Banlist Week Proposal

Banlist Week Proposal

Thunder Dome’s recent thread for his case on banning Deathscyther brought up a lot of things—past issues, questions of purpose, etc. These points, along with many others, sort of clicked into my head into this proposal. Feel free to find a better name.

Testing is dead. This is not very surprising: testing, while valuable, requires you to sit down to perform a repetitive, often monotonous task (read: stamina tests). The importance of testing has also changed over the years: where test results nearly had law status on what was good and bad at the peak of Metal Fight activity, their influence has considerably waned following the hiatus between Metal Fight and Burst. Before, hard evidence—testing—was needed to support a claim on the effectiveness of a part; today, we look to punctual observations and tournament results to support those same claims.

We could argue the merits of both approaches, we could assume that formal testing will pick back up to some degree when Hasbro releases Beyblade internationally, but that is neither here nor there. What is here is the fact that most of our information, now, comes from our competitive scene and what is there is my feeling that we should try to integrate the creative customizing you can find in a testing environment—where there are no stakes (win or lose, it’s just a test)—into the environment that currently produces the most data, i.e. tournaments, where there are stakes.

The current experimental format initiative by the committee looks to reenvision the structure of WBO tournaments and to some extent fix the somewhat shoddy balance of the Burst format; this proposal instead looks to provide a new avenue for parts testing and to enable a better portrayal of the “metagame”. Brad mentioned in Thunder Dome’s thread that banning Deathscyther would simply, and I paraphrase, steer the metagame towards the next safest thing: adding evidence to the theory (no matter how foregone the conclusion seems to be) would be one thing I think could be achieved with this proposal.

In essence, this proposal is similar to how Limited was originally imagined: with a mostly rigid monthly update. While Limited failed in that particular regard, there are a few key differences that could make it work, but first, here is what a banlist week would entail:

Every (Burst) tournament on the first week of every month would have to use the banlist: To avoid confusion with weeks that start in one month and end in another, that would be every first Monday up to and including the following Sunday. Of course, another timeframe could be used (e.g. every other month), another week, etc.

The banlist changes every month as long as one tournament was held during the Banlist Week: The goal is to promote new ideas, so as long as the banlist was used at least once, it should be changed. Changes would happen on the eighth day following the end of the week (i.e. the third Monday), leaving one week to discuss the next change and two weeks for participants and hosts to acknowledge the changes before the next Banlist Week.

Differences with Limited

Timeframe for updates: Limited is and has always been an optional format—available at all times—with no clear definition on how many tournaments is enough to determine that something needs to be changed (more on that in the next point). The Banlist Week would be mandatory (only way to opt out would be to host your tournament on another week) and the rules for when to enact changes would be perfectly clear.

Change is the end game: The scope of Limited is specific: to offer the most balanced and most open play field, where a great variety of parts can be used for different strategies, and changes to the banlist are made with finding that sweet spot in mind. For Banlist Week, each banlist can have a unique goal, from bettering regular play by removing the most powerful parts to see what happens, to trying out completely new things by removing a complete line of Drivers: A banlist can be smart, it can be crazy, but the Banlist Week itself is all about letting those changes, as ludicrous or as sensible as they are, happen. It is about offering, once a month, a new experience, something different, for better or for fun.

Of course, I invite everyone to state their opinion, offer their insight, propose the changes they feel would better the proposal. I invite the committee to offer their approval or disapproval, and most importantly, I invite hosts to share their thoughts on how they feel this would affect their workload and how easy or difficult they would find enforcing a banlist changing monthly. I was a silent disapprover of making the listing of winning combinations mandatory. I will happily say that the decision to do so turned out for the best, but my general stance that we should not impose overmuch work on hosts, who already put in a lot, voluntarily, remains. This project and its success, should it be approved, would hinge on you and your acceptance to take on extra workload during the tournaments.

Evidently, the implementation of this idea, again should it be approved or well received, could wait after the matters of the Experimental Format and the Deathscyther ban are resolved, but I do not think it would hurt to discuss. Just something I wanted to throw out there.
This is an interesting proposal. The reason why the Limited format does not experience changes as often as it may have been meant to initially, however, is that with only one event, you really cannot know if you had enough time to adapt to the change, especially in a tournament environment rather than in your home alone. For the Zero-G ban of F230CF/GCF, I also insisted that at least one region should try the temporary ban twice to be able to form a good idea of whether they like it or not. I would like to apply the same reasoning to disapprove that the banlist would change each month, not to mention that changing so often brings up the question of whether tournaments that are processed are really all that similar, forming a coherent and accurate whole in the rankings.


But yes, testing is completely dead, but nothing can rely only on tournament results though, there are way too many variables, such as parts not even being tried at all.


Statistically, from memory, few tournaments are hosted on the first week of a month, for some reason hah.
This sounds like a really nice idea! I was initially concerned that hosts may be annoyed by the altered rules for events that fall within the designated weeks (and thus may avoid scheduling then), but I think that the potential for experimentation and wacky ideas makes it very appealing. It seems like a good way to keep things fresh and give new ideas a trial run without dramatically affecting the normal series of events.
(Sep. 26, 2016  3:12 AM)Kai-V Wrote: This is an interesting proposal. The reason why the Limited format does not experience changes as often as it may have been meant to initially, however, is that with only one event, you really cannot know if you had enough time to adapt to the change, especially in a tournament environment rather than in your home alone.

No, no, I definitely agree. Limited is about opening up the play, so you need at least modicum of information before you make changes. Six months can seem long, but if you only had say two tournaments, then you haven't really got six months worth of data.

(Sep. 26, 2016  3:12 AM)Kai-V Wrote: I would like to apply the same reasoning to disapprove that the banlist would change each month, not to mention that changing so often brings up the question of whether tournaments that are processed are really all that similar, forming a coherent and accurate whole in the rankings.

But yes, testing is completely dead, but nothing can rely only on tournament results though, there are way too many variables, such as parts not even being tried at all.

I would disagree here, at least partly. I really want this to be a way to promote new, interesting scenarios. I've drawn comparison with Limited, but I actually wouldn't really want this to be just Limited 2.0: Burst Edition. Beating in a habit of change is also the best way to ensure that changes can happen on a regular basis.

You're right to say that a tournament can't completely replace formal testing and its control of variables, but a tournament does have the advantage of being a more organic experience, so I still think we could extract some valuable or interesting data from the whole thing.

As for rankings, if we were resetting it every year, I'd be with you, and I'd probably propose we do it unranked, but, as it is, I feel there's enough of a time bias in the rankings already for it to be a non issue. I put more stock in the Face/medal system.

(Sep. 26, 2016  3:12 AM)Kai-V Wrote: Statistically, from memory, few tournaments are hosted on the first week of a month, for some reason hah.

All the more reasons to implement it, if it doesn't hurt anyone!

(Sep. 26, 2016  3:14 AM)Cake Wrote: This sounds like a really nice idea! I was initially concerned that hosts may be annoyed by the altered rules for events that fall within the designated weeks (and thus may avoid scheduling then), but I think that the potential for experimentation and wacky ideas makes it very appealing. It seems like a good way to keep things fresh and give new ideas a trial run without dramatically affecting the normal series of events.

Glad you like it! The additional burden on hosts was definitely my biggest concern with this proposal.
I would totally host a tournament like this given enough time to get things rolling and make sure everyone has diverse enough parts to get a good accurate reading of the meta game. But i like this concept. It would be fun to say the least. And the data that is collected from each region would be interesting to see.
(Sep. 26, 2016  5:41 PM)Nocto Wrote:
(Sep. 26, 2016  3:12 AM)Kai-V Wrote: This is an interesting proposal. The reason why the Limited format does not experience changes as often as it may have been meant to initially, however, is that with only one event, you really cannot know if you had enough time to adapt to the change, especially in a tournament environment rather than in your home alone.

No, no, I definitely agree. Limited is about opening up the play, so you need at least modicum of information before you make changes. Six months can seem long, but if you only had say two tournaments, then you haven't really got six months worth of data.

(Sep. 26, 2016  3:12 AM)Kai-V Wrote: I would like to apply the same reasoning to disapprove that the banlist would change each month, not to mention that changing so often brings up the question of whether tournaments that are processed are really all that similar, forming a coherent and accurate whole in the rankings.

But yes, testing is completely dead, but nothing can rely only on tournament results though, there are way too many variables, such as parts not even being tried at all.

I would disagree here, at least partly. I really want this to be a way to promote new, interesting scenarios. I've drawn comparison with Limited, but I actually wouldn't really want this to be just Limited 2.0: Burst Edition. Beating in a habit of change is also the best way to ensure that changes can happen on a regular basis.

You're right to say that a tournament can't completely replace formal testing and its control of variables, but a tournament does have the advantage of being a more organic experience, so I still think we could extract some valuable or interesting data from the whole thing.

As for rankings, if we were resetting it every year, I'd be with you, and I'd probably propose we do it unranked, but, as it is, I feel there's enough of a time bias in the rankings already for it to be a non issue. I put more stock in the Face/medal system.

While I agree that "beating in a habit of change is the best way to ensure that changes happen on a regular basis" (read: play more limited, make the banlist easier to change, update it more frequently), I do think that changing it once per month is a bit too frequently. While that does allow us to correct any mistakes we make relatively quickly, I don't think it really gives us enough time ("tournament time" and real time) to accurately analyze the snapshot of the metagame, at least not with the current amount of Burst tournaments held monthly across different regions. I would prefer a bi-monthly update. As of now, the current monthly Burst tournament schedule looks something like "2 Torontos, and choose one or two of the following: North Carolina, Los Angeles, Montreal, London, and Maryland". Kai-V makes a good point about giving areas time to react and adapt to the ban, which usually spans multiple tournaments and at this point is usually only possible for Toronto, disregarding the fact that the banlist only lasts for one week.

Another issue I could potentially see is that if this idea is implemented, it would create an "unfair" playing field for most players, which is probably what Kai-V is getting at with the "coherent and accurate whole in the rankings". Say Part A is the main component of a versatile and popular "safe" combo that can beat the majority of the meta, and Part B is the main component of the combo that beats the combo Part A uses, and Part B's popularity is primarily an effect of Part A's popularity. As Part B beats Part A 1v1, we mistakenly identify it as the "problem" and temporarily ban it, unwittingly creating a horrid meta for the next experimental tournament where it's nothing but people spamming Part A. (I didn't say any names, but parts and combos definitely popped into your head, didn't they?) While keeping these experimental events unranked makes "unfair" wins and losses less consequential, I can't help but think all this experimentation going on right now is detracting from actual ranked events. This may be a thing of the past, but I recall reading on several occasions that people would not drive out for an event if it was unranked. Despite being not even a real thing, points and rank are a pretty big incentive for us "killer/conquerer" type players. I've got a bone to pick with the BeyPoint system myself, especially with regards to point inflation, but this isn't the time or the place for that and I don't understand the root of the problem well enough to suggest a solution.

As for testing vs. tournaments, I actually want to believe that the lack of testing is due to the fact that two of the most popular Layers in the game, Valkyrie and Deathscyther, wear down too quickly. Test results don't carry as much weight unless they use mint versions of these parts, and no one wants to "waste" their Valk or DS on a test when they could save it for when points are on the line. There's also the factor of some parts working well for some people and then poorly for others - Armed is my favorite example but I also made a post a while back that I'm too tired to dig up right now about the wide range of viable Limited Attack Wheels. Then there's the whole issue of testing alone in a low-pressure environment than a high-stakes tournament. Players behave differently when points are on the line, which brings back the point that I made earlier about these events being ranked or unranked: In a ranked event, players will reveal their "true" natures and use what is really good, while an unranked event will allow players to be more creative with their combos and strategies (Xcalibur Heavy Orbit), but they won't be playing at their absolute best. Not many people are going to donate their ego to science.

With all that being said, I think that this system could be implemented at some point (once we iron out Deck Rotation, there's been more experimental formats than I can count lately), but this just occurred to me: How would we decide on what gets changed each update? As of right now, it seems pretty arbitrary. And what happens when we don't feel that anything needs to be changed (if it ain't broke, don't fix it)?
(Sep. 28, 2016  9:52 AM)Wombat Wrote: While I agree that "beating in a habit of change is the best way to ensure that changes happen on a regular basis" (read: play more limited, make the banlist easier to change, update it more frequently), I do think that changing it once per month is a bit too frequently. While that does allow us to correct any mistakes we make relatively quickly, I don't think it really gives us enough time ("tournament time" and real time) to accurately analyze the snapshot of the metagame, at least not with the current amount of Burst tournaments held monthly across different regions. I would prefer a bi-monthly update. As of now, the current monthly Burst tournament schedule looks something like "2 Torontos, and choose one or two of the following: North Carolina, Los Angeles, Montreal, London, and Maryland". Kai-V makes a good point about giving areas time to react and adapt to the ban, which usually spans multiple tournaments and at this point is usually only possible for Toronto, disregarding the fact that the banlist only lasts for one week.

Having had time to sit on it, I suppose I'm not opposed to having longer periods of time so that people and communities have enough leeway to react to changes, but I still think regularity has to be paramount for it to stay interesting. To me, the data adds a real, extremely interesting value, but it's secondary to actually playing different scenarios.

Given your tournament data, your suggested update schedule and what Kai-V said about the frequency of tournaments in the first week, I think the system could stay as I first presented it, but we could adjust the value of tournaments to, say, three. At one possible week per month, updates could easily take place every two, three or even four months, which, if I understood both of you correctly, would fall within your preferences, while still maintaining a form of regularity.

(Sep. 28, 2016  9:52 AM)Wombat Wrote: Another issue I could potentially see is that if this idea is implemented, it would create an "unfair" playing field for most players, which is probably what Kai-V is getting at with the "coherent and accurate whole in the rankings". Say Part A is the main component of a versatile and popular "safe" combo that can beat the majority of the meta, and Part B is the main component of the combo that beats the combo Part A uses, and Part B's popularity is primarily an effect of Part A's popularity. As Part B beats Part A 1v1, we mistakenly identify it as the "problem" and temporarily ban it, unwittingly creating a horrid meta for the next experimental tournament where it's nothing but people spamming Part A. (I didn't say any names, but parts and combos definitely popped into your head, didn't they?) While keeping these experimental events unranked makes "unfair" wins and losses less consequential, I can't help but think all this experimentation going on right now is detracting from actual ranked events. This may be a thing of the past, but I recall reading on several occasions that people would not drive out for an event if it was unranked. Despite being not even a real thing, points and rank are a pretty big incentive for us "killer/conquerer" type players. I've got a bone to pick with the BeyPoint system myself, especially with regards to point inflation, but this isn't the time or the place for that and I don't understand the root of the problem well enough to suggest a solution.

(Sep. 28, 2016  9:52 AM)Wombat Wrote: With all that being said, I think that this system could be implemented at some point (once we iron out Deck Rotation, there's been more experimental formats than I can count lately), but this just occurred to me: How would we decide on what gets changed each update? As of right now, it seems pretty arbitrary. And what happens when we don't feel that anything needs to be changed (if it ain't broke, don't fix it)?

Since we both agree to some extent that the ranking system has some failings, but that it's still a large motivator for tournament attendance, I would propose a counterargument to unranking theses events on the basis that Beyblade is ultimately largely a rock-paper-scissors based heavily on luck and preparation. Being skilled, I would say, relies heavily on the preparation aspect, and that aspect builds primarily upon knowledge of the game. It seems to me that being able to adapt to new scenarios and win would show a better understanding of the game and level of preparation on the part of the player than being able to memorize the top-tier list (at least in a slow-moving metagame like Burst is in right now). Recently, card games have been presented as good arguments to change the way tournaments are run: I think we could draw a similar parallel in this case to the fact that available cards often rotate in these types of game.

Overall, though, it's not a point I feel strongly about, and I'd be happy if the proposal went ahead ranked or unranked. To respond to part of what you said, I'd like to emphasize the fact that it isn't my goal for this proposal to be just about banning things that are strong in regular play, but to remove parts from the game following any theme.

(Sep. 28, 2016  9:52 AM)Wombat Wrote: As for testing vs. tournaments, I actually want to believe that the lack of testing is due to the fact that two of the most popular Layers in the game, Valkyrie and Deathscyther, wear down too quickly. Test results don't carry as much weight unless they use mint versions of these parts, and no one wants to "waste" their Valk or DS on a test when they could save it for when points are on the line. There's also the factor of some parts working well for some people and then poorly for others - Armed is my favorite example but I also made a post a while back that I'm too tired to dig up right now about the wide range of viable Limited Attack Wheels. Then there's the whole issue of testing alone in a low-pressure environment than a high-stakes tournament. Players behave differently when points are on the line, which brings back the point that I made earlier about these events being ranked or unranked: In a ranked event, players will reveal their "true" natures and use what is really good, while an unranked event will allow players to be more creative with their combos and strategies (Xcalibur Heavy Orbit), but they won't be playing at their absolute best. Not many people are going to donate their ego to science.

The durability of parts probably is a big factor in the amount (or lack thereof) of testing in Burst (wasn't it 1234Beyblade who went through over 30 Valkyrie in a relatively short period of time), and I think you made a compelling case for yourself as to other reasons why the amount of testing has decreased. Add to that the fact that testing can be long and boring, and often performed alone, while tournaments can be exciting and played with other people, and you can make a fair assumption that testing is likely not coming back in full force.

As to your other point, the format is sort of designed (whether it will absolutely work is still up in the air) to allow for ban lists that force players to play differently, as opposed to going to the next best thing, as might happen in a typical long-term, game-balancing ban list.


Apologies for the late reply, and if I missed any of your talking points.