Balance Discussions Part 2

[orangeandblack5]

My question is why a role "designed to be top tier" is good

[MysticMismagius]

My question is why a role "designed to be top tier" is good

obligatory i'm not betty but imo it's less that it's a good thing and more that this design means that trying to do something about it is infinitely more difficult
it's nigh impossible to make all the things it can do really "balanced" because it just does so much
so you can either nerf it such that the biggest pain points it causes are resolved (which often requires multiple passes and/or gives the role Little Mac status), rework it entirely, or yeetus deletus

[Achilles]

I will be revisiting everything in this thread in about 2 weeks when the new Town Traitor patch is live and bug free.

[BustedBetty]

You should realize that nerfs make gamethrowing as that role less effective, not more. (Real gamethrowing, I mean.)

Jailor
Jailor cannot jail the same player on consecutive nights (similar to Pirate). If a Jailor executes a Town member they will commit suicide the following night (like Vigilante).


How does that make gamethrowing less effective exactly?

Also I have no idea what you're talking about with that Jailor threw stuff.

[ak521]

You should realize that nerfs make gamethrowing as that role less effective, not more. (Real gamethrowing, I mean.)

Jailor
Jailor cannot jail the same player on consecutive nights (similar to Pirate). If a Jailor executes a Town member they will commit suicide the following night (like Vigilante).


How does that make gamethrowing less effective exactly?

Also I have no idea what you're talking about with that Jailor threw stuff.

sigh this kid again. Well, people will be rewarded less for gamethrowing if they suicide the next night. A Jailor can't jail a vigilante or a similar role multiple times for evils to maintain majority. How would these help gamethrowing? They don't.

[Chemist1422]

gamethrowing isn't the issue

the issue is that most people consider jailor unfun to play against since it's capable of utterly destroying the mafia with no fault on their part

[Transcender]

guys, isn't the escort vigilante doctor who can talk to their targets at night sooo balanced?


TOWN
POWER

[Brilliand]

To be fair, escort+doctor by itself isn't so bad. Adding the Escort ability to the Doctor is used as a Doctor nerf in some games.

...which is why the Jailor so seldom goes out of his way to protect people

[PatrykSzczescie]

What do you think about jailor removing from ranked rolelist and replacing it with TK?

[MysticMismagius]

I think that having a guaranteed role for the sake of having a guaranteed role is kinda dumb, and thus would be attracted to the idea of replacing Jailor with TK, but it's not enough.

Making Jailor spawn only sometimes solves the problems caused by Jailor... only sometimes. This leads to swing where the game is heavily influenced by whether or not Jailor spawns. We can already see this with Mayor and Ret.

[Transcender]

To be fair, escort+doctor by itself isn't so bad. Adding the Escort ability to the Doctor is used as a Doctor nerf in some games.

...which is why the Jailor so seldom goes out of his way to protect people

But then mayor also rolls cause lol and boom

[orangeandblack5]

As some always point out, jailing Mayor is often not the ideal play.

However, if you stop a kill, then yes it was, and given Doctor is randomly ineffective on Mayor just because...

[Transcender]

As some always point out, jailing Mayor is often not the ideal play.

However, if you stop a kill, then yes it was, and given Doctor is randomly ineffective on Mayor just because...

often, yes
but you always have the choice
so anyways




Town Power

[Chemist1422]

To be fair, escort+doctor by itself isn't so bad. Adding the Escort ability to the Doctor is used as a Doctor nerf in some games.

...which is why the Jailor so seldom goes out of his way to protect people

"nerf"

I'm probably biased but I've always considered Jailkeeper (escort+doctor) to be stronger than doctor or escort individually

[Transcender]

To be fair, escort+doctor by itself isn't so bad. Adding the Escort ability to the Doctor is used as a Doctor nerf in some games.

...which is why the Jailor so seldom goes out of his way to protect people

"nerf"

I'm probably biased but I've always considered Jailkeeper (escort+doctor) to be stronger than doctor or escort individually

it depends on the kind of jailkeeper you got there

[BasicFourLife]

To be fair, escort+doctor by itself isn't so bad. Adding the Escort ability to the Doctor is used as a Doctor nerf in some games.

...which is why the Jailor so seldom goes out of his way to protect people

"nerf"

I'm probably biased but I've always considered Jailkeeper (escort+doctor) to be stronger than doctor or escort individually

it depends on the kind of jailkeeper you got there

whomegalul

[Transcender]

To be fair, escort+doctor by itself isn't so bad. Adding the Escort ability to the Doctor is used as a Doctor nerf in some games.

...which is why the Jailor so seldom goes out of his way to protect people

"nerf"

I'm probably biased but I've always considered Jailkeeper (escort+doctor) to be stronger than doctor or escort individually

it depends on the kind of jailkeeper you got there

whomegalul

Please stop saying this to me i dont know what it means and im very concerned for my safety and youve said it to me multiple times today and why to i hear voices help

[BlastingOff]

I will be revisiting everything in this thread in about 2 weeks when the new Town Traitor patch is live and bug free.

Thank you very much, excited to see what's gonna get in

[QuetzalcoatI]

Too many clueless low-elo (or rather low-winrate) players in this thread who have no idea what they are talking about.

The town/mafia/NK split is already pretty balanced. Town has 9/14 members and should represent about 64% of wins, Mafia 29%, and NK 7%. People don't enjoy roles where they lose more often, and that's the main reason of why people don't enjoy evil roles as much, not because they are underpowered. These numbers are roughly in-line with what it is now. Town winrate may be about 75% in higher elo, but that's because towns are more coordinated and many evils don't try or care (some outright leave or afk). In low elo however, town winrate would be completely in the gutter if these changes went through, as town winrate there is already pretty low as it is (very little skill expression in these elos).

There are really two main issues:

1. Evils don't try or care because being evil is more difficult, lower winrate, and it's more abstract to see how your actions directly correlate to your success. A lot of times it can feel like minor, arbitrary things lose you the game as evil, and oftentimes they do. If evils actually try, even if it just is as simple as not leaving, not AFKing, making a will, and not taking 20 seconds to claim, the mafia winrate goes way up. One good mafia player who actually tries can substantially increase mafia's winrate, even in high elo. Now imagine if everyone who got evil roles actually tried, evil winrates would be 30-40% winrate or more even in the highest of elo.

2. The elo system is very poorly configured. It is just an inflated grind. Players can reach master elo with 40% winrate just by grinding out games. If your winrate isn't good and you don't consistently win games, you shouldn't be climbing. It's an elo ranking system, not an exp system.

Definitely remove the buffer if you die N1 where you lose only half elo and gain full, it just inflates elo over time. Also make the ranges smaller and make the floors/ceilings not so extreme. ToS obviously has skill expression, but for example: a high elo WW in a low elo town gets a +1/-2 ratio while vice versa gets +27/-1. I don't care how high or low elo you are, there is never even going to be remotely close to that big of a gap in winrate regardless of skill level. There is only so much a high elo player can do right and only so much a low elo player can screw up, because there is luck and so many other factors beyond their control that skill can't account for in a game like ToS. The high elo player would need to win 66% of their games to breakeven and the low elo player only needs to win about 3%. This is even more true for town/mafia where even just one bad player can completely screw your game. Also due to these extreme ranges, many games in high elo come down to either gaining a little or losing a ton if you mess up or gaining a lot and losing little if you mess up. To climb up in elo you should need to be a consistently strong player who is able to win more as your factions than the average player at your elo rank does, not just simply grind out games.

Another possible change is to have the ratios adjust based on the elo. In a 1300 elo lobby, town winning can be an accomplishment that deserves some elo reward, but in a 2300 elo loby, where town is expected to win more, the ratios should be adjusted so that the reward should proportionally less, and the punishment proportionally more.

Anyways the RM changes are okay but the mafioso change is beyond horrible. Jailor or ret just die the second they are revealed in any game. That's nerfing town (which doesn't even need nerfs if evils actually tried and were competent) in the worst way possible. I think ret is fine as it is, but don't like the idea of the townie being back only temporarily. I don't see why mafia is being buffed and town is being nerfed (and has been nerfed repeatedly in recent balance patches).

[PatrykSzczescie]

To nerf jailor meta quite a bit, make jailor forced to execute on N1 when witched into the jailee.

[orangeandblack5]

The town/mafia/NK split is already pretty balanced. Town has 9/14 members and should represent about 64% of wins, Mafia 29%, and NK 7%. People don't enjoy roles where they lose more often, and that's the main reason of why people don't enjoy evil roles as much, not because they are underpowered. These numbers are roughly in-line with what it is now. Town winrate may be about 75% in higher elo, but that's because towns are more coordinated and many evils don't try or care (some outright leave or afk). In low elo however, town winrate would be completely in the gutter if these changes went through, as town winrate there is already pretty low as it is (very little skill expression in these elos).

I am completely unable to see any reason that we should assume that winrates "should" be proportional. You can say "Town should represent X% of wins" all day, but just stating such does not do a very good job discussing why you believe that is better than alternative options. In particular, it is, at the end of the day, a team-based game. Proportion should not factor into it very much at all when looking at the Town and the Mafia - they should have, provided both teams have an equal skill level, roughly even chances to win, as that both feels fair to players (you'll note this would significantly help this issue with people leaving as Mafia because they lose more often that you seem to just accept) and allows the outcome to be more dictated on the relative skill of the players involved. Making it strictly proportional throws away these gains for very little benefit - the only advantage is that, on average, people win more games. But these wins are not truly earned, and are effectively meaningless if everybody is winning more. At the end of the day, this just doesn't outweigh the downsides of going with target winrates based purely on proportion.

There are really two main issues:

1. Evils don't try or care because being evil is more difficult, lower winrate, and it's more abstract to see how your actions directly correlate to your success. A lot of times it can feel like minor, arbitrary things lose you the game as evil, and oftentimes they do. If evils actually try, even if it just is as simple as not leaving, not AFKing, making a will, and not taking 20 seconds to claim, the mafia winrate goes way up. One good mafia player who actually tries can substantially increase mafia's winrate, even in high elo. Now imagine if everyone who got evil roles actually tried, evil winrates would be 30-40% winrate or more even in the highest of elo.

Putting aside the fact that this is still not a high enough scum winrate, the fact of the matter is that this was not and reportedly still is not the case in higher elo lobbies. You can't convince me that all of the high elo players nowadays only care about playing Town, and it certainly wasn't true in the past. Yet higher elo games have often been even more townsided than the lower elo games, something you literally just acknowledged in your first paragraph! This is, plain and simple, either a scathing attack on every high-elo player or a flat-out contradiction.

On one hand, let's say you honestly believe almost no high elo players try as scum and they all suck at it. Then shouldn't those people not be in high elo? If you don't try any time you play one of the two major factions in the game, you are hardcapping yourself at 50% of the skills required to be a good player. If you honestly believe that players like this are making it into high elo, then you should in no way be supporting proportional winrates, which are what would allow this sort of inflation based purely on town games in the first place! So, if this is how you feel, you should be against winrates favoring proportionality over skill.

On the other hand, let's say you think that high elo players do try as scum, but them trying as town is just more effective and so the scum lose anyways. In that case, that just stands to prove that, when utilized correctly, the strength of town is too high relative to that of the scum. When we look at lower elo games then, which you admit see a lot of evil players leave because they don't feel like they can win or even make an impact, the problem is the same - scum consistently do not perform well across all elo ranges. It is effectively a contradiction, then, to argue both that "scum always leave in low elo because they lose too much" and "if scum were better town would never win." The scum can't both be weak enough to cause people to leave constantly because they lose too much and on the verge of being overpowered if given a few buffs. The idea that "oh but if they tried" is nonsensical because the players that do try should naturally rise through the ranks while the players that don't should fall down, and again we see that the solution is to abandon winrates based on proportionality and move towards winrates more focused on determining the winner based on player skill.

A Ranked mode should be balanced to promote skilled players rising through the ranks.

2. The elo system is very poorly configured. It is just an inflated grind. Players can reach master elo with 40% winrate just by grinding out games. If your winrate isn't good and you don't consistently win games, you shouldn't be climbing. It's an elo ranking system, not an exp system.

The elo system itself is something I have not studied enough to have a proper opinion on the finer points of the algorithm. However, it doesn't take a genius to see that if town is winning 64% of the time, and you are town 64% of the time, you are going to keep rising in the ranks as time goes on, as you are just going to win more and more games. On a long enough timescale, proportional winrates lead to inflated elo. Yet again, the clear first step towards fixing this problem is to change our target winrates to be closer to 50/50 across Town/Mafia.

Definitely remove the buffer if you die N1 where you lose only half elo and gain full, it just inflates elo over time. Also make the ranges smaller and make the floors/ceilings not so extreme. ToS obviously has skill expression, but for example: a high elo WW in a low elo town gets a +1/-2 ratio while vice versa gets +27/-1. I don't care how high or low elo you are, there is never even going to be remotely close to that big of a gap in winrate regardless of skill level. There is only so much a high elo player can do right and only so much a low elo player can screw up, because there is luck and so many other factors beyond their control that skill can't account for in a game like ToS. The high elo player would need to win 66% of their games to breakeven and the low elo player only needs to win about 3%. This is even more true for town/mafia where even just one bad player can completely screw your game. Also due to these extreme ranges, many games in high elo come down to either gaining a little or losing a ton if you mess up or gaining a lot and losing little if you mess up. To climb up in elo you should need to be a consistently strong player who is able to win more as your factions than the average player at your elo rank does, not just simply grind out games.

Another possible change is to have the ratios adjust based on the elo. In a 1300 elo lobby, town winning can be an accomplishment that deserves some elo reward, but in a 2300 elo loby, where town is expected to win more, the ratios should be adjusted so that the reward should proportionally less, and the punishment proportionally more.

Again I feel like we're eye to eye in principle - however, I think something you might be overlooking is that a lot of people arguing for changes in this thread are doing so with the express intent of reducing swing first and foremost. You mention "luck and so many other factors beyond their control" and that is exactly what many of us here are aiming to try and fix. Many changes you might see suggested and look at as "but that isn't imbalanced" are being suggested expressly to take care of this exact problem. I encourage you to look at things again with this in mind before you start writing off people as "clueless" and having "no idea what they are talking about."

Anyways the RM changes are okay but the mafioso change is beyond horrible. Jailor or ret just die the second they are revealed in any game. That's nerfing town (which doesn't even need nerfs if evils actually tried and were competent) in the worst way possible. I think ret is fine as it is, but don't like the idea of the townie being back only temporarily. I don't see why mafia is being buffed and town is being nerfed (and has been nerfed repeatedly in recent balance patches).

I still fail to see how you can argue so many things that point towards evening out the factions and then end with "but we don't need to even out the factions that's dumb." Sure, I'm also not a huge fan of the Mafioso change at all (you're forgetting Doctor, which means that this only works 50% of the time, which is, for reasons discussed earlier, more of a problem than if it was 100% reliable), but the idea that the factions don't need changes at least somewhat in line with what is in the OP is radically divergent from your own logic on how things "should" be. I'm not sure exactly where or why you pull a 180 from disliking certain issues with Ranked to advocating against steps to help fix them, but I would be interested in hearing more about how you see things changing for the better if nothing actually changes.

[QuetzalcoatI]

You don't play ranked and you don't play the game at any high-level with the best players out there and it shows man. Also it irritates me when people spam these huge walls of text (while failing to bring up any valid points), so I'm just going to cut to your replies only so it doesn't flood the thread.

I am completely unable to see any reason that we should assume that winrates "should" be proportional. You can say "Town should represent X% of wins" all day, but just stating such does not do a very good job discussing why you believe that is better than alternative options. In particular, it is, at the end of the day, a team-based game. Proportion should not factor into it very much at all when looking at the Town and the Mafia - they should have, provided both teams have an equal skill level, roughly even chances to win, as that both feels fair to players (you'll note this would significantly help this issue with people leaving as Mafia because they lose more often that you seem to just accept) and allows the outcome to be more dictated on the relative skill of the players involved. Making it strictly proportional throws away these gains for very little benefit - the only advantage is that, on average, people win more games. But these wins are not truly earned, and are effectively meaningless if everybody is winning more. At the end of the day, this just doesn't outweigh the downsides of going with target winrates based purely on proportion.

Winrates should be proportional because that is the most fair to the most players and it also keeps it so that 50% is roughly the "average" winrate. Yes, obviously if you raised mafia winrate up, that would make mafia more fun to play because people would have more chance of win, but it would also make town less fun to play for over twice as many people every single game. Also a major part of it is that if you buff evils to increase the high elo evil winrate, then the evils will just run rampant in lower elos since their winrate is already quite high there and you already have very little agency and town is extremely passive as it is. I never said strictly proportional, obviously that's never going to happen, but that should be used as a general guideline for a good experience for all players and a balanced game.

Putting aside the fact that this is still not a high enough scum winrate, the fact of the matter is that this was not and reportedly still is not the case in higher elo lobbies. You can't convince me that all of the high elo players nowadays only care about playing Town, and it certainly wasn't true in the past. Yet higher elo games have often been even more townsided than the lower elo games, something you literally just acknowledged in your first paragraph! This is, plain and simple, either a scathing attack on every high-elo player or a flat-out contradiction.

You don't play ranked, you don't play high elo, and you have no experience to speak from at all. Evil is a much different skillset than town and it's also one that people don't like as much. Evil has a very high skill ceiling and requires coordination, knowing your win condition, being able to read both town and scums, leading town effectively, convincing others of your role despite having very limited information, etc. to be able to play at a truly high level (although if all evil players tried as hard for evil as they did town, evil winrate would see a huge spike). Look at some of the top players over the years play evil; there's a ton going on at once and it's a ton of information to process while also fooling the town. There are only a handful of truly great evil players while there are many, many great town players. Evil can be difficult, it can be crushing, and oftentimes you lose due to minor decision making or arbitrary things that you couldn't have known would make you lose until they do, so it requires a ton of experience, game knowledge, deception, and a strong mental to win consistently. The reason why evils may be unsatisfying to some is that the winrate is lower and they get it less often; people struggle seeing over the long-term what actually works because they don't get mafia as often and losses tend to stick stronger in their memory than unfair wins.

On one hand, let's say you honestly believe almost no high elo players try as scum and they all suck at it. Then shouldn't those people not be in high elo? If you don't try any time you play one of the two major factions in the game, you are hardcapping yourself at 50% of the skills required to be a good player. If you honestly believe that players like this are making it into high elo, then you should in no way be supporting proportional winrates, which are what would allow this sort of inflation based purely on town games in the first place! So, if this is how you feel, you should be against winrates favoring proportionality over skill.

Yes, of course they are in high elo even if they are bad at scum. They really shouldn't be, but when the elo system as it is serves as nothing more than a glorified exp grinding system rather than a system that puts players of similar skill level together, then yes, there are tons of unskilled players in high elo. You should have to win more games than players within that elo range to climb, but you don't... and this creates complacency from players to not ever have to try as evils because they will continually climb regardless due to how broken and worthless the elo system is. Again, essentially the only thing high elo represents currently is experience in playing a ranked and competitive mode. As long as you are over roughly 40% winrate (which is laughably low, some of the top players are at 55-63%), you will reach highest elo queues soon enough. The games quality as a whole is still miles better in high-elo ranked than it is in low-elo ranked or RP or any other modes, but individually, the players can be of significantly different skill levels, because the only thing they have in common is that they have thousands of games played and that they are more serious players on average.

On the other hand, let's say you think that high elo players do try as scum, but them trying as town is just more effective and so the scum lose anyways. In that case, that just stands to prove that, when utilized correctly, the strength of town is too high relative to that of the scum. When we look at lower elo games then, which you admit see a lot of evil players leave because they don't feel like they can win or even make an impact, the problem is the same - scum consistently do not perform well across all elo ranges. It is effectively a contradiction, then, to argue both that "scum always leave in low elo because they lose too much" and "if scum were better town would never win." The scum can't both be weak enough to cause people to leave constantly because they lose too much and on the verge of being overpowered if given a few buffs. The idea that "oh but if they tried" is nonsensical because the players that do try should naturally rise through the ranks while the players that don't should fall down, and again we see that the solution is to abandon winrates based on proportionality and move towards winrates more focused on determining the winner based on player skill.

I already addressed this but yes, if you do try then the evils will win substantially more games. It's not even about being some extremely skilled evil team making crazy coordinated plays. There are so, so many games I see where town wins but wins just barely because some evil(s) left N1, never claimed or had a will, took 20+ seconds to claim, afk'ed, or purposely trolled thinking they didn't stand a chance. Obviously, because I do try as evil, I see more of these in my games because we get a lot closer to winning on average, but if players literally just put forth some modicum of effort and see how the game turns out it would dramatically increase evil winrate. As stated previously, the elo system does not work as intended (which is the major issue) which leads to players not caring about evil roles because they can basically completely write them off and still climb. This creates a vicious cycle, because even if you want to try as evil, it's disheartening to see so many evils just give up and not care in the slightest.

There is a similar issue in League (and many other team games), where if people fall behind, then they just simply give up, troll, afk, and stop caring. When if they all did actually try, they would see that they would win many of these games that they completely wrote off. But because everyone complains about how the game is snowbally and because everyone also has a weak mental, the second something goes wrong and a team falls behind, they tend to give up and then lose. Even if a player only won 10% of these deficit games as opposed to just giving up, that would be a massive increase in win-rate, because so many games consist of one team taking an early lead. This could be the difference between an average 51% winrate player and a 56% winrate player. One is a completely average, hardstuck player who will (should) stay in the same elo range forever or very, very slowly climb, whereas a player who is 56% winrate will climb at a steady pace. It's the same thing in ToS, except it's even worse since people who don't try as mafia will climb regardless, simply because you don't need a high win-rate at all to climb, it just makes it a bit faster.

Additionally, global win-rates tend to actually increase in average in high elo since global town winrate is higher. This is another reason for elo inflation that I already discussed. With that being said, high-elo winrate is much more relevant because there is so much more skill expression and agency from an individual player, whereas in low-elo there are just so many genuinely new players who simply don't know game mechanics at all so it's very difficult to tell if someone is scummy because they could be (and oftentimes are) just a clueless player who is new to the game, or you just have clueless players on your team, you try to help them out and they just ignore you and screw your entire faction. Anyways, no matter how good you are at town, your winrate will plateau at around 54-57% or so if you aren't good at evils. A lot of people are good at town, and with 9 players, even if you are not as good, you can still win a lot of town games. However, the evil winrate is what really separates a top-tier 62% winrate players from 55% winrate decent and consistent players. Unfortunately none of this matters since they are both playing with the afk troll who has 40% winrate and thousands of games (and sometimes they even get sub 1.5k elo new players in their lobby during slow times).

A Ranked mode should be balanced to promote skilled players rising through the ranks.

Exactly, that is what the end-goal should be. The issue is not game balance, the issue is the completely broken elo system, and to an extent the punishment system for allowing this to occur. High-elo should be high-winrate skilled players, not just those who have grinded tons of ranked games out. You shouldn't be able to just leave or afk as evil without getting punished, and you shouldn't be able to just not try as evil and still climb and be in the high elo queues. The game is so much fun when you get competent evils who are deceptive and aggressive. What's not fun though is just simply nerfing town and buffing mafi which leads to many of the frustrating and hopeless situations where the game is lost by N2, or where you have literally 5+ TP or TK claims and you have to pick which ones you think are evil with essentially no information to go off.

The elo system itself is something I have not studied enough to have a proper opinion on the finer points of the algorithm. However, it doesn't take a genius to see that if town is winning 64% of the time, and you are town 64% of the time, you are going to keep rising in the ranks as time goes on, as you are just going to win more and more games. On a long enough timescale, proportional winrates lead to inflated elo. Yet again, the clear first step towards fixing this problem is to change our target winrates to be closer to 50/50 across Town/Mafia.

Yes, it's quite obvious you don't understand how the elo system in ToS works, or how an elo system works in general. If town is a higher winrate, then their ratio should reflect that. Because the ratio doesn't change based on elo range, this means that in high-elo, town is a bit more generous with elo, and mafia a bit more strict. If town (on average) wins 50% of games in low-elo and 75% of games in high-elo, then low-elo town range should be (purely example of my point) say -5 to +5 while high elo town range should be -6 to +2. Now, there are a number of ways you could implement this, and I'm not sure what the best would be, but I think it's quite obvious that some form of change does need to be implemented. When global faction winrates vary so substantially by elo, you can't just the same ratio for all elo ranges as this leads to major problems (such as having many dozens of 40% winrate players in high masters). This might be a controversial change though, as most players are bad and like to be able to just climb purely by grinding games even if their winrate is low and never improves.

What's funny is the game could literally could just give you +1 for a win and -1 for a loss and that very basic system would still be miles better than we have right now. It would stop low winrate players from climbing, it would make players try more as evil since an evil loss loses you just as much elo as a town loss, and it would remove the absurd ranges that reward you for queuing at specific times to get into a higher elo lobby. In the current system, a low-elo player in a high-elo lobby basically has a totally free game because if they win, they would get a ton of elo, but if they lose they just get -1, and vice-versa, a high-elo player in a low-elo lobby who has to fight hard for a win (which is very frustrating because they have low-elo teammates and low agency) just to get +1, but if they lose the game then they lose tons of elo.

Again I feel like we're eye to eye in principle - however, I think something you might be overlooking is that a lot of people arguing for changes in this thread are doing so with the express intent of reducing swing first and foremost. You mention "luck and so many other factors beyond their control" and that is exactly what many of us here are aiming to try and fix. Many changes you might see suggested and look at as "but that isn't imbalanced" are being suggested expressly to take care of this exact problem. I encourage you to look at things again with this in mind before you start writing off people as "clueless" and having "no idea what they are talking about."

Those are all factors of a social deception-based party game like ToS. Luck and randomness are inherent to ToS because if there was not some level of uncertainty, scumreading, and tells then it would just be one giant math equation. As it is now, it's a nice mix of using scumreads and processing large amounts of information to piece it together. Many heavily skill-intensive games have some degree of luck. The luck and uncertainty factor is what is the appeal of many of these types of games. And that's fine, because that's what makes each game unique and interesting and gives people claimspace, but the elo system needs to be adjusted to reflect that; a skilled player in a low-elo lobby can never be expected to win 93% of their town games (or even remotely close) if they get into a low-elo game, just like an unskilled player shouldn't get rewarded for simply winning 7% of their games in a high-elo game. Not every game, and definitely not every team game, will always have a skillful player stomp an unskillful player. Rather, it is through large sample and a slight edge that makes them a better player. In a game where 50% is the average, a 55% player is not just 10% better than a 50% winrate player, they are substantially more skilled, and a 60-65% winrate isn't just 20-30% better, when it is essentially the cap of human skill.

I still fail to see how you can argue so many things that point towards evening out the factions and then end with "but we don't need to even out the factions that's dumb." Sure, I'm also not a huge fan of the Mafioso change at all (you're forgetting Doctor, which means that this only works 50% of the time, which is, for reasons discussed earlier, more of a problem than if it was 100% reliable), but the idea that the factions don't need changes at least somewhat in line with what is in the OP is radically divergent from your own logic on how things "should" be. I'm not sure exactly where or why you pull a 180 from disliking certain issues with Ranked to advocating against steps to help fix them, but I would be interested in hearing more about how you see things changing for the better if nothing actually changes.

I am saying we should even out factions, but it would make much more sense for them to be even based on number of members rather than simply existing. 66% of the games in a 14 player lobby be won by a team of 1 or a team of one is not fun for the average player. With each faction being at equal winrate, the average winrate falls to 33%, whereas the average is 50% if factions have roughly representative winrates. I've already said what the issue is. The game is pretty balanced as is. There could be some minor changes to keep things fresh and make RM have more agency (I don't like the framer change presented, but I think disguiser and forger could def use a buff). Changes should be about adding reciprocal counter-play to both sides.

tl;dr I think the game is fairly balanced as it is, and there should be some gameplay changes just to keep things fresh not so much to completely try to shift the winrates. The main issues are that people are not punished for not caring at all or trying with evil, whether is be leaving as evil, afk'ing as evil, or just simply not trying and taking ages to claim on stand or in jail. This can be through punishment (mostly for afk or leaving), but I'm talking mainly through the elo system not punishing them as it should, since it allows players with very low winrates to climb when they never try as evil. Check the winrate of almost any evil afk'er or leaver and they are almost always below 50% (good high elo players tend to be at around 53-60% winrate). When a few players don't try then it creates a vicious cycle, because even good players who try as evil start to get annoyed and discouraged when they see their teammates and allies simply not trying at all or outright leaving without penalty. The game is extremely fun when played at a high level with competent and aggressive evils, but simply buffing evils and nerfing towns leads to frustrating situations where the game is lost N2 or you have to guess which of the 5+ TP claims are evil with little to no information to go off.

[wozearly]

Also it irritates me when people spam these huge walls of text (while failing to bring up any valid points), so I'm just going to cut to your replies only so it doesn't flood the thread.

Hah. I literally just pulled a reply I was making to you, as your reply to orangeandblack explained your thinking a lot more clearly. I'll return to it, though.

I agree with a number of the points you're making, but not necessarily your conclusion on balancing. In the case of ToS, "Evils just don't seem to try, and there are very few good evil players" is, in my view, a symptom of a combination of a number of underlying balancing issues; many of which are tilted in Town's favour. I'll come back to this when I've got time to reframe what I'd written.

Either way, its refreshing to have someone else join the conversation from the angle of "Luck and randomness are inherent to ToS...and are what makes each game unique and interesting". I think this is a regularly misunderstood aspect by people on the opposite side, busy crusading to eliminate as much "swing" as possible.

[MysticMismagius]

Either way, its refreshing to have someone else join the conversation from the angle of "Luck and randomness are inherent to ToS...and are what makes each game unique and interesting". I think this is a regularly misunderstood aspect by people on the opposite side, busy crusading to eliminate as much "swing" as possible.

I've addressed this earlier so I'm going to just repost what I said regarding luck/RNG in ToS here

Well yes, the sheer fact that this game isn't a 1v1 means that there is luck involved in every single role, because you can get fucked over by your teammates or other people through no fault of your own, but that kind of luck is not preventable
I agree that some element of luck can make a game more entertaining or even help bolster skill, but there is such thing as too much
[some "swingy" things that people have been advocating getting rid of ITT] cross that line imo
Wow I really did say "luck" a whole lot

[wozearly]

some "swingy" things that people have been advocating getting rid of ITT cross that line imo

The difficulty is that we (the playerbase) don't all agree on where the line should be drawn.