a percentage of the total points obtainable in that category, how is it that a deck can ever achieve 100%? Surely this means that one deck received all the points obtainable that week, and no other decks got anything?
Honestly I'm quite perplexed that I find it this confusing because I teach physics so I'm no stranger to maths haha. What have I missed?
lim (a+n)/(b+n) = 1 as n grows.
numerator = number of weeks you have been tier 1
denominator = number of weeks total
If you reach tier 1 every week you add 1 to both numerator and denominator, and you get the limit.
Yeah, makes sense. Thanks! My confusion was thinking that the "total available" applied to multiple decks and so the percentage was a percentage control of the metagame. My bad, so thanks for clearing it up.
For what it's worth though, I don't think it's worded very well in the description?
Why not just add a tier 3 sub-forum? Seems like a simpler/cleaner solution to the problem, and it's a reasonable distinction in such a deep and diverse format.
Sorta like when you give a mouse a cookie situation. Thats why we have deck constructed and developing competitive in the first place. For all intensive purposes developing competitive is tier 3.
Would it be confusing to call it tier 3? Seems like a fairly easy change to make.
Of course you may then ask the question "what's the point. It's just nomenclature" and you'd be right to ask. I just think tier 3 sounds better is all. More fitting with the rest of the forum sections in modern. Nothing particularly deep or whatever
Just keeps going.
The pin is a visable elegant way to just put a line in the sand and move on. Its not a big deal. anyone who cares THAT much about what are tier 3 and 4 decks can just use the ranking threads, which are literally pinned to the front of the sub.
MTGsal is a communication hub for deck building, not the go to place for rankings.
Some interesting uptrending charts
Some interesting downtrending charts
Zac Pinales' weekly ladder and source data
All time ladder construction: each week every deck above 3.50% meta share gets a tier1-point, every other deck above 1.50% gets a tier2-point, every other deck above 0.50% gets a tier3-point. The following ratios are then calculated for each deck: tier1points/total number of weeks, (tier1points+tier2points)/total number of weeks, (tier1points+tier2points+tier3points)/total number of weeks. These ratios tell how many weeks, over the total number of weeks, a deck has been tier1, at least tier2 and at least tier3.
Unless there is a change in policy for release of data, the actively-chosen 5-0 lists no longer represent even a random sampling of the format (which can normalize over time). So the only thing we have to go on moving forward are a handful of paper lists every few weeks and months. Until they give us MTGO data, the days of meta tracking are effectively over.
UR ....... WUBR ........... WB ............. RGW ........ UBR ....... WUB .... BGU
Spells / Blink & Combo / Token Grind / Dino Tribal / Draw Cards / Zombies / Reanimate
Which is truly unfortunate. A handful of lists every couple weeks or months is too few data points too far apart to be anywhere near as accurate and relevant as MTGO data.
UR ....... WUBR ........... WB ............. RGW ........ UBR ....... WUB .... BGU
Spells / Blink & Combo / Token Grind / Dino Tribal / Draw Cards / Zombies / Reanimate
Spirits
And that is pretty limiting. And paper and online formats differ so much do to price of all things.
Quote.
Some interesting uptrending charts
Some interesting downtrending charts
All time ladder construction: each week every deck above 3.50% meta share gets a tier1-point, every other deck above 1.50% gets a tier2-point, every other deck above 0.50% gets a tier3-point. The following ratios are then calculated for each deck: tier1points/total number of weeks, (tier1points+tier2points)/total number of weeks, (tier1points+tier2points+tier3points)/total number of weeks. These ratios tell how many weeks, over the total number of weeks, a deck has been tier1, at least tier2 and at least tier3.