Rethinking the tournament legal counter-picks?

[Bana]

I was recently looking at the history of stage legality in Melee. We have five neutral stages: Battlefield, Final Destination, Yoshi's Island, Dreamland 64, and Fountain of Dreams. Counter-picks have slowly been banned over time, and all that remains is Pokemon Stadium.

This makes me feel like the current rule set has unfortunately, unintentionally become tailored to favoring certain characters. Irregardless of what stages are legal, Fox and Falco would probably top the tier list. However, I can't help but feel that the current stages are only encouraging 20XX. Why is the only legal counterpick a stage that favors spacies?

Lots of stages have been eliminated for good reason. Moving stages, stages with walk-offs, stages that allow circle camping, stages with walls (that allow for infinite combos), and stages that allow floaty camping (Kongo Jungle 64) all make sense to remove. But there are a few stages that I think may not have an incredibly strong reason to have banned.

The inclusion of Pokemon Stadium already implies that we can accept stages that favor certain match-ups, and that we can accept stages that are somewhat unconventional (regarding the stage transformations, some of which include walls and places for infinites or camping). Brinstar and Mute City are two stages that come to mind, but even other past stages should (I think) be reconsidered.

I'm just saying that I think we should reopen the discussion for bringing back some counter-pick stages, and I think it has strong potential to affect the meta in a positive way. I'm relatively new to the scene, so I don't have a good handle on the history of banned stages. Please let me know if there are convincingly strong reasons that the current set of legal stages should be the final stage set.

 

[Bones0]

This topic has been discussed in great length in the official ruleset thread, so I'd recommend just reading that if you want to see the overall opinion on the topic. It also gets brought up about once a month so it's certainly not a new thing. Many people try to argue the same points you do, but here is a breakdown of why your logic is flawed:

1. Pokemon Stadium used to be a neutral. The only reason it was relegated to the counterpick list was because striking was introduced as an alternative to random select for game 1, and you cannot strike from 6 stages. If FD was originally chosen as the lone "counterpick" instead of PS, there's no doubt that people would be complaining that the spacies' worst stage is excluded from striking...

2. Related to #1, PS being excluded from the striking process is a disadvantage to characters that excel on it.

3. PS being good for Fox and Falco is an opinion, not a fact. Personally, I think most floaties do better overall on PS than the 3-plats since PS is effectively an FD with 2 side plats very low to the ground.

4. Even if the vast majority of people believe PS is good for spacies, that does not change what the metagame might be in the future. Stages have always evolved to suit different characters over the years. Falco used to be considered an amazing character on FD because of his laser control, especially vs. Fox. These days, Falcos are constantly striking and banning FD as it is usually their last choice of stage vs. most of the cast. If people think their character is bad on a stage, they should try to develop strategies that enhance their performance on the stage, not try to get it banned or use it as leverage to get better stages added back into the ruleset. Simply looking objectively at the dimensions of PS, it's clear that the stage is not some radical force that skews matchups. If any stage is guilty of that, it's FD.

5. Even if everyone agreed spacies are dominant on PS and agreed that will never change, that STILL doesn't mean anything because you shouldn't be selecting stages to balance matchups. It's comparable to disallowing bans vs. low tiers to make the game more fair. Should an opponent playing vs. Kirby really need a ban to win? It's irrelevant because as soon as you start changing the ruleset to artificially buff or nerf certain characters, the game loses its integrity. Anyone who ever wins with Kirby wouldn't have really earned the win because the ruleset basically gave them a handicap. You should look at the attributes of each stage and ask yourself if it is conducive to competitive play regardless of which characters do good or bad on it.

 

[1MachGO]

It's irrelevant because as soon as you start changing the ruleset to artificially buff or nerf certain characters, the game loses its integrity.

This isn't to say that I disagree with all of your overarching point(s), but I disagree with this statement on the grounds that the ruleset itself is entirely arbitrary. Selecting a "legal" stage list, 4 stocks, and an 8 minute timer is entirely of community design and the metagame has revolved around these (objectively speaking) arbitrary rules.

Technically speaking, the most fair way to conduct matches in our current ruleset would be to do a Bo3 on every single legal stage and then the aggregate results would determine the winner. Since this isn't realistically achievable, the next best thing would be to permanently ban every single stage and only make one stage legal (either battlefield or FoD). IIRC, the 64 community had already done this. A stage list, in and of itself, is so fundamentally flawed that tradition is the only reason I can think of for why it still exists.

 

[Bones0]

This isn't to say that I disagree with all of your overarching point(s), but I disagree with this statement on the grounds that the ruleset itself is entirely arbitrary. Selecting a "legal" stage list, 4 stocks, and an 8 minute timer is entirely of community design and the metagame has revolved around these (objectively speaking) arbitrary rules.

Technically speaking, the most fair way to conduct matches in our current ruleset would be to do a Bo3 on every single legal stage and then the aggregate results would determine the winner. Since this isn't realistically achievable, the next best thing would be to permanently ban every single stage and only make one stage legal (either battlefield or FoD). IIRC, the 64 community had already done this. A stage list, in and of itself, is so fundamentally flawed that tradition is the only reason I can think of for why it still exists.

The ruleset is arbitrary, but none of the other aspects of the list (timer, stock count, games per set, etc.) were based around balancing characters. Why would we determine the legality of stages any differently?

 

[1MachGO]

The ruleset is arbitrary, but none of the other aspects of the list (timer, stock count, games per set, etc.) were based around balancing characters. Why would we determine the legality of stages any differently?

If the ruleset isn't supposed to balance characters, we should remove bans from Bo3's and DSR/MDSR entirely. While the initial reasoning for Bo3's in lower bracket is to save time, the additional rule of bans purely exists to balance MUs. While you could argue that Bo3's have bans due to increased match significance over Bo5's, you'd still be acknowledging the fact that stages should be soft-banned to prevent MU imbalance. Furthermore, mid and low level players are playing a completely different game than top level players (the most pathetic aspect of our ruleset, IMO). The same concept applies to DSR and Modified DSR. We are soft banning at the micro level for the sake of character balance.

Its also worth noting that bans in Bo3s and DSR/MDSR in Bo5s favor spacies more than any other character. They already outplace just about every character in the game on average, and while this has a lot to do with their popularity, I guarantee their placings would drop at least 15-20% if the aforementioned rules were removed. That, or every space animal player below Mango/PP level would pick up Sheik/Falcon to deal with the FD counterpicks up the ***.

Regardless, the current ruleset is incredibly dated.

 

[Bana]

Let's accept your first point that stages should not be selected around certain match-ups. My secondary point in originally bringing Pokemon Stadium up was to show that we still allow a stage that is pretty unconventional. PS is not static. PS has transformations with walls that allow for infinite combo's. PS has transformations that greatly give advantage to certain characters. PS has transformations that sometimes force players to camp.

If you want to have all stages be static and fitting certain conventions, fine, but the very fact that PS is legal shows that this is not the case. Why is PS allowed? PS is fun. PS is unconventional, yes, but what it brings to competitive play is fun and exciting. I would argue that it's NOT the best choice for a purely competitive game, but I would argue that the competitive scene benefits from it being legal. Perhaps the same could be said of Randall or the fly guys on Yoshi's Story.

Why are Mute City and Brinstar banned?

 

[The Soap]

Mute City- Very small stage, no grababble ledges, car hazards

Brinstar- Acid, can break main platform apart

 

[Bones0]

If the ruleset isn't supposed to balance characters, we should remove bans from Bo3's and DSR/MDSR entirely. While the initial reasoning for Bo3's in lower bracket is to save time, the additional rule of bans purely exists to balance MUs. While you could argue that Bo3's have bans due to increased match significance over Bo5's, you'd still be acknowledging the fact that stages should be soft-banned to prevent MU imbalance. Furthermore, mid and low level players are playing a completely different game than top level players (the most pathetic aspect of our ruleset, IMO). The same concept applies to DSR and Modified DSR. We are soft banning at the micro level for the sake of character balance.

Bo3s DO have bans due to increased match significance, but that doesn't mean I acknowledge stages should be banned from the RULESET. There's a huge difference between determining a stage's legality and determining the stage selection process. Bo3s with bans is no more an attempt to balance matchups than bo3s without bans. Which character benefits will vary from matchup to matchup, but if you just base your decision on ditto matchups, it's pretty easy to see why bans in bo3s is superior to no bans.

It is unfortunate that not all players can play bo5s, but that is simply logistics. While it technically becomes a "different game" when using bo3s, that is true for all games. Even if a set is all Yoshi's Story, the fact that players have less time to adapt can make a set go very differently than if it were played as a bo5. This is exactly why I've made the suggestion that TOs implement flex sets to minimize the disparity between bo3s and bo5s (see sig). Even with this understanding, idk what you are proposing. Whether you have bans or no bans doesn't really help to mitigate this effect, and none of this is at all related to the original point concerning balancing the cast by adding or removing stages from the stage list. You said earlier we should either have all of the stages or just a single one, but you fail to realize that picking a single stage is just as arbitrary as picking a list of 6. All I'm saying is that since we're arbitrarily selecting stages, we should do our best to remove bias towards or against specific characters.

Its also worth noting that bans in Bo3s and DSR/MDSR in Bo5s favor spacies more than any other character. They already outplace just about every character in the game on average, and while this has a lot to do with their popularity, I guarantee their placings would drop at least 15-20% if the aforementioned rules were removed. That, or every space animal player below Mango/PP level would pick up Sheik/Falcon to deal with the FD counterpicks up the ***.

[citation needed]

Let's accept your first point that stages should not be selected around certain match-ups. My secondary point in originally bringing Pokemon Stadium up was to show that we still allow a stage that is pretty unconventional. PS is not static. PS has transformations with walls that allow for infinite combo's. PS has transformations that greatly give advantage to certain characters. PS has transformations that sometimes force players to camp.

If you want to have all stages be static and fitting certain conventions, fine, but the very fact that PS is legal shows that this is not the case. Why is PS allowed? PS is fun. PS is unconventional, yes, but what it brings to competitive play is fun and exciting. I would argue that it's NOT the best choice for a purely competitive game, but I would argue that the competitive scene benefits from it being legal. Perhaps the same could be said of Randall or the fly guys on Yoshi's Story.

Why are Mute City and Brinstar banned?

Personally, I draw the line at stage hazards. I don't want to play on a stage that can do damage to a player. If you draw the line somewhere else, that's fine, but the entirety of your original post said very little about the benefits of including counterpicks like Brinstar and Mute City in the stage list. Instead, you simply rationalized that PS was good for certain characters, therefore other characters deserve a stage of equal strength.

I always tell people the same thing when they want more stages: If you want a stage unbanned, you should be willing to keep it on your random select. The 6 stages that are currently legal are the same 6 stages that people have kept on random select for years. The rest were sort of just stages that people played on in tournament because they could cheese or abuse certain aspects of the stage. There were people who practiced specifically for their counterpicks, but even they were few and far between. I think most people would change their mind about including a stage like Brinstar or Mute City if they consistently played on it all the time.

 

[1MachGO]

Bo3s DO have bans due to increased match significance, but that doesn't mean I acknowledge stages should be banned from the RULESET. There's a huge difference between determining a stage's legality and determining the stage selection process. Bo3s with bans is no more an attempt to balance matchups than bo3s without bans. Which character benefits will vary from matchup to matchup, but if you just base your decision on ditto matchups, it's pretty easy to see why bans in bo3s is superior to no bans.

It is unfortunate that not all players can play bo5s, but that is simply logistics. While it technically becomes a "different game" when using bo3s, that is true for all games. Even if a set is all Yoshi's Story, the fact that players have less time to adapt can make a set go very differently than if it were played as a bo5. This is exactly why I've made the suggestion that TOs implement flex sets to minimize the disparity between bo3s and bo5s (see sig). Even with this understanding, idk what you are proposing. Whether you have bans or no bans doesn't really help to mitigate this effect, and none of this is at all related to the original point concerning balancing the cast by adding or removing stages from the stage list. You said earlier we should either have all of the stages or just a single one, but you fail to realize that picking a single stage is just as arbitrary as picking a list of 6. All I'm saying is that since we're arbitrarily selecting stages, we should do our best to remove bias towards or against specific characters.

I numbered things so its a bit more organized

1. Well, for starters, I am not really trying to propose anything. I am just outlining some problems with our ruleset and how we do (though indirectly) attempt to balance the game with it.

2. In regards to the whole banning thing, I specifically said you'd be acknowledging soft-bans to balance MUs in Bo3s. If this doesn't make sense, think of it like this: Is FD a legal CP in Fox/Marth MU during Bo3s? The statistical likelihood of FD being played in this MU is probably so low, that if you randomly decided to put in the ruleset: "Any Bo3 MU involving Marth and Fox cannot go to FD" it would essentially have the same effect as the current meta trend.

The thing to keep in mind is that since matches are more significant in Bo3s, so are stages and the polarity they bring to certain MUs. By allowing the players to micro-ban, we are acknowledging the polarity of stages themselves and, by extension, preventing the really ugly polarization from occurring in Bo3s; thus, balancing certain MUs that would be skewed without the rule.

Another idea to consider is this: If CP banning and DSR/MDSR don't "pseudo- balance" MUs at their respective match lengths, why can't we remove these rules? Particularly DSR/MDSR?

3. To address what you are saying about 1 stage being more arbitrary than 6, I strongly disagree. The way the CP system works is that the winner of the first match is in the most favorable position. In theory, whoever wins the first match between two similarly skilled players should also win the set because the last match will be played on their CP. If there is only 1 stage to play on, this problem is negated.

It is really no different than playing on an asymmetrical map in a competitive FPS. Lets say side A has a 60% winrate over side B. This means that whichever team wins the RPS/coin toss to Side A is more likely to win the set because they'll be playing the last match on that side.

Consistency and symmetry are the most ideal conditions for competitive play.

4. I like your intent with flex sets, but it doesn't necessarily solve the logistical problem associated with Bo5s. It all depends on the likelihood of two similarly skilled players meeting. Have you gathered any stats on how often matches go to game 3?

[citation needed]

So lets say that Fox and Falco are hard-countered by Marth on FD. If a Marth goes up against a Fox/Falco of equal skill, all they have to do is win game one and they've (theoretically) won the set. In Bo5's, the Marth can CP to FD twice and in Bo3's, the Fox/Falco player can't ban FD.

Since DSR/MDSR and Bo3's with bans have been in effect for ages now, I obviously can't cite real life examples. However, I feel like this can be reasoned with the above logic. An FD counterpick against Fox and Falco is probably one of, if not THE hardest counterpick in the game for certain characters (namely Marth and Pikachu, but it also applies to the Mario bros, Peach, and a few others).

 

[Bones0]

2. I feel like you're confusing two totally different concepts when it comes to balancing. There is balance in the sense of "character A should be given buffs through the ruleset so he can better compete with character B". That is a distinctly separate concept from introducing bans or striking to select stages as fairly as possible. If you have a Fox vs. Marth set, you may consider Fox to gain more from adding bans than Marth does, but ultimately that is not why the bans were added. In Fox vs. Falco, I'd consider having bans to be more in Falco's favor than Fox's because of FD. These are examples of how bans influence the balance of matchups, but every alteration you make to a ruleset will inevitably change the balance of matchups because you are effectively playing a different game. Since every ruleset influences the balance of matchups, how are we supposed to pick a ruleset? In my mind, we should choose a ruleset independent of character balance, and let the chips fall where they may.

3. Your theoretical situation totally ignores the possibility that the player who lost game 1 can win on his opponent's counterpick... Varying skills on different stages is the whole reason we have developed a ruleset that encourages as much stage variety as possible. Your FPS example is flawed because it essentially boils down to winning a totally random coin toss. There are no coin tosses when both players get a ban and counterpick; it's perfectly symmetrical. Even when players fail to win on their opponent's counterpick, the root cause of their loss can simply be traced back to game 1 and they'd have no reason to complain. If you are of the belief that players should simply strike for the stage every game, that's fine, but most people would just take issue with the skill and depth lost in requiring good play on all of the stages.

4. Why can't spacies ban FD in a bo3? I agree being able to counterpick FD twice is a dumb side effect of our current ruleset (one that is fixed in my ruleset by having bans for bo5s, but requiring players to change their ban after their second win). I agree FD is a really strong counterpick vs. spacies, but I don't understand your point. You want to get rid of bans so spacies are worse? The whole idea behind bans is to prevent the more extreme stages from appearing in shorter sets. That's not an attempt to nerf or buff any specific character, and like I described in my initial post, these things are totally opinionated and subject to change over time. Marth mains can complain about FD being banned all the time, but if we remove bans then there may come a day where they all complain about having to play on DL. Again, this is why I suggest choosing a ruleset based on ditto matchups where you aren't concerned with balancing specific characters.

 

[1MachGO]

2. I feel like you're confusing two totally different concepts when it comes to balancing. There is balance in the sense of "character A should be given buffs through the ruleset so he can better compete with character B". That is a distinctly separate concept from introducing bans or striking to select stages as fairly as possible. If you have a Fox vs. Marth set, you may consider Fox to gain more from adding bans than Marth does, but ultimately that is not why the bans were added. In Fox vs. Falco, I'd consider having bans to be more in Falco's favor than Fox's because of FD. These are examples of how bans influence the balance of matchups, but every alteration you make to a ruleset will inevitably change the balance of matchups because you are effectively playing a different game. Since every ruleset influences the balance of matchups, how are we supposed to pick a ruleset? In my mind, we should choose a ruleset independent of character balance, and let the chips fall where they may.

While the ruleset isn't directly buffing characters, it indirectly does as you have pointed out. Removing bans from Bo3s and MDSR/DSR would technically make the game more "vanilla."

3. Your theoretical situation totally ignores the possibility that the player who lost game 1 can win on his opponent's counterpick... Varying skills on different stages is the whole reason we have developed a ruleset that encourages as much stage variety as possible. Your FPS example is flawed because it essentially boils down to winning a totally random coin toss. There are no coin tosses when both players get a ban and counterpick; it's perfectly symmetrical. Even when players fail to win on their opponent's counterpick, the root cause of their loss can simply be traced back to game 1 and they'd have no reason to complain. If you are of the belief that players should simply strike for the stage every game, that's fine, but most people would just take issue with the skill and depth lost in requiring good play on all of the stages.

I specified that this problem applies to two similar skilled players. If two players are of equal skill then winning on a neutral stage is theoretically the equivalent of a coin toss.

And I am not saying that I don't personally want stage variety (I think it compounds a lot of strategy in the game) but I'll admit that having consistency and neutrality is objectively "better".

However, I think I have devised a solution for the aforementioned CPing problem. I'll be posting a thread about it soon after some research.

4. Why can't spacies ban FD in a bo3? I agree being able to counterpick FD twice is a dumb side effect of our current ruleset (one that is fixed in my ruleset by having bans for bo5s, but requiring players to change their ban after their second win). I agree FD is a really strong counterpick vs. spacies, but I don't understand your point. You want to get rid of bans so spacies are worse? The whole idea behind bans is to prevent the more extreme stages from appearing in shorter sets. That's not an attempt to nerf or buff any specific character, and like I described in my initial post, these things are totally opinionated and subject to change over time. Marth mains can complain about FD being banned all the time, but if we remove bans then there may come a day where they all complain about having to play on DL. Again, this is why I suggest choosing a ruleset based on ditto matchups where you aren't concerned with balancing specific characters.

Again, I am not trying to propose any solutions or "nerf" spacies; all I am doing is trying to show some flaws in our ruleset and how it affects different characters. If banning and DSR/MDSR removed, then spacies, Falcon, Puff, and Samus would be worse. Marth/Sheik would swap with spacies on the tier list, and Pikachu, Peach, and ICs would be slightly better. I couldn't care less which meta we had.

As a side nitpick, you honestly can't thing DL is as polarizing as FD; especially since most of our legal stages have the "battlefield" platform arrangement.

 

[Bones0]

I'm not sure how no bans is more vanilla than bans. Using either as a means of selecting stages is totally arbitrary. Even if you are right, enabling items on neutral and playing 2-minute time matches would also make the game more vanilla. I don't see any reason to consider "vanilla"-ness more superior to our current ruleset so long as we don't actively try to balance the cast based on popular opinion.

LOL Two evenly matched players playing on a struck stage is not a coin flip. The winner will be determined based on whoever plays better. Just because they each tend to have a 50% chance of winning any one game doesn't mean the winner is random like a coin flip. Saying any ruleset is objectively more competitive or better at determining skill is ridiculous because skill is circularly defined by the ruleset you are playing in. Cactuar Stamina Pancake isn't any more or less "better" than the regular ruleset. It's just different, and the majority prefers to compete seriously on the latter.

You keep discussing the differences in the metagame when bans are removed and I've never denied them, but I don't know how you can consider it a flaw. Every ruleset you could possibly conceive of will benefit some characters and hurt others. How is our ruleset flawed when being used for two players character locked on Mario?

Just because DL has 3 platforms doesn't mean it can't one day become more powerful of a counterpick for spacies than FD currently is for Marth. If you had a stage the size of Hyrule Temple that was also a 3-plat, that would still be by far the most polarizing stage, so clearly layout similarity doesn't mean everything when it comes to which counterpicks are best. Even if DL is never viewed as better for spacies than FD is for Marth, it doesn't negate my point which is that the strengths of counterpicks are just as flexible as the rest of the metagame, and when you consider these are all opinions in the first place (filled with bias as we are all human), it seems pretty dumb to try to balance the cast with stages instead of just picking stages based on neutral criteria.

 

[flwns]

I have not read the the entirety of the book that is this thread.. but I just want to say that I think if the 20XX hack pack was made the de facto version of the game for competitive play, it would really expand the list of viable stages and the options available at the competitive level.

Stages in the pack like, no transforms pokemon, cleaned up green greens, cleaned up kongo jungle, mini hyrule, smashville fourside, and now in 3.0 there is even battlfield where you can change the platform heights. These are all very viable stages and I think they would add a lot of freshness into the competitive environment.

I'm rather new to the scene and I'm just kind of baffled that we have this great community project that is used commonly outside of tournaments when it could really expand the options at the competitive level.

 

[Dragoomba]

Stages in the pack like, no transforms pokemon, cleaned up green greens, cleaned up kongo jungle, mini hyrule, smashville fourside, and now in 3.0 there is even battlfield where you can change the platform heights. These are all very viable stages and I think they would add a lot of freshness into the competitive environment.

Cleaned up Green Greens is my personal favorite out of these. I'd always wanted to see an experimental sort of tournament (or at least a side tournament) with these stages made legal just to see how things would play out. I'm not sure about it becoming the tournament standard though, due to accessibility. There's also the problem of the large scale tournaments like MLG and EVO that legally cannot feature a modified version of the game. If this were the tournament standard for grassroots events, it would be problematic that the tournament standard wouldn't be able to be played at EVO and such.

 

[flwns]

Cleaned up Green Greens is my personal favorite out of these. I'd always wanted to see an experimental sort of tournament (or at least a side tournament) with these stages made legal just to see how things would play out. I'm not sure about it becoming the tournament standard though, due to accessibility. There's also the problem of the large scale tournaments like MLG and EVO that legally cannot feature a modified version of the game. If this were the tournament standard for grassroots events, it would be problematic that the tournament standard wouldn't be able to be played at EVO and such.

I do agree it would be fun to see some events host the hack pack stages.

Regarding accessibility, modding the wii software to run the hack pack off an SD card, while not requiring a wiimote is pretty straightforward once know how to do it.

I was unaware about possible conflicts regarding legalities of transitioning the modded game from smaller events to those large events you mentioned. Good to know.

 

[1MachGO]

I'm not sure how no bans is more vanilla than bans. Using either as a means of selecting stages is totally arbitrary. Even if you are right, enabling items on neutral and playing 2-minute time matches would also make the game more vanilla. I don't see any reason to consider "vanilla"-ness more superior to our current ruleset so long as we don't actively try to balance the cast based on popular opinion.

I find your claim that 2-minute matches with items being more vanilla is non sequitur since we are speaking in the context of competitive play. Items introduce randomness and aren't conducive to competitive play.

Saying that both options are arbitrary comes off as special pleading. The ruleset, in and of itself, is arbitrary, but your belief is that we shouldn't have any rules that try to balance the cast. Bans and DSR/MDSR actively prevent players from using the most optimal strategies thy can against certain characters; pseudo-buffing said characters as a result. Marth/Pikachu not being allowed to go to FD whenever the lose a game to Fox/Falco can loosely be compared to Wobbling bans at some tournaments. The ruleset is nerfing them to some degree.

LOL Two evenly matched players playing on a struck stage is not a coin flip. The winner will be determined based on whoever plays better. Just because they each tend to have a 50% chance of winning any one game doesn't mean the winner is random like a coin flip. Saying any ruleset is objectively more competitive or better at determining skill is ridiculous because skill is circularly defined by the ruleset you are playing in. Cactuar Stamina Pancake isn't any more or less "better" than the regular ruleset. It's just different, and the majority prefers to compete seriously on the latter.

You are getting far too technical about it. You might as well argue that a coin flip isn't truly random because its ultimately determined by the weight of coin, the amount of force being applied to it, and the height its being dropped from. The actual elements which create the outcome isn't random but the expectation is.

You shouldn't be able to predict the outcome of a set between two evenly matched players. If there is only one, neutral stage to play on, this statement holds true. However, once you add CPing into the mix, you should predict that the winner of the first match wins the set because the CPing system favors them.

You keep discussing the differences in the metagame when bans are removed and I've never denied them, but I don't know how you can consider it a flaw. Every ruleset you could possibly conceive of will benefit some characters and hurt others. How is our ruleset flawed when being used for two players character locked on Mario?

You keep bringing up the example of dittos, but dittos are irrelevant since the MU is even and any stage they select would have no impact, empirically, on their character. We should only talk about non-dittos since stages can affect their advantages and create polarization.

Just because DL has 3 platforms doesn't mean it can't one day become more powerful of a counterpick for spacies than FD currently is for Marth. If you had a stage the size of Hyrule Temple that was also a 3-plat, that would still be by far the most polarizing stage, so clearly layout similarity doesn't mean everything when it comes to which counterpicks are best. Even if DL is never viewed as better for spacies than FD is for Marth, it doesn't negate my point which is that the strengths of counterpicks are just as flexible as the rest of the metagame, and when you consider these are all opinions in the first place (filled with bias as we are all human), it seems pretty dumb to try to balance the cast with stages instead of just picking stages based on neutral criteria.

Well obviously that would be the most polarizing stage in existence, but its purely hypothetical. Dreamland is only a bit bigger than BF in stage size; its blast zones and wind are the only really distinguishing characteristics. Literally, no legal stage in the game creates as much polarization as FD (except for maybe FoD or Yoshis in some limited cases).

 

[Bones0]

I find your claim that 2-minute matches with items being more vanilla is non sequitur since we are speaking in the context of competitive play. Items introduce randomness and aren't conducive to competitive play.

Saying that both options are arbitrary comes off as special pleading. The ruleset, in and of itself, is arbitrary, but your belief is that we shouldn't have any rules that try to balance the cast. Bans and DSR/MDSR actively prevent players from using the most optimal strategies thy can against certain characters; pseudo-buffing said characters as a result. Marth/Pikachu not being allowed to go to FD whenever the lose a game to Fox/Falco can loosely be compared to Wobbling bans at some tournaments. The ruleset is nerfing them to some degree.

Striking also prevents Marth and Pikachu from taking spacies to FD. Is the striking system also an attempt to balance the game? You still fail to understand the simple difference between a ruleset that nerfs and buffs incidentally vs. a ruleset that actively adds in stages to benefit specific characters like the OP was suggesting (add in Brinstar/Mute City because spacies are good on PS).

You are getting far too technical about it. You might as well argue that a coin flip isn't truly random because its ultimately determined by the weight of coin, the amount of force being applied to it, and the height its being dropped from. The actual elements which create the outcome isn't random but the expectation is.

You shouldn't be able to predict the outcome of a set between two evenly matched players. If there is only one, neutral stage to play on, this statement holds true. However, once you add CPing into the mix, you should predict that the winner of the first match wins the set because the CPing system favors them.

Idk what you're talking about at all. If you only play on a single stage the entire set, each player has a 50% chance of winning. If you play in our current ruleset, each player also has a 50% chance of winning. Your complaint is that you can predict the winner of the set based on who wins game 1, but how is that not also true for a set played on a single stage? If we're talking coin-flip-level randomness, whoever wins game 1 has a 75% chance of winning because in order for the player who lost game 1 to win the set, they would have to win 2 games in a row. If the counterpicks odds are 60:40, then the player who lost game 1 has a 60% chance of winning game 2 and a 40% chance of winning game 3. That amounts to a 24% chance of them winning the set. So unless I'm ****ing up my basic probability mathematics here, a player who loses game 1 is only 1% less likely to win the set in the current ruleset than in a single-stage ruleset. It's clear that if you value counterpick odds as worse than 60:40 that the gap becomes wider and wider, but even 60:40 seems like a stretch and I'd want to see some actual hard data to back up a claim that counterpicks matter that much WITH BANS.

You keep bringing up the example of dittos, but dittos are irrelevant since the MU is even and any stage they select would have no impact, empirically, on their character. We should only talk about non-dittos since stages can affect their advantages and create polarization.

I keep bringing up dittos specifically BECAUSE the mu is even. If you base a ruleset around dittos, then you can be sure you aren't actively buffing or nerfing certain characters based on popular opinion.

Well obviously that would be the most polarizing stage in existence, but its purely hypothetical. Dreamland is only a bit bigger than BF in stage size; its blast zones and wind are the only really distinguishing characteristics. Literally, no legal stage in the game creates as much polarization as FD (except for maybe FoD or Yoshis in some limited cases).

It depends a lot on the mu. If FD was really such a polarizing stage, players would never strike to it. While it isn't the most commonly struck stage, it definitely still happens.

 

[DeepDish]

Can someone provide an empirical example as to why Brinstar is no longer a counter-pick? I get that some people don't like the acid, but if it was any other stage and you fall off the sides, you'd be dead. Also, players familiar with the stage should be able to understand how far the acid levels rise at specific points.

 

[1MachGO]

Striking also prevents Marth and Pikachu from taking spacies to FD. Is the striking system also an attempt to balance the game? You still fail to understand the simple difference between a ruleset that nerfs and buffs incidentally vs. a ruleset that actively adds in stages to benefit specific characters like the OP was suggesting (add in Brinstar/Mute City because spacies are good on PS).

Striking and banning are apples and oranges. When you strike, you are compromising to a stage you both are willing to play the first match on. Banning denies your opponent their best counterpick advantage.

And why aren't incidental buffs/nerfs a problem?

Idk what you're talking about at all. If you only play on a single stage the entire set, each player has a 50% chance of winning. If you play in our current ruleset, each player also has a 50% chance of winning. Your complaint is that you can predict the winner of the set based on who wins game 1, but how is that not also true for a set played on a single stage? If we're talking coin-flip-level randomness, whoever wins game 1 has a 75% chance of winning because in order for the player who lost game 1 to win the set, they would have to win 2 games in a row. If the counterpicks odds are 60:40, then the player who lost game 1 has a 60% chance of winning game 2 and a 40% chance of winning game 3. That amounts to a 24% chance of them winning the set. So unless I'm ****ing up my basic probability mathematics here, a player who loses game 1 is only 1% less likely to win the set in the current ruleset than in a single-stage ruleset. It's clear that if you value counterpick odds as worse than 60:40 that the gap becomes wider and wider, but even 60:40 seems like a stretch and I'd want to see some actual hard data to back up a claim that counterpicks matter that much WITH BANS.

You are ****ing up basic probability mathematics. The odds in each match always remain 1:1. This is because each match is an individual instance that should go unaffected by whichever event preceded it. With counterpicking, the odds can change from match to match so the winner of the first match is always favored.

I keep bringing up dittos specifically BECAUSE the mu is even. If you base a ruleset around dittos, then you can be sure you aren't actively buffing or nerfing certain characters based on popular opinion.

By that logic, KJ64 should be legal because its very fair in a ditto. However, its banned because it allows game breaking strategies such as circle camping to occur (something which is entirely MU based). Jungle Japes might be in a similar situation (not necessarily because it facilitate circle camping, but it probably gives ludicrous advantages to certain characters)

It depends a lot on the mu. If FD was really such a polarizing stage, players would never strike to it. While it isn't the most commonly struck stage, it definitely still happens.

I don't understand what your point is. Do you believe that there is a legal stage which creates more polarization in more MUs than FD?

 

[Elyssa Xey Hexen]

// File: StageLegality.c
// Author: Xeylode

#include <stdio.h>

void main(void){
int stageLegality;
stageLegality = fox && Falco;
if( stageLegality == fox && Falco){
printf("Remove Fox&Falco, then discuss.");
}
return 0;
}

 

[Bones0]

Striking and banning are apples and oranges. When you strike, you are compromising to a stage you both are willing to play the first match on. Banning denies your opponent their best counterpick advantage.

And why aren't incidental buffs/nerfs a problem?

Striking denies the opponent their best counterpick in the same exact way. The only difference between striking and banning is we only allow players to get rid of 1 stage instead of 2. The counterpick process is a way of compromising to a stage; we've just changed the process so that we have a tendency to play on different stages for each game.

You are ****ing up basic probability mathematics. The odds in each match always remain 1:1. This is because each match is an individual instance that should go unaffected by whichever event preceded it. With counterpicking, the odds can change from match to match so the winner of the first match is always favored.

How is the winner of the first match favored more with bans than with a single stage? You say each match is an individual instance, but you keep talking about the probability of the winner of game 1 winning the whole set. I'm fairly certain my math is correct, and insisting every match is a 50-50 doesn't do anything to explain why counterpicking hurts the chances of the player who loses game 1. He may be less likely to win game 3, but he's also more likely to win game 2 whereas without counterpicks both games are an equal shot of losing.

By that logic, KJ64 should be legal because its very fair in a ditto. However, its banned because it allows game breaking strategies such as circle camping to occur (something which is entirely MU based). Jungle Japes might be in a similar situation (not necessarily because it facilitate circle camping, but it probably gives ludicrous advantages to certain characters)

When I said we should construct a ruleset focused on ditto matchups, I meant for banning and counterpicking, not necessarily for stage lists. Obviously if a single or small set of characters has an extreme edge over the rest of the cast that has to be taken into account. I don't consider that even slightly comparable to including or excluding bans based on how bad spacies are on FD or how good floaties are on Brinstar.

 

[1MachGO]

Striking denies the opponent their best counterpick in the same exact way. The only difference between striking and banning is we only allow players to get rid of 1 stage instead of 2. The counterpick process is a way of compromising to a stage; we've just changed the process so that we have a tendency to play on different stages for each game.

Lol but one player isn't counterpicking when you're striking... so how is their their best counterpick being denied? The relationship between two players within the circumstances of the set are completely different before and after game 1.

How is the winner of the first match favored more with bans than with a single stage? You say each match is an individual instance, but you keep talking about the probability of the winner of game 1 winning the whole set. I'm fairly certain my math is correct, and insisting every match is a 50-50 doesn't do anything to explain why counterpicking hurts the chances of the player who loses game 1. He may be less likely to win game 3, but he's also more likely to win game 2 whereas without counterpicks both games are an equal shot of losing.

I've already explained this. Counterpicking changes odds whereas a constant change keeps them static.

Comparison:

W/ Counterpicking (winner of game 1 is favored):
>Game 1, stage is neutral, players are of equal skill, odds are 50-50 (Both players are just as likely to win)
>Game 2, Loser of game 1 CPs to a non-neutral stage, the odds tip into his favor 60-40 (CPing player is more likely to win)
>Game 3, Loser of game 2 CPs to a non-neutral stage, the odds tip into his favor 60-40 (CPing player is more likely to win)

Single stage only (winner is entirely random):
>Game 1, stage is neutral, players are of equal skill, odds are 50-50 (Both players are just as likely to win)
>Game 2, stage is neutral, players are of equal skill, odds are 50-50 (Both players are just as likely to win)
>Game 3, stage is neutral, players are of equal skill, odds are 50-50 (Both players are just as likely to win)

You're belief that odds change based on the outcome of the last game is wrong and its basically a gamblers fallacy (http://en.wikipedia.org/wiki/Gambler's_fallacy). Each match is statistically independent and the odds which determine the outcome won't change match to match.

When I said we should construct a ruleset focused on ditto matchups, I meant for banning and counterpicking, not necessarily for stage lists. Obviously if a single or small set of characters has an extreme edge over the rest of the cast that has to be taken into account. I don't consider that even slightly comparable to including or excluding bans based on how bad spacies are on FD or how good floaties are on Brinstar.

Well this is a convenient moving of the goalposts. Regardless, dittos are a terrible reference when it comes to stage selection since (as I've stated earlier) picking a stage has no effect on the respective advantages. You might as well pick "random" in a ditto.

 

[Combo Blaze]

IIRC, the 64 community had already done this. A stage list, in and of itself, is so fundamentally flawed that tradition is the only reason I can think of for why it still exists.

Japan ruleset is DL only.
NA ruleset is DL neutral stage, Congo and Peach's counterpick.
Peru ruleset is hyrule only.
Brazil ruleset is any stage except for Sector Z (don't know the specifics).

 

[tauKhan]

I've already explained this. Counterpicking changes odds whereas a constant change keeps them static.

Comparison:

W/ Counterpicking (winner of game 1 is favored):
>Game 1, stage is neutral, players are of equal skill, odds are 50-50 (Both players are just as likely to win)
>Game 2, Loser of game 1 CPs to a non-neutral stage, the odds tip into his favor 60-40 (CPing player is more likely to win)
>Game 3, Loser of game 2 CPs to a non-neutral stage, the odds tip into his favor 60-40 (CPing player is more likely to win)

Single stage only (winner is entirely random):
>Game 1, stage is neutral, players are of equal skill, odds are 50-50 (Both players are just as likely to win)
>Game 2, stage is neutral, players are of equal skill, odds are 50-50 (Both players are just as likely to win)
>Game 3, stage is neutral, players are of equal skill, odds are 50-50 (Both players are just as likely to win)

Yes, the individual probabilities of each match look like that, but @ Bones0 correctly calculated the actual probabilities of winner of match 1 winning the set, which in first case is 76% and the second one 75%, so the difference is very small if cp's don't change probabilities drastically.

 

[1MachGO]

Yes, the individual probabilities of each match look like that, but @ Bones0 correctly calculated the actual probabilities of winner of match 1 winning the set, which in first case is 76% and the second one 75%, so the difference is very small if cp's don't change probabilities drastically.

Did you read up on the gambler's fallacy? (http://en.wikipedia.org/wiki/Gambler's_fallacy)

Assuming that a coin will show heads after getting tails 10 times in a row is no different than assuming a coin will flip heads after getting tails once. The likelihood of that assumption occurring is theoretically (theoretically being the key word here) higher, but the actual odds per coin flip will always remain the same (50:50). Look at the the coin toss simulator: the rate of increase between heads and tails is not systematic. Furthermore, a set of three games is such a small sample size that we shouldn't be surprised by complete randomness in the outcome.

 

[MookieRah]

@ E Elyssa Xey Hexen why do you include stdio.h?

@ 1 1MachGO
That isn't Gambler's Fallacy, actually. He isn't stating that it is more likely for the event to happen again simply because it has already happened, he's simply deducing that the player who took game 1 only needs to win one more flip. That means that there is a 75% chance that the winner of game one will win the set, because the chances of two coin flips being in the favor of the other player is two 50/50, which is indeed 25%. Obviously, you already know this, but you need to understand the fallacy before you argue with it.

We can see from the above that, if one flips a fair coin 21 times, then the probability of 21 heads is 1 in 2,097,152. However, the probability of flipping a head after having already flipped 20 heads in a row is simply 1⁄2. This is an application of Bayes' theorem.

This illustrates my argument. Taken as a whole, the chances are 75%, but each individual match is still 50/50.

 

[ShrieK1295]

We should change some of the current neutrals to counterpicks, not add more stages. All the other stages SUCK BADLY. This the least any ruleset has ever favored spacies. Having 3 neutrals and 3 CPs would be best because some of the current neutrals are not very neutral, and spacies have a big advantage in striking in many matchups.

 

[1MachGO]

@ E Elyssa Xey Hexen why do you include stdio.h?

@ 1 1MachGO
That isn't Gambler's Fallacy, actually. He isn't stating that it is more likely for the event to happen again simply because it has already happened, he's simply deducing that the player who took game 1 only needs to win one more flip. That means that there is a 75% chance that the winner of game one will win the set, because the chances of two coin flips being in the favor of the other player is two 50/50, which is indeed 25%. Obviously, you already know this, but you need to understand the fallacy before you argue with it.


This illustrates my argument. Taken as a whole, the chances are 75%, but each individual match is still 50/50.

Actually, a gambler's fallacy is when you believe the last result/repetition of a result means its less likely to reoccur. Your second sentence is speaking to the reversal of the fallacy (which is also a fallacy) but not the one I was specifically referring to.

In the context of a set between two players, saying player 2 is going to win the second match after losing game 1 is an instance of the fallacy because the chances of that occurring are independently 50:50. (Results of fair odds do not occur systematically) This fallacy would then repeat itself when going into game 3.

Like I said, the assumption is correct theoretically/mathematically, but since each match result is totally random and statistically independent we shouldn't be surprised from any result.

Some other stuff on the odds:

>However, with enough data, the winner of game one would be moderately favored due to the Bo3 system
If the two same exact players with a 50/50 MU against each other for 100+ sets (the rule of large numbers), we would probably find that the winner of game 1 wins the set 66% of the time (not 75%). This is because there are only three possible outcomes; each with an equal chance of occurring:

Player 1 takes game one> Player 1 takes game two (Player 1 wins set)
Player 1 takes game one>Player 2 takes game two>Player 1 takes game three (Player 1 wins set)
Player 1 takes game one>Player 2 takes game two>Player 2 takes game two (Player 2 wins set)

While this would favor the winner of game 1, I doubt we would see the same exact MU between two evenly skilled players long enough for the rule of large numbers to take effect.

>If playing by your rules (saying that repeated results are less likely) then neutral-stage-only still doesn't favor game 1's winner as much as Bones0 originally postulated (A 1% difference)
The comparison between a set with a neutral stage and one with counterpicks is being oversimplified since more variables are involved in counterpicks.

In a neutral stage only scenario, we are only looking at the relationship in player skill. In counterpicking, we have to consider both player skill and the additional effects of the counterpick.

Bones0 math:

Neutral stage only (the stage isn't a factor for the outcome):
50/50 game one <<< based on skill since the stage is a non element
75/25 (loser of game 1's favor)
75/25 (loser of game 2's favor)

Counterpicks (the stage affects the outcome):
50/50 game one <<< based on skill since the stage is a non element
60/40 (loser of game 1's favor) <<< Better odds for game 1's winner than neutral stage??
76/24 (loser of game 2's favor) <<< result of .6 * .4

To elaborate, if we are saying that two players are so equally matched that the odds of either of them winning is a 50:50 coin toss (which only occurs when the variable of the stage is removed), then theoretically, the odds of the same person winning twice in a row would be 25% BEFORE we even factor in the effects of the stage:

Correct math:

Neutral stage only (the stage isn't a factor for the outcome):
50/50 game one <<<based on skill since the stage is a non element
75/25 (loser of game 1's favor)
75/25 (loser of game 2's favor)

Counterpicks (the stage affects the outcome)
50/50 game one <<< based on skill since the stage is a non element
75/25 + effects of 60/40 counterpick (loser of game 1's favor) <<< Since the players have a coin toss relationship in terms of skill, the repeated occurrence of game 1's result is only 25%. Their chances are then made lower by the effects of the counterpick
75/25 + effects of 60/40 counterpick (loser of game 2's favor)

Assuming this is correct, these are the conclusions we can make in regards to NSO vs. CPs

>The likelihood that game 1's winner is going to take the set is high in both of them (but not as infinitesimal as Bones0 originally implied)
>CPing has a higher chance of going to game 3, but a smaller chance that game 1's loser will win
>Evenly skilled players should probably play a Bo1 because there is at least a 75% they are going to waste each other's time.

 

[Elyssa Xey Hexen]

@ E Elyssa Xey Hexen why do you include stdio.h?

printf function.

 

[MookieRah]

@ 1 1MachGO
It's not the fallacy, I even posted an example from the very thing you referenced showing you that it isn't. If you take each match individually, then every match has a 50/50. If you give one player the first win, and look at the next two potential matches as a set, there is a 25% chance for the loser of the first match to win. That doesn't negate the fact that those matches have a 50/50 shot each, but to say that the chances for someone to flip heads twice in a row is the same as someone flipping tails once, is obviously not true.

Also, I wasn't including the stage advantage as a part of this, because I was discussing from a ditto standpoint. I did that for simplicity, because you didn't understand the Gambler's Fallacy correctly.

 

[Elyssa Xey Hexen]

If I am to legitimately talk about this discussion I think one should bring into mind what stages one wants to bother playing on in the first place? What other stages would one see being remotely usable as a stage anymore?

Within my own opinion I have gotten rather annoyed with stages affecting matches. The forefront of competition should be putting on displays that competes one person's skill against another. The question to me is should stages be included within this idea of displaying skill? I do not believe there are many stages where this is a feasible assumption. If anything stages do nothing more than to further increase division among advantages. I suppose this is the point of a counterpick, but sometimes stage selections can feel like a total overall of a match-up. Stages should only include small variances within a match-up and nothing very drastic.

The biggest annoyance with many of the old counterpick stages were the positional advantages being far too strong. Let us use Falco or Peach on Congo jungle for example. These two characters get a huge advantage over many other characters with slow jump heights. The current stage list is basically the same stage with minor variations. DL has high platforms and large boundaries. Characters like Falco can get an advantage over Marth here. While on YS Marth's disadvantage is minimized. However, the effects are not glaringly one-sided. Falco can easily destory Marth on YS if it turns against him.

 

[1MachGO]

@ 1 1MachGO
It's not the fallacy, I even posted an example from the very thing you referenced showing you that it isn't. If you take each match individually, then every match has a 50/50. If you give one player the first win, and look at the next two potential matches as a set, there is a 25% chance for the loser of the first match to win. That doesn't negate the fact that those matches have a 50/50 shot each, but to say that the chances for someone to flip heads twice in a row is the same as someone flipping tails once, is obviously not true.

Also, I wasn't including the stage advantage as a part of this, because I was discussing from a ditto standpoint. I did that for simplicity, because you didn't understand the Gambler's Fallacy correctly.

Like I said, mathematically/theoretically, you are correct.

However, the gambler's fallacy is in place because you are assuming that the winner of game 2 has a 25% chance of winning game 3 when it is in fact 50/50. Saying you are "looking at it as a whole" is basically special pleading because the prior assumption has to be in place in order for you to be correct.

Again, go look at the coin flip simulator on that wikipedia page and see how unsystematic the coin flips are. A coin flip is statistically independent and will ALWAYS have 50/50 odds. You can calculate the odds of a certain run occurring but it isn't necessarily going to be true in practice. IMO, saying that the winner of game 1 has a 66% chance of winning (since 2 out of the 3 equally likely outcomes are in their favor) is more likely.

 

[MookieRah]

Like I said, mathematically/theoretically, you are correct.

Then it isn't Gambler's Fallacy, as I'm not at any point suggesting that because of Event A, Event B is less likely to happen even though the chance of even B is still 50/50.

Now, if what you said is true and that it's more like 66%, then that is simply because I'm not a mathematician and/or statistician. Excuse my simplification of this less obvious mathematics and just move onto your point instead of derailing the thread with this tangental math crepe.

 

[1MachGO]

Then it isn't Gambler's Fallacy, as I'm not at any point suggesting that because of Event A, Event B is less likely to happen even though the chance of even B is still 50/50.

Okay, well what about this: player 1 won game 1, what are the odds of player 1 winning game 2?

Even if you aren't outright saying it, at some point you are observing the fallacy. Otherwise, you would say that player 2 has a 50% chance of winning if it goes to game 3.

Now, if what you said is true and that it's more like 66%, then that is simply because I'm not a mathematician and/or statistician. Excuse my simplification of this less obvious mathematics and just move onto your point instead of derailing the thread with this tangental math crepe.

Well I explained why this was the case in my long *** post. And yeah, I agree, this has gone on way too long.

My argument has basically been that a single, neutral stage is more fair than CPing.

Though, personally, I like a varied stage list; I just think its just interesting that we use a faulty ruleset.

 

[tauKhan]

Did you read up on the gambler's fallacy? (http://en.wikipedia.org/wiki/Gambler's_fallacy)

Yes, I know the fallacy. I didn't fall for it, and I calculated probability of winner of game 1 winning the set assuming that both players have 50/50 chance of winning each game.

P(winner of game 1 loses game 2) = 1/2 and P(winner of game 1 loses game 3 if it happens) = 1/2. Therefore P(winner of game 1 loses the set) = 1 - (1/2)*(1/2)

>However, with enough data, the winner of game one would be moderately favored due to the Bo3 system
If the two same exact players with a 50/50 MU against each other for 100+ sets (the rule of large numbers), we would probably find that the winner of game 1 wins the set 66% of the time (not 75%). This is because there are only three possible outcomes; each with an equal chance of occurring:

Player 1 takes game one> Player 1 takes game two (Player 1 wins set)
Player 1 takes game one>Player 2 takes game two>Player 1 takes game three (Player 1 wins set)
Player 1 takes game one>Player 2 takes game two>Player 2 takes game two (Player 2 wins set)

While this would favor the winner of game 1, I doubt we would see the same exact MU between two evenly skilled players long enough for the rule of large numbers to take effect.

First of all, the amount of data is irrelevant to the probability of an outcome of a set. I don't even understand why you bring that up.

Your analysis goes wrong with the bolded part: The chances are NOT equal. Player 1 has 50/50 chance of winning game two. So it's equally likely that case 1 happens or either of the other cases to happen. Therefore the other cases have individually smaller probability (25%) of happening.

Another way of looking at this is assuming that game 3 is played out even if Player 1 already won the set.
Then you have 4 equally probable cases, 3 of which result in Player 1 winning.

Correct math:

Neutral stage only (the stage isn't a factor for the outcome):
50/50 game one <<<based on skill since the stage is a non element
75/25 (loser of game 1's favor)
75/25 (loser of game 2's favor)

Now this is an example of the Gambler's fallacy.

 

[1MachGO]

Your analysis goes wrong with the bolded part: The chances are NOT equal. Player 1 has 50/50 chance of winning game two. So it's equally likely that case 1 happens or either of the other cases to happen. Therefore the other cases have individually smaller probability (25%) of happening.

Another way of looking at this is assuming that game 3 is played out even if Player 1 already won the set.
Then you have 4 equally probable cases, 3 of which result in Player 1 winning.

The amount of data is relevant if we are trying to gauge this information in a real life scenario. I brought it up because I had been emphasizing how unsystematic 50/50 results are. Game 1 favoritism would only be noticeable in this ridiculous scenario.

However, you are correct about there being a 50% chance of going to game 3, thus, there is a 75% chance favoritism to the winner of game 1. I WAS WRONG. My fervent disagreement stemmed from these kind of statements:

P(winner of game 1 loses game 2) = 1/2 and P(winner of game 1 loses game 3 if it happens) = 1/2. Therefore P(winner of game 1 loses the set) = 1 - (1/2)*(1/2)

Which, to me, looks like the odds of game 3 are based off the result of game 2 (gambler's fallacy).

However, when comparing every possible outcome, I understand now. I had a mental lapse.

Now this is an example of the Gambler's fallacy.

I prefaced all of that information with the statement "playing by your rules". All of that was done to illustrate how even if we say that odds change based on the last result (gambler's fallacy), the odds are still more favorable to game 2's loser in a neutral stage scenario than a CPing one. The illustration was more or less a contingency plan and wasn't my core argument.

With that said, since I was wrong about there being an instance of the gambler's fallacy, I will use some aspects of the aforementioned contingency plan. Namely that CPing only changing the odds by 1% is probably not true (even so, 75% is still fairer than 76%).

 

[Ringedge]

Can someone give me a TL;DR of this? I am curious of what it has to say. I was debating allowing a few Counter-picks being played a smash club I am trying to start at my school.

 

[tauKhan]

With that said, since I was wrong about there being an instance of the gambler's fallacy, I will use some aspects of the aforementioned contingency plan. Namely that CPing only changing the odds by 1% is probably not true (even so, 75% is still fairer than 76%).

The 76% is arrived at the same way, just changing parameters. It's definitely the correct value when win% on cp stage is assumed to be 60%. P(winner of game 1 loses game 2) = 6/10 and P(winner of game 1 loses game 3 if it happens) = 4/10. Therefore P(winner of game 1 loses the set) = 1 - (6/10)*(4/10) = 0,76

I used to think that the cp system emphasizes game 1 too much, but after running these calculations I don't anymore. I think that in bo3 with bans it's reasonable to assume mu's don't usually go from 50/50 to better than 70/30 at bad cases, and those cases the difference between cps and single stage is still acceptable. Of course real mu's aren't necessarily symmetrical like that and having more stages changes mu's. However, if we assume that a player has a good counter pick while his opponent has none, the first one has a higher chance of winning under ruleset which has cps whether he wins or loses the game. I think it's reasonable to assume that the current stage ruleset doesn't make game 1 much more important than it would be if we played on a single stage.

 

[1MachGO]

Eh, I find the 1% difference hard to believe as it seems like there are more variables associated with counterpicks. Its hard to say.

Anyway, I may have come up with a solution that utilizes counterpicks and de-emphasizes game 1. I'll make a thread for it soon.

 

[theMagnumDragon]

I think Poke Stadium is fine only if we start to allow a ban in Best of Five. It is too dominate for Fox/Falco, and the transformations make it random which doesn't follow competitive play