LMF and the community input

[pooispoois]

As soon as we come to a decision about the variables involved. We need to decide on the following:

1. How many players a tournament must have to qualify.
2. How far back we want to go chronologically.
3. The minimum number of times a deck must place to qualify.
4. How often we intend to update the LMF with new results.
5. Whether or not we include foreign tournaments.

Currently, the proposed variables are as follows:

1. 50
2. 3 months
3. Twice
4. Once a month
5. Probably, yes.

Note, and this is important: none of these variables are currently set in stone. There needs to be a more thorough analyzation of the data and more comprehensive discussion amongst mods and adepts to decide if these are the variables we want to use. We'll keep you posted.

I think this system is pretty good with these variables, only I will suggest a system I feel is more useful than solution #2 to keep a fairly updated LMF section that also dismisses flukes (like possibly Reanimator as a recent exmaple). Every deck in that section has its "Consistency Points". These points are awarded at the end of the month, for each 2 Top 8 showings [in accountable tournaments] it made. For example, at the end of a month where ******** Top 8'd 6 times, it would be awarded 3 consistency points. Then, at the end of each month where a deck in the LMF area didn't post at least two top 8 showings, it gets a consistency point removed, and a deck without consistency points gets entirely removed from the LMF.
An hypotetical example: Reanimator would get added to the LMF and awarded a consistency point for two Top 8 showings in the month, but then the next month only one Reanimator deck manages to Top 8, so the deck loses its only consistency point and gets removed (it is what would be considered "a fluke"). Also, one could evluate the strength of a deck in the metagame by its consistency points, as well as how long it has been performing and, naturally and given the name I gave the points, its consistency at achieving Top 8s.

[dontbiteitholmes]

As soon as we come to a decision about the variables involved. We need to decide on the following:

1. How many players a tournament must have to qualify.
2. How far back we want to go chronologically.
3. The minimum number of times a deck must place to qualify.
4. How often we intend to update the LMF with new results.
5. Whether or not we include foreign tournaments.

Currently, the proposed variables are as follows:

1. 50
2. 3 months
3. Twice
4. Once a month
5. Probably, yes.

Note, and this is important: none of these variables are currently set in stone. There needs to be a more thorough analyzation of the data and more comprehensive discussion amongst mods and adepts to decide if these are the variables we want to use. We'll keep you posted.

Sounds good but might I suggest.

1. 50
2. 6 months (due to that shortage of 50+ player tourneys but taking into account the changing meta)
3. Twice (prevents another Pikula fiasco)
4. Once a month. (unless something outragous happens on a given month)
5. Yes, but if a deck is a deck to beat in Europe but not played in America it should have (EU) next to it so people don't wonder why LoamTog isn't winning them all types of tourneys.

[Whit3 Ghost]

I agree with Holmes.

6 Months is about right, because 3 month updates are too frequent and don't allow for multiple tournament results.

[Zilla]

I agree with Holmes.

6 Months is about right, because 3 month updates are too frequent and don't allow for multiple tournament results.

A lot can happen in 6 months, particular considering how many more large tournaments we've been seeing in recent months. Nevertheless, it's not out of the question. However, if we do go with a 6 month timeline, we will likely increase the number of minimum placements to qualify to 3 because there will be a wider range of data to work from. This would also lower the "fluke factor" mentioned earlier.

Note that if we do go with the 3 month timeline, we would disregard months without any tournaments over 50 players. So for example, if there are results in January, but then no tournaments in February or March, then some in April, then none until July, then January, April and July would count as "the last 3 months" for data collection purposes. Essentially, for a month to be taken into consideration, it must have at least one tournament with over 50 players in it.

[Phantom]

Is their any thought to weighting the results based on the number of players entered (as well as cutting off all results with < 50 players)?

Basically, I'm asking if the winner of a GP that drew 200+ players will count more than a weekly they run in Portland?

[Zilla]

Is their any thought to weighting the results based on the number of players entered (as well as cutting off all results with < 50 players)?

No. We want to keep this relatively simple.

Basically, I'm asking if the winner of a GP that drew 200+ players will count more than a weekly they run in Portland?

I am not aware of any weekly tournaments which regularly draw more than 50 players, or even close to it. If there are any, then this number will likely be raised.

[AnwarA101]

Is their any thought to weighting the results based on the number of players entered (as well as cutting off all results with < 50 players)?

Basically, I'm asking if the winner of a GP that drew 200+ players will count more than a weekly they run in Portland?

A weekly run in Portland won't count unless it had at least 50 people, but then again you might want to check with Zilla.

[DeathwingZERO]

Just out of curiosity, but if the most recent list of possible credentials were to be instituted, wouldn't this mean at the very least Deadguy Ale was in fact still an LMF deck, for instance? It's had 4 placings in the last 6 months, almost as much as any incarnation of UGx Thresh, which is in contention along with Solidarity and Vial Goblins to be the top tier of the format.

Also, wouldn't it be a good idea to look into adding in maybe a little bit of a push in the instance of getting substantial tournaments (24 players +) on weekly/bi weekly basis as well as the obvious 50+ "big" tournaments we see maybe once a month or so? I would hardly shape the metagame of a format still in it's maturing level based on 12-13 tournaments a year, we're much more akin to Extended's various "seasonal changes" than Type 1's "New top deck" theories. It would seem if we can get people/places to give some results from regularly scheduled events in the top 8, that could also prove to be somewhat helpful in at least determining what metagame shifts certain areas are going through, even if it is a lesser turnout. I personally have noticed a huge difference between the West Coast, East Coast, and European decks, at least in the range of the top 8. This must mean that in playtesting groups and smaller tournaments, there's much more than just 3-4 decks these guys are going up against on a weekly basis.

[scrumdogg]

Sounds good but might I suggest.

1. 50
2. 6 months (due to that shortage of 50+ player tourneys but taking into account the changing meta)
3. Twice (prevents another Pikula fiasco)
4. Once a month. (unless something outragous happens on a given month)
5. Yes, but if a deck is a deck to beat in Europe but not played in America it should have (EU) next to it so people don't wonder why LoamTog isn't winning them all types of tourneys.

Agreed except for point number five - a good deck in a large tournament environment is a good deck in a large tournament environment. Besides, completely discounting play skill & familiarity with deck, most players are not going to be using the LMF data for major tournament purposes anyway. They are looking at A) what to expect (for part of a field) B) stuff that has been successful C) something they can play at Bob's Cards, BBQ, & Carwash with the other local 10-20 Legacy mutants & dominate. If Bavarian Cream Loam Control gets monkeyfucked down at Bob's, you obviously either shouldn't play it as it is bad for your local meta or practice a lot more.

[tivadar]

So does two appearances mean two appearances in DIFFERENT tourneys, or do two appearances in the same tourney count? It seems like it should be different ones because this will prevent decks that succeed in weird metas from automatically getting in should they double place in a tourney.

[scrumdogg]

So does two appearances mean two appearances in DIFFERENT tourneys, or do two appearances in the same tourney count? It seems like it should be different ones because this will prevent decks that succeed in weird metas from automatically getting in should they double place in a tourney.

Any weird meta with 50+ person tournaments deserves some respect anyway (or a roadtrip to sodomize them with good decks...). Besides, being a DTB doesn't mean you personally have to play it, like it, respect it or test it - you always have the option to ignore a deck....at your peril.

[Lego]

How close do decklists have to be for them to be considered the same deck? I gather that the two Reanimator decks which T8ed the Mana Leak were about as similar as Deadguy is to Red Death.

I mean, Goblins is Goblins whether it is mono-red or runs a green splash, whether it runs 4 maindeck Siege-Gang or maindeck Goblin King or Goblin Tinkerer. But what about when it runs 4 Ancient Tomb and 4 City of Traitors? Is it a different deck now, even if the rest of the creature base is exactly the same?

The same question can be asked for any number of decks. Have the Adepts and/or Mods been discussing this?

[Whit3 Ghost]

A lot can happen in 6 months, particular considering how many more large tournaments we've been seeing in recent months. Nevertheless, it's not out of the question. However, if we do go with a 6 month timeline, we will likely increase the number of minimum placements to qualify to 3 because there will be a wider range of data to work from. This would also lower the "fluke factor" mentioned earlier.

Note that if we do go with the 3 month timeline, we would disregard months without any tournaments over 50 players. So for example, if there are results in January, but then no tournaments in February or March, then some in April, then none until July, then January, April and July would count as "the last 3 months" for data collection purposes. Essentially, for a month to be taken into consideration, it must have at least one tournament with over 50 players in it.

That makes a lot more sense. I think either system is pretty good.

[TheInfamousBearAssassin]

This is a poor analogy in that it assumes there is a given, unchanging set of data from which to work that allows for accurate predicition of this nature. This applies to an unchanging 60 card decklist. It does not apply to a constantly changing metagame containing hundreds of different deck types. Furthermore, your analogy encourages prediction based on knowledge of statistical data, not gut feelings, which makes it a crappy way to end that sentence.

Combining some rough guidelines with subjective judgement based upon interpretation is good enough for professors, lawyers, judges, police officers, doctors, soldiers, every single driver, and pretty much any professional in any walk of life. It's an inescapable aspect of the world that we live in. Are you seriously suggesting that either the LMF is so direly important that the same standards of judgement which allow the rest of the world to get by are too imperfect for it, or that the Adepts who vote on these things are so much more inherently retarded than the average human being that their judgement is completely untrustworthy.

Can people just stop trying to use the fact that some responsibility for not being a moron is required to decide the LMF as an excuse to claim the system as flawed and revert to some blinder, also subjectively determined system?

[Finn]

Damn IBA, you certainly do have your minions. Every time you speak, some person wants to chime in to cheer.

On topic, am I understanding that you are putting limitations on what decks can be considered? It seems reasonable that the adepts can determine for themsleves far more fluidly which decks should be considered. Or are those numerical requirements which decks get the final nod? I certianly hope not for the same reasons.

There is a decision making body here. They should be allowed to make their decisions in any way they see fit. I personally would feel unnecessarily encumbered if I were among them and these sorts of quantitave shackles were placed upon my opinion.

[TheInfamousBearAssassin]

That's because I'm awesome, and a role model to asshole players who are primarily interested in making sure their opponents can't win instead of doing anything themselves everywhere.

Anyway, no, Godzilla has been advocating a system that would eliminate the conscious human decision-making element from the LMF entirely; so that, for instance, right now, new players to the format would be advised that Raffinity, Reanimator and Landstill are decks that they are highly likely to have to play against in a competitive Legacy tournament and will probably have trouble beating without a cohesive gameplan.

[Bardo]

I hope it doesn't read like a maintenance contract for a washing machine, but feel free to read our recently updated DtB criteria at TMD/TML and give us your opinion.

[AnwarA101]

Anyway, no, Godzilla has been advocating a system that would eliminate the conscious human decision-making element from the LMF entirely; so that, for instance, right now, new players to the format would be advised that Raffinity, Reanimator and Landstill are decks that they are highly likely to have to play against in a competitive Legacy tournament and will probably have trouble beating without a cohesive gameplan.

Well he just wants it to be a reflection of what is actually making Top8. These decks are supposedly the decks that you should need to beat to make Top8. If we only want a reflection of what makes Top8 then no human input is required. Make some criteria and apply it to the data and get some result. Whether that result represents whatever the idea of a "deck to beat" depends on your definition. Zilla's system seems to take the route that we should prepare for the most successful decks. But I think Zilla is predisposed to create criteria that allow for a large number (greater than 5) to be in the LMF. I'm not sure we ever answered the question - what is a deck to beat?

By the way, advocating is the wrong word you mean implementing.

[Zilla]

Combining some rough guidelines with subjective judgement based upon interpretation is good enough for professors, lawyers, judges, police officers, doctors, soldiers, every single driver, and pretty much any professional in any walk of life.

Actually no, it's not. The systems upon which judges and police officers base their profession, for example, are anything but rough guidelines. They are some of the most complex, convoluted systems ever conceived, and it takes an entirely seperate profession (lawyers) just to decypher it all. If a comparison to the American legal system as an argument for the way way we currently decide the LMF decks is the best you've got, I'm not impressed.

Are you seriously suggesting that either the LMF is so direly important that the same standards of judgement which allow the rest of the world to get by are too imperfect for it, or that the Adepts who vote on these things are so much more inherently retarded than the average human being that their judgement is completely untrustworthy.

I'm suggesting that there has thus far been no compelling argument that personal opinion is a better way to decide these things than cold hard facts.

Anyway, no, Godzilla has been advocating a system that would eliminate the conscious human decision-making element from the LMF entirely; so that, for instance, right now, new players to the format would be advised that Raffinity, Reanimator and Landstill are decks that they are highly likely to have to play against in a competitive Legacy tournament and will probably have trouble beating without a cohesive gameplan.

And with your system, we get people (I won't name any names) arguing that Truffle Shuffle belongs in the LMF. At least the archetypes listed above are well-known archetypes which have proven themselves at large, competitive tournaments.

That you're arguing against certain archetypes based on personal bias against them despite their performance only bolsters my assertion that a sytem which eliminates personal bias is preferable.

Also, as previously stated, we're still tweaking the variables involved in the way the new system will work. Once we've fleshed out the details, we'll present it to the community for review before final implementation.

I'm not sure we ever answered the question - what is a deck to beat?

The purpose of the LMF, in rough terms:

1. To give new players a foundation for understanding the format by introducing them to at least some of the relatively viable, prevalent archetypes which make up the metagame at any given time.

2. To give people who have taken a break from the format a rough idea of what is currently being played as a foundation for re-entering the format.

3. To provide people with the tools they need to create a testing gauntlet for their decks when preparing for an upcoming tournament, by providing them with a decent range of some of the more competitive, relatively prevalent archetypes to test against.

4. To provide decklists of some of the more popular archetypes for netdeckers.

5. To be used as a means of spotting changes in metagame trends for more advanced players.

By the way, advocating is the wrong word you mean implementing.

QFT.

[scrumdogg]

I just had a horrible thought, both the way we're currently running the LMF & the proposed system has no real way to address a deck that would be predominant but not really representing. Let me use a real example (fortunately covered because it IS a DTB) and one hypothetical situation. Goblins apparently was rampant at GenCon but put 1 in Top 8. If you were not able to run with Goblins, you were in for a rough tournament apparently. It gets worse if Goblins gets played everywhere but isn't cracking Top 8 (not happening, but bear with my reasoning). It IS however, in the Top 16 & constituting a significant percentage of decks at the tournaments. It gets worse if something like Fluctuator were rampant but the decks in Top 8 would be anything with blue - Thresh of all varieties, Landstill, Fish, BBS. The aggro builds you would cunningly plan against the Top 8, however, get EATEN by the 'unseen monster' in the Legacy pool. This is an extreme example, but how do we accurately show trends in Legacy that aren't well represented in Top 8 or can we even do it, beyond telling people 'you need to test against these certain decks & be able to beat randomness as well?'