Competitive and Uncompetitive Definition Discussion

[shrang]

Warning: Super tl;dr ahead


Competitive vs "Uncompetitive" (mostly in relation to Ubers, but applicable to other metagames too)

NOTE: THIS IS A DISCUSSION ON THE COMPETITIVE/UNCOMPETITIVE, DISCUSSIONS ON BALANCE ARE NOT WHAT WE ARE DISCUSSING

As you all know, we recently concluded a Suspect Test in Ubers on Shadow Tag, which ended... controversially to say the least. Among many issues that came out of that test, one of the sticking points in the many pages of back-and-forth was the issue of the criterion in which we were testing S-tag for - "uncompetitiveness". It was originally brought up from the definition used by the OU council in the original thread discussing S-tag and Mega Gengar in Inside Scoop, and was continued to be used as the default definition in the suspect tests that followed. I, like many others, thought this definition was not only ambiguous, but also contained circular logic in it which meant it was difficult to be interpreted. I think it's time we came up with a definition that is both meaningful and applicable, especially to Ubers, which this definition will mostly be directed. This thread is an attempt to put a clearer and more objective definition to this important criterion in which we are using to ban Pokemon from our metagames.

Firstly, let's see what's good and not so good about the the current definition of uncompetitive:

Uncompetitive game aspects (or strategies) are those that take away autonomy (control of the game's events), take it out of the hand's of player's decisions--and do so to a degree that can be considered uncompetitive.

Pros:
- It defines clearly on what uncompetitive game elements do

Cons:
- It does not define to what degree of control the game's events have to be limited: This is done in a half-assed way in saying "to a degree which can be considered uncompetitive", which is just a cop-out seeing as that's just circular logic and therefore doesn't say anything at all.
- The degree of control lost is important here. We have plenty of things in the game that actually takes the control away from the player. As noted in the suspect test itself, you have things like Taunt that limit certain actions, things like Whirlwind and Roar picks randomly which mons come in and out. We even have Magnet Pull/Arena Trap that trap a smaller group of mons but is still trapping nonetheless. This is not a red herring because all of these things need to be accomodated by the definition, which is why the clause on "degree that can be considered uncompetitive" needs to be properly defined, and a pure "X limits choice/control = uncompetitive" approach does not work.

But anyway, flawed as this definition is, I think it's a good place to start.

So before we actually try to define what uncompetitive is, I think it's high time we define what "competitive" is. I haven't seen the original thread for when this definiton was put out, but out of all the discussion in the S-tag and Mega Gengar threads, I don't think I've seen people try to define what competitive or competition means, and how it relates to Pokemon and Ubers, so I think trying to prove something is the opposite is a pointless exercise.

Competition in the pure sense:

Hide

This isn't THAT important to this discussion, but I'm going to leave this in here for completeness. It is important remember that competition in its purest form does not have to be fair, balanced, or even have rules, as long the things being contested are the same in both sides. The very classic example for this is the Cold War - the USA was able to collapse the Soviet Union due to a number of factors, but one very important one is the establishment of the petrodollar which mandated that all crude oil (and therefore ALL of its derivatives, things that pervade into all forms of society) has to be paid for in US dollars. I'm not going to bore everyone with the details, if you're interested, you can look everything up yourselves. What was clear was that for the arms race that followed the establishment of the petrodollar meant the US could print as much money as it needed to fund the arms race while the USSR could not. It was like playing a game of poker where the US had an unlimited supply of chips and the USSR did not. It was still a competition for the hegemony of the world, but in no way was it fair or balanced.

Competition as applicable to Pokemon (high yield principles):
I think a good way of defining competition in relation to Pokemon would be to compare it to competitive sports. In every competitive sport, there are things in common (HALLMARKS OF COMPETITION):
1) Reciprocity/fairness: All variables of competition are kept constant EXCEPT the one in which is being tested - you are allowed to do everything that your opponent is allowed to do. An example would be the 100m sprint: the distance is the same, the track surface is the same, the air resistance is kept the same to a point where any difference is negligible, the only variable being tested is the athlete's ability to run faster than everyone else
2) There are a set of rules and boundaries in which the game operates, for example you are not allowed to use your hands in football/soccer unless you're the goalkeeper in which case you're only allowed to do so in the goal-square, the ball cannot go out of bounds, etc etc. You can also extend this to that you cannot change what the competition is primarily testing for either, for example swimsuits in swimming. There is a fine line of making a good swimsuit for a swimmer to maximise their potential vs making them in the way that the sport ends up competing who can make the most aerodynamic swimsuit than who is faster at swimming.
3) There must be a winner - any draws that are allowed to happen (for example in football/soccer) contribute to a larger competition (for example a ladder) or are solved via other events (example sudden death, penalty shoot-outs, etc)


One thing I want to make very clear now is that competition =/= fun. One might find a sport fun, but in terms of competitive sports, the majority of players do not play with the sole purpose of having fun, but as a livelihood, or if it's an amateur sport (ie not professional and making money), to be better than someone else. I'm not saying you can't have fun while you're at it, but I don't think fun is the primary motivating factor of competition. Also, fun is a subjective term. I find XY Ubers incredibly boring but others find it a very good metagame (minus the things they hate about it, eg S-tag). For the sake of having an argument that's as objective as possible, let's just leave fun out of it for the time being.

So, if we have a look at competition and therefore what is competitive and compare it to Pokemon, it actually gives us a remarkable amount of room to work with. I'm going to make the set of rules and boundaries that is available to a player purely what GF has given us before actually moving to the rules we as a community have imposed on top of that.

If we have a look at the actual game on its own without any rules that we as a community have imposed on it (ie Streetmons), it is actually quite competitive in the 3 points that's laid about above, except for a few things which we use to patch up with the current clauses that we have (except Species Clause, which is an oddball). What makes Streetmons "uncompetitive" is something we'll explore in more detail in the next section.


Uncompetitive:
So, when we use the word "uncompetitive", I think everyone can agree that it literally means the opposite of competitive, which I have laid out the hallmarks above. Anything that allows for unfair play, anything that allows you to circumvent the basic rules of the game or anything that prevents a winner from emerging would be deemed uncompetitive. There really isn't anything that does the 2nd condition. One might say S-tag's ability to prevent switching circumvents the mechanic of switching, but like it or not, it's not actually subverting any rules. Switching is something we, as competitive players, have assigned a major importance (with good reason, I admit), but technically stopping switching is not against the rules of the game, per se. Ditto to Sleep clause and the topic of preventing meaningful switches. I think it is important to distinguish between what is a "rule" and what we assign as fundamental game mechanics. The only thing that circumvents the rules of Pokemon is Endless Battle Clause, which I'd touch on next.

One of our clauses, the Endless Battle Clause, is directed at conditions (2) and (3) - that there must be a winner of the game. In the event of endless battle, there is no winner (at least through the rules and boundaries of Pokemon as we know it). The winner is decided by whoever is willing to wait until the opponent has had enough and forfeits, which circumvents how Pokemon battles should be won (having all 6 of your opponent's Pokemon killed).

This leaves with the first hallmark of competition, which is that of reciprocity and fairness. In this case, the Clauses that we have are directed at this hallmark of competition. I have gone through this many times before in the S-tag and Gengarite suspect tests. Basically, the running theme in the object of all of our Clauses (apart from Endless battle, which I just discussed) have a means to make the game unfair. That means is the RNG. A common criticism of my arguments is that I focus too much on luck and ignore other things. That may be so, but the thing is I look at the clauses and look and the running theme, and luck plays a major part in all of them, even Sleep Clause. I'm pretty sure everyone knows how stuff like Swagger/Moody/OHKO use the RNG, so I'm not going to bore everyone here. I think Sleep Clause requires more explaining.

Sleep Clause and the RNG
Sleep Clause is one of the things that I've seen people reference a lot as something that's not luck oriented, but I have argued that there are important elements of the RNG involved what I call uncontrolled sleep. Just for the record, uncontrolled Sleep would be when you have the freedom to sleep all of your opponent's Pokemon, while controlled Sleep is the variant where we have it under Sleep Clause. The unreliability of Sleep Talk is well known, as is the randomness of the Sleep counter. These two elements make the counterplay of Sleep (in most cases) RNG based. Furthermore, as was pointed out late in the Gengarite thread, that Sleep Clause came into being before Sleep Talk existed. However, even then, uncontrolled has an important RNG element - speed ties. I think this excerpt from one of my previous posts in the Gengarite thread explains:

Actually, if you look at the actual problem of sleep, you can still get it down to a problem of luck. The reason is simple actually. Even without Sleep Talk and berries and sleep-preventing abilitites (ie Gen 1, when Sleep Talk did not exist), let's just imagine if we removed Sleep Clause how we would play the game. You would either 1) wait for sleep to run out (which the length is random) and hope you don't get killed first, or 2) preemptively use sleep against your opponent by sending in the fastest sleep inducer you know exists. The first scenario is partly based on luck but I will give you the point that sleep effectively removes choice (which I'll address in a bit). However, who wins in the second scenario, assuming that both players try to do the same thing to each other (since the first scenario is obviously putting the receiver at a disadvantage)? The answer is easy - whoever wins the speed tie and whoever hits the sleep-inducing move. What is that based on? Oh wait, that's based on luck. I guess this is the sort of situation which is relatively close to that thought experiment you brought up like 15 pages back. Here, you have something that resembles something that wins the game for you automatically (it isn't, but for all intents and purposes of the argument, let's just pretend it is). I gave you my answer how I would proceed. If it was skill-based on that winning condition, then we keep it. If the two "auto-win" mons cancelled each other out, we should keep it. If it's decided by the RNG (which in this case it's based on speed ties and accuracy), then we should ban it, which is why Sleep Clause exists. So, like it or not, sleep is still very much luck-based in its very core even if you take away what we now associate with luck aspects. Add to that what we actually have now (Sleep Talk), I'd still argue that sleep is very much luck-based even you can't see it that way.

The removal of choice by uncontrolled Sleep is a consequence, a consequence, which by most voters to be too much and therefore was uncompetitive. You could say that you could run Vital Spirit/Insomnia, but the distribution of sleep users was so large that no amount of Vital Spirit/Insomnia could defeat a team of sleep inducers, which forced you to play RNG games, either through pre-emptive sleep (speed ties) or playing RNG games with Sleep Talk and the sleep counter.


Why is the RNG important?
- The RNG is inherently discriminatory. Sure, you have an equal chance of winning and losing in theory, but anyone who plays Pokemon on a regular basis knows that it doesn't always work that way. It violates the reciprocity principle above because the person who is more lucky in the game wins due to an OP RNG factor and not due to skill that they displayed. Whoever is more lucky wins at the expense of whoever is less lucky, and the RNG "chooses" a winner (I know the RNG is not a sentient being, but what happens is basically this).
- In terms of loss of control, I'm just going to rehash what I said in the S-tag test:

Anyway, I'm not here to overhype luck. I am here to give you an explanation on why luck has been a major factor (and usually deciding factor) in every ban we've had so far, and why, if you're going to even consider referring to past bans in Ubers, that you're going to have to come to terms of what made luck such an important factor in all the bans we've had. Like I said in my last post which you don't seem to have registered (of course I understand you don't even think what I have to say is worth reading any more), what makes Shadow Tag different from all the bans we've had so far is the fact that everything that Shadow Tag is based on are choices made by you and your opponent. Everything that Shadow Tag depends on are human decisions. You have intelligent, conscious choices that you make to either avoid/succumb to getting trapped by S-tag. If you lose to someone with Shadow Tag, it's not because it was something out of your control, but because you chose (whether you knew what was going to happen or not, whether you were forced to through your opponent's plays or not) to go in a sequence of events that led you to being beaten by Shadow Tag. If you want a quick summary of the last few lines, it comes down to "you were outplayed". For the most part apart from the intrinsic luck elements that we couldn't ban, if you outplayed your opponent (factoring in decisions regarding Shadow Tag), you won. The result depended on decisions made by both players. It's different from every other ban we've had (with the exception of Sleep Clause, which I've gone into in detail). For the other bans based on uncompetitiveness, be it Moody or Swagger or OHKO or Evasion in the past, you could outplay your opponent and still lose because the result depended on an external factor that is not in the control of either player. That external factor is conveniently what we refer to as luck. It has nothing to do with consistency/inconsistency. We didn't ban these things because it let you beat your opponent sometimes with luck and therefore something as consistent as S-tag deserves to banned. This is what is meant by the OU council's definition of "taking away control of the player", not whatever garbage to do with choice of switching, or what have you.

So, to what degree is uncompetitive?

I think this is the million dollar question. I think to most part, that loss of control is a good way to describe uncompetitive elements, and that loss of control in turn makes the game unfair, as I just highlighted above. Again, like I said at the start of this overly long post, many things in the game do this so therefore a distinction needs to be drawn on where we draw the line. My opinion is mostly formed by what the history of bans that we currently have, so please don't tell me off for preaching my opinion onto everyone else. Of course, I do have my own opinion which I'd appreciate people to think critically and accept if possible, but I'm just going say now that it my interpretation of what has already taken place.

So, I'm willing to keep the first part of the definition we currently have, as long as we clearly define what "removal of control" means.

As for the degree, I think it should be as follows:
- The control removed must violate one of the three hallmarks of competition
- The amount of control removed must have a direct and causative effect on the outcome of the game. Now, this might seem simple, in reality it really isn't. Something like "remove X to sweep with Y" doesn't always cut it because other things can happen. Just because removing X makes a Y sweep easier doesn't mean the actual thing itself caused it. The word causative is very important here. This is closely related with:
- The element in question must have contributed most to the victory (or loss). An example of this would be OHKO clause. Like I've mentioned previously, it wasn't the fact that something like Kyogre can luck past Latias with Sheer Cold that made me think it was too much, but the fact that it was perfectly possible for something like ScarfOgre to just spam Sheer Cold and win the game by itself if luck was on your side.
- The suspect in question cannot be prepared for with any tools used in Pokemon battling, including teambuilding.

How do they do this?
- Like I said earlier in this discussion, the RNG is a big part of this. The reason was highlighted in the passages I quoted from my posts in the suspect threads.
- In a similar vein, EITHER:
- Both players must have control removed (ie RNG-based bans), OR:
- If one player has control removed (eg Sleep/S-tag), the way that it is decided who has control removed is through the RNG (reciprocity principle) or through any means that violate principles (2) and (3) of the hallmarks of competition that I listed above.


So, tl;dr, and in a nicer format:

=====
A competitive metagame is (Hallmarks of Competition):
1) Fair: One player can do to the opponent what the opponent can do in return to them
2) Not one which has the means to subvert given rules of the game OR change the mechanics in which the game is played
3) One that must have a clear winner and loser

An uncompetitive game element, therefore, must subvert any one of the above Hallmarks of Competition. This is done via taking away autonomy (ie control of the game's mechanics) to the extent which:
- The control removed must violate one of the above three hallmarks of competition AND
- The amount of control removed must have a direct and causative effect on the outcome of the game AND
- The element in question must have contributed most to the victory (or loss) AND
- The element in question cannot be prepared for with any tools used in Pokemon battling, including teambuilding.

This is done by:
- Both players must have control removed (ie RNG-based bans), OR:
- If one player has control removed, the way that it is decided who has control removed is through the RNG (reciprocity principle, hallmark 1) OR through any means that violate principles (2) and (3) of the hallmarks of competition that is listed above.
======

I'm going to throw this out to everyone else now, I've said my piece. Feel free to point out mistakes in my lines of thinking as long as it is relevant to the argument as a whole. Also, please be civil in your discussion here. I'm aware of how dramatic these threads can become and I'm aware of what some people might think about me. This is not the place for you to vent those things.


*raises flameshield*

 

[ryan]

couldn't agree more

 

[chaos]

This is a good post shrang! ty for sharing.

I've also thought a lot about this issue, and I think you missed a fundamental facet of competition. In my opinion, the purpose of competition is to establish a relation A < B, which states that with high likelihood (L%), B should defeat A in combat. The volatility of the game is how many matches one must play to determine A < B.

The volatility of chess, or any other game with equal starting conditions and without an RNG, is 1. The volatility of rock-paper-scissors is infinity, and therefore there is no point in playing RPS competitively (EDIT: online at least, where you can just use a bot. In person may be different). At the moment, we consider the volatility of RBY OU to be 3 (SPL is bo3), and the volatility of Random Battles to be 5. All else being equal, it is preferable to have a lower volatility. There are a number of reasons why, but a particularly math-y reason is that the higher the volatility, the less matchsets will be completed, and therefore we are less able to compile statistics on the metagame. Ladder systems in particular are adversely affected.

In general we are going to want to put a threshold on volatility U, and manipulate the rules of the game such that the volatility stays below this threshold. This allows us to make best-of-U tournaments for the game. We can define an aspect of the game as uncompetitive if it increases the volatility of the game beyond that threshold--U matches would not be sufficient to determine A < B and therefore the tournament is bunk.

Now for the money shot: if likelihood L% and threshold U are determined, whether or not something is uncompetitive can be empirically tested. Ideally we would not be using vague paragraphs and what not to determine such things. It's likely that we will continue doing so because the effort of empirically testing is too high, but I hope that clarification of the fundamentals will improve how we approach these issues anyway.

How exactly you determine L% and U is beyond the scope of this post. In particular, U will depend on 1) how many matches people are willing to play, 2) external considerations like the Ladder, and 3) the "fundamental character" of the metagame. I think we can all agree that the fundamental character of Ubers is to preserve as many Pokemon and abilities as possible. Somehow I doubt people would be OK with playing 5 matches of ubers to determine who won.

Food for thought. And I guess this is a good enough post for my 10k, so wooooo.

EDIT: lots of edits for clarity

 

[MMII]

My personal favorite route is to not try to attach a definition to some term made up ages ago by the community. I'd much rather just clearly define Ubers without using those sort of words.

 

[shrang]

The problem is, that you can choose to not use buzzwords for tiering philosophy, but in the end, if you want to convey that philosophy to someone else, you will have to use language and words to do so. When that happens, you're going to have to clarify and specify every word that you use in that philosophy. Sure, the word "uncompetitive" has gotten a bad rep so far, but that's because we've been using it so flippantly with no specifics and no degree and therefore it's pretty much a cop-out. I'm using the word "uncompetitive" here at the moment because it's something we have as a starting point AND it is a buzzword that people are using regardless of whether we like it or not. If we come up with a philosophy without that word, then fine, but you still have to define every important, subjective criteria used in your new creed to the most specific and objective form possible anyway.

 

[jpw234]

I have a bunch of problems with this definition and don't think that we should be using it for policy purposes, particularly for banning. I think as laid out shrang's definition doesn't actually provide any guidance for policymakers. If we take a look at these "Hallmarks of Competition":

I think a good way of defining competition in relation to Pokemon would be to compare it to competitive sports. In every competitive sport, there are things in common (HALLMARKS OF COMPETITION):
1) Reciprocity/fairness: All variables of competition are kept constant EXCEPT the one in which is being tested - you are allowed to do everything that your opponent is allowed to do. An example would be the 100m sprint: the distance is the same, the track surface is the same, the air resistance is kept the same to a point where any difference is negligible, the only variable being tested is the athlete's ability to run faster than everyone else
2) There are a set of rules and boundaries in which the game operates, for example you are not allowed to use your hands in football/soccer unless you're the goalkeeper in which case you're only allowed to do so in the goal-square, the ball cannot go out of bounds, etc etc. You can also extend this to that you cannot change what the competition is primarily testing for either, for example swimsuits in swimming. There is a fine line of making a good swimsuit for a swimmer to maximise their potential vs making them in the way that the sport ends up competing who can make the most aerodynamic swimsuit than who is faster at swimming.
3) There must be a winner - any draws that are allowed to happen (for example in football/soccer) contribute to a larger competition (for example a ladder) or are solved via other events (example sudden death, penalty shoot-outs, etc)

The second and third ones are effectively pointless for policy purposes. The second point says "you can't break the rules of the game", and since we're constrained by the simulator we can't do that (you do point this out). The third one basically only implies the endless battle clause, which is good, but doesn't help with anything else.
Basically everything you're arguing for comes with the first point (which you call reciprocity/fairness), but you handwave way too much and stretch these terms way past their breaking point.
Reciprocity means that both players have access to the same set of options. Reciprocity as a standard can only meaningfully be upheld prior to a battle (when building teams), where it is true that every player should have access to the exact same amount of choices. But this is vacuously the case - we will always all have the same Pokemon/items/EVs available to us in the simulator (unless zarel is giving himself the capability to distribute 800 EVs per Pokemon or something). Reciprocity cannot be a meaningful standard in-battle, though - since the teams each person chooses to bring provide a different set of choices each turn. This distinction of reciprocity being upheld pre- and during-battle is borne out by the example you give (of the 100m sprint), where the infrastructure of the race (track conditions/length/air resistance/etc.) corresponds to the available teambuilding options and must be reciprocal, but the decisions in control of the competitor (the racing ability and the battling ability) are not reciprocal. So when you say things like

The RNG is inherently discriminatory. Sure, you have an equal chance of winning and losing in theory, but anyone who plays Pokemon on a regular basis knows that it doesn't always work that way. It violates the reciprocity principle above because the person who is more lucky in the game wins due to an OP RNG factor and not due to skill that they displayed

which appears to be the linchpin of your entire argument, you aren't basing it on any standards you've set out previously. RNG doesn't actually violate any "reciprocity principle" unless you're trying to say there should be "reciprocity" in a battle, but you haven't explained what that would even look like. How do users with two different teams get "reciprocal" choices in a battle? Why is RNG not reciprocal? Both players have the ability to choose to use any RNG-based move or condition.
It is unclear to me why reciprocity (as I understand it) is distinct from the 2nd point of "following the rules" (since pre-battle reciprocity is enforced by the sim), and it is unclear to me why any sort of "in-battle reciprocity" is a desirable or comprehensible standard to use.

You call your first point "reciprocity/fairness", but only define reciprocity. Then you appeal to "fairness" separate from reciprocity, which feels like a rhetorical cheat, since you haven't justified any standard of fairness at any point. Regardless:

Basically, the running theme in the object of all of our Clauses (apart from Endless battle, which I just discussed) have a means to make the game unfair. That means is the RNG.

I'm assuming that you take "unfair" = RNG. I don't think it's uncharitable of me to attribute this definition to you since you haven't explained what "unfair" means in any other terms.

You have intelligent, conscious choices that you make to either avoid/succumb to getting trapped by S-tag. If you lose to someone with Shadow Tag, it's not because it was something out of your control, but because you chose (whether you knew what was going to happen or not, whether you were forced to through your opponent's plays or not) to go in a sequence of events that led you to being beaten by Shadow Tag. If you want a quick summary of the last few lines, it comes down to "you were outplayed". For the most part apart from the intrinsic luck elements that we couldn't ban, if you outplayed your opponent (factoring in decisions regarding Shadow Tag), you won. The result depended on decisions made by both players. It's different from every other ban we've had (with the exception of Sleep Clause, which I've gone into in detail). For the other bans based on uncompetitiveness, be it Moody or Swagger or OHKO or Evasion in the past, you could outplay your opponent and still lose because the result depended on an external factor that is not in the control of either player. That external factor is conveniently what we refer to as luck. It has nothing to do with consistency/inconsistency. We didn't ban these things because it let you beat your opponent sometimes with luck and therefore something as consistent as S-tag deserves to banned.

(emphasis mine)

At this point, I don't see how your post doesn't just boils down to "hax = ban" couched in a bunch of fancy rhetoric. You now have to face down the entire barrage of complaints against this position, such as:
- What about non-100% accurate moves
- What about paralysis
- What level of hax is relevant/acceptable
- All the other already hashed-out reasons that this is an awful standard

The only time you address any of these objections is here

- The amount of control removed must have a direct and causative effect on the outcome of the game. Now, this might seem simple, in reality it really isn't. Something like "remove X to sweep with Y" doesn't always cut it because other things can happen. Just because removing X makes a Y sweep easier doesn't mean the actual thing itself caused it. The word causative is very important here. This is closely related with:
- The element in question must have contributed most to the victory (or loss). An example of this would be OHKO clause. Like I've mentioned previously, it wasn't the fact that something like Kyogre can luck past Latias with Sheer Cold that made me think it was too much, but the fact that it was perfectly possible for something like ScarfOgre to just spam Sheer Cold and win the game by itself if luck was on your side.
- The suspect in question cannot be prepared for with any tools used in Pokemon battling, including teambuilding.

where you appear to be trying to answer the question "what level of hax is relevant/acceptable" in bold. But you don't give any guidance as to what a "direct and causative effect on the outcome of the game" means. There's nothing that limits this to the clauses you've deemed acceptable. Paralysis, Hydro Pump misses, etc. can all have direct and causative effects on the outcome of games. They can also be elements that "have contributed most to the victory (or loss)".

Basically, the only relevant one of your "Hallmarks of Competition" is the first one, and the only thing I got from your post that you mean by this is that RNG-influenced mechanics are unfair. But this is a simplistic position that has been outdated for years and you don't appear to have added any nuance to it.


Rather than being entirely destructive, I would like to offer a different definition.
Here are some statements I'm presuming to come to my definition of competitiveness. I believe them to be axioms, but they are certainly up for debate.
1. A competitive game is played with the goal of determining a winner by testing the skill of the involved players.
- Not by randomness (a drawing) or arbitrary standards (your height)
2. In Pokemon, skill refers to the ability to make the best choices in team-construction and battle situations.
- Each player has the exact same breadth of choices available to them pre-battle (referring to Pokemon, items, EVs, etc.)
- Each player has the exact same breadth of choices available to them in-battle (using one of the legal moves or switching to a non-fainted Pokemon)
- Each choice is equally effective for every player (unlike in athletic events, where different competitors might execute the same choice to different results based on their strength, speed, etc.)
- Players can only differentiate themselves in terms of which choices they make
3. Having more choices better tests skill.
- Presume player A always makes the best choice, and player B selects from the available choices at random. If there are 2 choices, player A will be correct twice as often as player B. With 3 choices, 3 times as often - with N choices, N times as often. The more choices there are, the more likely player A is to be rewarded for being more skilled (which is desirable).
4. Having balanced choices better tests skill.
- If players have N choices that are each the best choice (1/N)% of the time, players will be unable to achieve good results by mindlessly selecting any choice - they will need to use skill to determine which choice best fits a given situation. If a single choice is the best a significantly higher portion of the time, players will achieve good results by constantly making that choice, which decreases the benefits of skill.

So, my definition: A "competitive" game element is one that increases the number of balanced choices available to each player. Contrastingly, an "uncompetitive" game element is one that decreases the number of balanced choices available to each player.

Some points in favor of this definition:
- It gives a clear metric by which to measure the competitiveness of game elements, which is good. For example, while appealing to RNG invites debates about whether Hydro Pump is uncompetitive, relying on the number of available choices clearly distinguishes sleep and freeze from other RNG-based elements (since they take away a player's ability to move entirely).
- It is applicable to all types of bans that we deliberate on, as opposed to perpetuating some arbitrary distinction between bans of Pokemon and other game elements. Bans of Pokemon are usually done when it is determined that a particular Pokemon is no longer a balanced choice.

 

[shrang]

I've actually addressed your definition before jpw234. I'll point out my previous post in the Gengarite thread in response to your definition (which is almost the same as the one you have here):

Before we talk about what is meaningful and not, I'm going to point out problems in your definition first. You point out that to be competitive, you must increase the number of meaningful choices (my emphasis). Does this imply that everything that doesn't increase a meaningful choice is therefore uncompetitive and should be banned? You asked this question in your previous post too, by asking what Shadow Tag adds to competitiveness. If this is true, then there will be a couple of glaring issues I would see with that:
1) Overcentralisation: Overcentralisation would result in the apparent choices you have being decreased because the metagame resolves around a couple of threats that are too powerful. Sure, you can say that people technically have the choice to run other things, but no-one making a competitive team would use them, therefore these overcentralised Pokemon would have, in your words, removed the number of "meaningful" and "balanced" and therefore uncompetitive. If you look around Ubers, pretty much every Pokemon we have meaningfully limits choice in some way such that each of them can be deemed uncompetitive through removing choice in your definition. We've specifically mentioned that we're not going to use overcentralisation as an argument for banning things, but I see that the word overcentralisation is currently being blurred with uncompetitive because of such definitions. This may be a good definition for OU, but it doesn't fit with Ubers.
2) Shit Pokemon: Technically, these Pokemon do not add any meaningful or balanced choices available to the user. No-one is going to consider Magikarp a meaningful choice for any team. If we are going by your definition, then these Pokemon are all uncompetitive. Should we ban them too? I know this is a rhetorical question, but if you're on the same track as I am, then it is obvious that there are two different types of uncompetitive, one that gives a player a significant advantage and one that is just shit, and therefore the whole problem of "choices" is a problem.

^Again, this is more applicable to Ubers, so if you want to address this for OU and below, you can make the appropriate changes. I'm aware the Mega Ray ban will have changed at how we look at balance in Ubers, but the main premise of that whole post is intact and we still have a relatively clear distinction on an "OP" ban and an "uncompetitive" ban.

Yes, "hallmarks" 2 and 3 aren't very policy relevant, but they were kind of there to highlight what is in any competitive game, not just Pokemon. If they aren't important because we have the apparatus to control them, then fine, we don't have to talk about them unless for some reason we run into a situation when they are relevant (eg let's pretend GF makes a pay-for-EV system where you can increase your Pokemon's EVs by 100 for every $10 you fork over or something evil like that).

3. Having more choices better tests skill.
- Presume player A always makes the best choice, and player B selects from the available choices at random. If there are 2 choices, player A will be correct twice as often as player B. With 3 choices, 3 times as often - with N choices, N times as often. The more choices there are, the more likely player A is to be rewarded for being more skilled (which is desirable).
4. Having balanced choices better tests skill.
- If players have N choices that are each the best choice (1/N)% of the time, players will be unable to achieve good results by mindlessly selecting any choice - they will need to use skill to determine which choice best fits a given situation. If a single choice is the best a significantly higher portion of the time, players will achieve good results by constantly making that choice, which decreases the benefits of skill.

I kind of agree with most of this, but I don't think that just because someone can limit the choice of their opponent means that the means in which they used it is suddenly uncompetitive. In any Pokemon game, you are actively working toward limiting your opponent's choices anyway. The more choices available to your opponent the worse off you are going to be. For example, you should always pick a lead that you deem to be the best to take on as many of your opponent's team as possible, or to disrupt their strategy (thus limiting their choice) whenever possible. So, therefore I still think this falls under balance. A piece of criticism of the original definition was that there was a very subjective "how much is too much" element in there. For example, Taunt limits certain choices, so do phazing moves, Stealth Rock limited the choice of using some Pokemon that are pretty good in and of themselves and would probably be good if it weren't for SR alone. Even more specific trapping abilities like Magnet Pull also limit choice but were never considered for an Ubers ban. Like I said, this isn't really a strawman argument because if we are to incorporate the notion of choice into competitiveness, you're going to have to address these anomalies. I know you put the word "balanced" in there to differentiate why certain things are bannable and certain things aren't, which is why I think this is a balance problem. Shadow Tag was suspected because it was made the choice reduction too easy, not because reduction of choice is an inherent wrong. If we bring this to something like Mega Rayquaza, it was banned because it was way too powerful and made sweeping too easy, not because sweeping itself is an inherent problem in itself. On the other hand, the RNG elements that we've banned so far, IMO are inherently problems in themselves (for the reasons I have outlined in my massive OP).


Just to set some things straight though:

At this point, I don't see how your post doesn't just boils down to "hax = ban" couched in a bunch of fancy rhetoric. You now have to face down the entire barrage of complaints against this position, such as:
- What about non-100% accurate moves - This is a bit different because you've chosen to use those moves yourself. I did address that IMO uncompetitive game elements take control away from both players. If you willingly give control of the game away and lose, that's your own problem.
- What about paralysis - I'd be pretty happy to ban T-wave and Glare and stuff like that, tbh
- What level of hax is relevant/acceptable - With the exception of the non-100% accurate moves, I think if you look at all the things we've banned so far, we've been actively minimising every bit of hax that we've been able to ban without changing the cartridge mechanics. If there was a legit cartridge of banning the 10% Ice Beam freeze chance, I reckon we would have done it by now. We've even banned things like Brightpowder (in gen 5 anyway), which shows that the tolerated level of hax is pretty low. This is why, when I look at our banning history, I think a T-wave ban or something like that would actually be appropriate.

 

[lilyhollow]

tldr - policy decisions usually happen for practicality reasons stemming from overwhelming community demand, with 'competitiveness' rationalizations largely happening in hindsight. i think these rationalizations, and therefore a formal definition of 'competitiveness,' are unnecessary. i also think redefining competitiveness is too much trouble in the first place, because it's a term that is already well in use.

i put the tldr first because this really got too long!

-------------------------------------------------------

I don't think it's appropriate to use the term "competitiveness" here. As mentioned before, yes, it is a buzzword. But more importantly, it's a term that's understood by many people in many different communities to mean all sorts of things, most of which don't have anything to do with the definitions being thrown around in this thread. There's a lot of baggage associated with that word, so giving it some formal definition that probably goes against all of that is an uphill battle. You're going to be fighting the expectations of certainly anyone who comes from a non-Pokemon competitive community, and probably most Pokemon players who aren't plugged into the nuances of Smogon policy at that.

Here are some examples of what I'm talking about

re: "There must be a winner" - Chess is one of the most respected competitive games in history. In top level play, draws happen more than 60% of the time. Seemingly most chess players are fine with this. Certainly it would be considered weird to most of them if you called Chess 'uncompetitive.' Probably you'd use a different word like 'boring' or something?

re: RNG - Poker is hugely, massively competitive. There are plenty of examples of rng-heavy games that are super successful and unquestionably considered 'competitive' by their playerbase. So, again, to most people serious about those games, it would be sort of meaningless to them if you told them that their games are 'uncompetitive.' Competitiveness already means something way way different to them than what we're trying to define here, so it's probably best we chose a different word.

re: Having more choices better tests skill - In every fighting game ever made, a large portion of the game is "try to put my opponent in a position where they are as helpless as possible and then capitalize on that. Sometimes, completely helpless, with literally zero counterplay available to them whatsoever." Actually, most other games are like this (Chess endgames, but I think basically every competitive game ever made also?). I... agree that it's good to give both players choices during a game of course! But at the same time it would be really weird, I think, to say "okay so an uncompetitive game element is one that restricts the choices of players. So for example, check in Chess, knockdowns in fighting games, destroying buildings in RTS games." It's true that those are all examples of a limitation on player choice, but the word 'uncompetitive' does not really feel useful when those features are basically at the core of their respective highly competitive games.

In general I think having a formalized definition for a thing that already means many different things to many different people is just not going to gain much traction. Jargon is already troublesome as it is, but jargon that redefines terms that are already in use is really a hassle.



I have another small issue with this 'having more choices better tests skill' thing, I think it's extremely common in this community and of course one of the classic arguments against Shadow Tag, or more commonly, 'overcentralization' in general.

I think it's more accurate to say 'having more choices tests more skill,' or something. Like, imagine a Pokemon game that has 10,000 Pokemon and each of them can carry 20 moves and you can use 10 pokemon on your team. Okay yes it would probably be true to say that you're doing more skill-testing there in the long run. I have to know what all these moves/Pokemon do and the team combinations are absurdly more complex, so the amount of skill it would take to even begin mastering that system would probably be way way higher. But would that new version of Pokemon be better at testing skill, for the purposes of actual human beings? Probably not, because many of the skills it would be testing would be cumbersome and boring. What's more, the game we have right now is already very good at testing skills like probability management, valuation, memorization, understanding of synergy, yomi, and so on. Adding thousands of new Pokemon and teambuilding options would improve our ability to test none of those skills, except maybe valuation and definitely a ton of memorization, which I don't think anyone even actually wants to test that much in the first place.

In other words, testing 'more' skill =/= being better at testing the skills we would like to test.

Whether either of those should be a part of some formalized definition of 'competitiveness,' I cannot say, primarily because I am not interested in a formalized definition of competitiveness.

Bringing this back to Shadow Tag, "Shadow Tag ban = more options for the player = more skill being tested" was a big meme that was always floatin around, certainly around gen 4 Wobbuffet when I was actually around to witness it. That's how a lot of people justified their desire for a Wobbuffet ban, after they had already pretty much decided that they hated Wobbuffet for being an annoying piece of shit.

Here is my problem with that rationalization (and basically my problem with any invocation of the idea of 'competitiveness' in a Pokemon policy discussion): the game we'd have before a Shadow Tag ban would almost certainly be super ultra omega good at testing all sorts of skills. To say that getting rid of Shadow Tag makes the game more 'skill-testing' because players have 'more options' is probably true! But, it's not a very important observation if our goal is to improve the health of the game or its community. You could also say that "if we made Chess players armwrestle eachother for an extra move after every capture in the endgame, the game would be more skill-testing." That would be equally true, and it would be equally unimportant, because the chess community has long understood the fact that chess is well past the point where it's "skill-testing enough" for their purposes. Pokemon, like chess, is way past the point where it needs any more skill-testing than it already has.

Arguably testing a little bit 'less' skill (by, say, allowing Shadow Tag) is only important if the players of the game are bumping up hard against the overall skill ceiling of the game. Adding 'more choices/skill' to Pokemon is like adding a packet of salt to a cup of chicken-flavored instant ramen or something.

So I think that the reason the Shadow Tag ban is good (or if you think T-wave or RBY Freezes or Greninja should be banned, the argument would be the same) has very little to do with "it allows us to test a little more skill in this extremely rich, high-skill-ceiling game, a game where nobody has even come close, really, to touching that ceiling." Who cares if we can raise the skill ceiling of a game where we can't even see the skill ceiling. What this community recognized about Shadow Tag is simply that most players completely despise its dynamics and feel like absolute shit playing against it. It's the same for many RNG elements like Bright Powder, which were easy to remove and very inconvenient for the short-set format that was and continues to be the tournament standard. And... I think that's enough! I don't really think we need to layer on this 'it's uncompetitive' rationale after the fact, particularly when we kind of don't even know how to define that meme without stepping on the toes of a million people who already have their own ideas of what is or isn't 'competitive.'

Sorry for super ramblyness but hey it's been years, I owed this place one