EDIT: UGH! Sorry for the huge post!
The traditional role of a referee (umpire etc.) is to adjudicate between two competing interests, such as the two sports teams on the field. In an RPG, there is only a single team on the field - the players, through their characters. Their opposition cannot stand up and call for a ruling - they are fictional. This is where we get the "killer DM" stereotype, the DM who treats the challenges he places as a measure of his own skills, and becomes an adversary to the players.
The wargame referee from which the DM evolved was also a neutral third party arbitrating rule disputes between two players (or teams) engaged in the game. The DM is much more a game participant (albeit one with a very different role) than the classic referee or umpire.
Interestingly from the terminology perspective, Kreogsspiel began as "Instructions for the Representation of Tactical Maneuvers under the Guise of a Wargame". IOW, it was viewed as the façade of a game, but not actually a game, by its own designers. You posit that the manner in which many play D&D makes it "no longer a game". I wonder when we pass the marker by which the creator of Kreogsspiel would perceive it being transformed into a game.
Reading this history, it seems like the move to "Free" Kriegsspiel is extremely similar to the evolution of the classic wargame into D&D and other role playing games. It is interesting that you classify this evolution as moving away from "a game", especially as I suspect the original Kriegsspiel designer would likely have perceived "Free" Kriegsspiel as more of a game than his own model. It being 64 years later, his views cannot actually be known, of course.
I note other games by the same name exist, though I presume the reference to "Free" Kriegsspiel indicates we are discussing the original and its evolution.
First, one criticism of the experience model in early versions of D&D was that the gaining of "points" (xp) often failed to correspond to successes of the characters. For example, if their goal was to protect the local village from the depredations of nearby monster tribes, they were rewarded for killing and looting the monsters. But they could also earn both xp and gold for killing the villagers and looting their village. They would not, however, be rewarded as well for driving off the monsters, and not at all for negotiating a peace treaty between the village and the monsters which could allow both to become far more prosperous.
Where are these defined "endgame conditions"? Many games, especially from OD&D to AD&D 2e, had no plan for an "end" to the game. Characters continued adventuring more or less in perpetuity, gaining higher and higher levels. At about 9 - 11 level (dependent on the class), advancement became extremely slow. 3e accelerated level gains, and also capped them, setting a target for character retirement, but not a campaign goal, however by that point, I suggest the game had evolved from the roots you are focused on.
Finally, from the player's perspective "dead character" and "retired character" have pretty similar outcomes - the player creates and brings in a new character. We certainly don't tally up victory points and declare a player "the winner" at the end of the typical campaign, even when it has "an end" rather than just dying out.
Emphasis added - this is commonly presented as a reason why D&D (and other RPG's) differ markedly from games people are more familiar with - they lack a defined endpoint, defined conditions of victory and a defined winner.
So your interpretation, then, is that this group continued for months despite the fact no one was having any fun. That seems highly unlikely to me. I would also note that the detractors of each and every edition of D&D typically point to some element which they refer to as "broken". Presumably, then, the game you are playing is just as objectively "broken"* as the game played by those in the anecdote.
* Which is to say, not at all. That people are still playing and enjoying it means, at least to me, that it cannot objectively be "broken".
[grammarnazi]Might we apply another "non-game's" rules? "You're" means "You are". I believe you are seeking a possessive which would be "your".[/grammarnazi]
To the reality, you have claimed the DM must adjudicate solely from the rules. The examples I provided can all be easily envisioned as arising in a game, and have no defined rules for adjudication in D&D, although greater structure has evolved over time. The jump to social skills in 3e is, IMO, the clearest example of a jump in evolution, but still leaves considerable judgment required by the DM. The only game I can think of where there is a rule given for dealing with matters which are not covered by the rules is Toon, which quite simply states there are two possibilities - a thing will happen, or it will not, so flip a coin. Every other game instructs the GM to exercise best judgment in assigning probabilities to results and adjudicating the outcome.
My hobgoblin example provided the state of the game board, saw the players take what I think most of us would consider a reasonable, permissible action and asked how you would apply the rules to set the die roll required and the possible outcomes. No such rules exist. GM improvisation is used to resolve such situations, with greater or lesser rules guidance depending on the rules set in use.
You said the players cannot move the tokens, by the rules. Now you are saying they can, but the action might be interrupted by factors unknown to them. It seems like the rules are much less clear-cut than you initially asserted.
Gygax's introduction to the 1e DMG, if read literally, indicates that no one who DM's a game (and thus needs to read the DMG) may ever play in a game again, as he has seen the DMG. I believe Gary himself played in others' games, and that some of his players ran their own games. In any event, it was in response to my request you cite a rule requiring a screen, not a rule requiring the DM have knowledge the players lack (whether rules knowledge, a precept of that 1e intro that many gamers dispute, or campaign knowledge such as the location, strength and other details of adversaries).
The fact my hobgoblin scenario (and those raised earlier) cannot be easily resolved by that statblock seems to pretty clearly indicate that this is not the statblock you previously described, which would permit easy adjudication/mechanical resolution of any interaction with the creature described in that statblock, sticking entirely to the rules with no improvisation required of the DM.
What you are describing strikes me as the first definition, which is role playing, but is not a game. D&D evolved from wargames (where players direct tactical units) to a game where the player controlled a single individual, which better fits the second definition of taking on the role of another person, in this case a fictional person created by the player. Such a fictional person, properly role played, will not be a pawn or playing piece, but will come to life through the role playing of the player. To me, at least, that is the "RP" in "RPG".
I can easily play Talisman, Dungeon or the D&D board games like Wrath of Ashardalon as a Game (no RP required or desired). At least to me, a true RPG goes beyond that "G" to add "RP". If we extract the "G", we remove the classical "game" element of rules as mechanics for adjudicating success and failure, and move to a pure collaborative theatre exercise (which can still be a "game" in many senses of the word), a very basic example being the "Bakery" game referred to some ways above.
I think the personality became a matter of greater focus as time wore on (let's remember that the first edition of D&D was published in 1974 and, if not the first RPG, is generally considered the first commercially available RPG. The first widely available edition was published in 1977, which was the first opportunity for the game to move outside the circle of wargamers. There is not a lot of pre-'80's RPG history to address, just its ancestors and historical roots in wargaming). The old Rogues Gallery book published some characters with CHARACTER. That was back in the Holmes Basic/1st print AD&D books, published in 1980 for AD&D. Perhaps the characters presented therein, with their somewhat limited personality sketches, may have lead to the gamers with whom you were familiar back in the '70s adopting personalities for their characters for the first time (speculation just because of your and the book's timing), but that in no way indicates that other groups did not play characters with CHARACTER in the '70s. It does indicate that the founders of the game were playing characters with personalities, as that is where these characters were drawn and adapted from.
Being one of those who bought The Rogues Gallery off the shelf back in 1980, I have my own history with the hobby to draw on. It bears noting, however, that I came to the hobby with no wargaming roots and preconceptions.
Emphasis and references added. I would note:
1. The reference to a large volume of dice rolling indicates this is a change from the manner in which a living DM would have generated a dungeon. This implies an expectation that DM's typically used their own improvisation and judgment in designing and stocking dungeons, even back in 1975 when this article would have been published. This is wholly inconsistent with your random table model - the tables here are clearly perceived as differing from the activity of a DM, even in the days of the game's infancy.
2. This "sealed envelope" mechanism reinforces the role of the DM in creating interesting and challenging encounters which simply cannot be simulated by any random roll mechanism. It also indicates that players were expected to possess the skills, and the rules knowledge, of a DM. It is players exchanging these sealed envelopes, but DMs creating their contents, so members of these various groups are acting as DM's for other groups while acting as players in their own groups. Of course, this was in 1975 - the DMG was not published until 1979, so the article writer could not incorporate Gary's intro.
3. This again suggests that the players have access to the full rules, including those relevant to the DM. Again, about four years before Gary's intro to the DMG saw print.
Overall, this also reinforces the need for a DM who does more than just roll dice and consult random tables, as even this simulation is expected to fail without adding the improvisation of a DM. That it is rarely if ever used (certainly from the late 1970's to now, in my experience) further demonstrates the failure of simple random tables to emulate a "true D&D experience". This is, as you say, why the game requires a DM, not just players.
Pemerton's claim matches my experience as well. I invite anyone who has ever extrapolated the DM's monster placement and/or dungeon design random table from game play to post their contrary experience. @pemerton also acknowledges there may be some statistical error in his sample. If your model is widespread, it seems like examples of such extrapolations should be similarly widespread.
It does depend largely on definitions and every table will vary in its desired balance of "RP" to "G". However, in my view, there is a point where the RP faces into such obscurity that we have only "G", a mechanistic and tactical exercise. One example is the suggested use of random charts for every aspect of dungeon design. Another is where the characters are mere ciphers - pawns lacking personality - and not Characters. At the other end of the continuum, the "G" fades away and we move to live theater. One example of this is the DM who ignores any mechanics of social skills (limited in older editions to reaction rolls; 3e added skills like Diplomacy, Intimidate and Bluff) in favour of pure role playing of such interactions. That impacts only one element. A broader example would return us to the resolution mechanics of "let's pretend", where there are no dice, only player consensus as to the results of any given effort.
If they are drawbacks for the player in the sense that they reduce the fun, I agree. However, often the fun is in playing the drawbacks of the character, not just the strengths. I enjoy playing my impatient and impulsive character because, while often less than tactically perfect, his impetuous tactics are fun. Ditto my berserker highlander who stares that Umber Hulk right in the eyes, and does not know the Monster Manual stats rhyme and verse, even if I myself am 100% conversant with the monster's stats. A steady stream of "I avert my eyes in the manner that provides the greatest save bonus with the least to hit penalties and pick the best tactic every time because I am a robot lacking any personality" is not nearly as memorable, and (perhaps because) it is not nearly as fun.
This is quite true. It requires the DM view those restrictions as challenges, not as impediments. If, instead, we present the Paladin with no-win situations (either he compromises his principals or he/the party dies; refusal to act dishonourably guarantees defeat; Paladin's Dilemma where every choice violates the code and causes a fall from grace) or just sweep his beliefs under the rug when they become inconvenient (routine "the Paladin leaves the room, whereupon we begin torturing the prisoner"; the Evil characters have lead shields so the Paladin never notices they are evil; "my code doesn't require me to do anything about my friends being completely dishonourable"), then there is no point playing such a character.
Superman doesn't kill. Instead, he finds alternatives to succeed without killing, even when they are difficult or inobvious. If we instead write Superman into a scenario where he either kills or loses ("well, since you refused to kill Lex Luthor, Earth is destroyed"), we violate the character. This is no different from designing a game world where the Paladin can never succeed. If you don't want LG characters, run a world where LG behaviour equals losing. If you want to encourage less bloodthirsty characters, they must achieve positive results from less bloodthirsty solutions (not "every prisoner we take betrays us; every enemy we spare comes back to attack us again; every helpless prisoner we rescue backstabs us"). I'm always amazed by GM's who will complain bitterly that their players don't follow genre tropes, when they ensure that every instance of such tropes results in adverse consequences (not only no benefits, but a penalty).
Flanking provides a bonus, so players try to flank. Using a weapon with which one is not proficient imposes a penalty, so players avoid using such weapons. If honourable behaviour provides bonuses, players will be inclined to play honourable characters. If honourable behaviour virtually always results in the character being taken advantage of, expect a lot more backstabbers and murderhobos in your PC group.
I would say if role playing results in challenges (it doesn't have to result in bonuses), it makes the game more fun. That, in itself, is a reward - bonuses not required. But if it results in failure, plain and simple, then role playing will simply disappear. Like the dinosaurs, it simply could not survive in that environment.
You are ignoring the role of a referee (umpire, etc.) in a game. They stop the game, usually, to make a call. It can be done while the game continues on, but the actual players who are playing the game to achieve goals in it need to be informed of the judgment calls. Of course, most players are refereeing themselves and others too, but neutral referees are used - as in D&D - to insure adherence to the rules.
The traditional role of a referee (umpire etc.) is to adjudicate between two competing interests, such as the two sports teams on the field. In an RPG, there is only a single team on the field - the players, through their characters. Their opposition cannot stand up and call for a ruling - they are fictional. This is where we get the "killer DM" stereotype, the DM who treats the challenges he places as a measure of his own skills, and becomes an adversary to the players.
The wargame referee from which the DM evolved was also a neutral third party arbitrating rule disputes between two players (or teams) engaged in the game. The DM is much more a game participant (albeit one with a very different role) than the classic referee or umpire.
Kreigsspiel is a great game. "Free" Kriegsspiel isn't a game the same way the Calvinball isn't a game. People can attempt to treat such as an illusion as a game, but as I said before to @pemerton, most folks wouldn't play such as an actual game except ironically.
Interestingly from the terminology perspective, Kreogsspiel began as "Instructions for the Representation of Tactical Maneuvers under the Guise of a Wargame". IOW, it was viewed as the façade of a game, but not actually a game, by its own designers. You posit that the manner in which many play D&D makes it "no longer a game". I wonder when we pass the marker by which the creator of Kreogsspiel would perceive it being transformed into a game.
Wikipedia said:Kriegsspiel in its original form was not particularly popular among the Prussian officer corps; The rules were cumbersome and games took much longer than the battles that they were supposed to represent. It was not until 1876 that General Julius von Verdy du Vernois had the idea of placing more power in the hands of the gamemaster in order to speed up the game and reduce the number of rules. von Verdy's “Free” Kriegspiel did away with many of the movement and combat rules in order to save time, giving the duty of deciding the effects of orders and combat to the gamemaster. This allowed players to play a game in real time, giving the players a better feel for the tension of actual combat. To retain military accuracy, von Verdy emphasized the necessity of using military experts as gamemasters. The new “Free” Kriegspiel soon gained more popularity than its predecessor (now known as “Strict” Kriegsspiel”); The Prussian (later German) General Staff used it both for its internal exercises and as a training tool.
Reading this history, it seems like the move to "Free" Kriegsspiel is extremely similar to the evolution of the classic wargame into D&D and other role playing games. It is interesting that you classify this evolution as moving away from "a game", especially as I suspect the original Kriegsspiel designer would likely have perceived "Free" Kriegsspiel as more of a game than his own model. It being 64 years later, his views cannot actually be known, of course.
I note other games by the same name exist, though I presume the reference to "Free" Kriegsspiel indicates we are discussing the original and its evolution.
But there are points scored and definitive endgame conditions for every player and the campaign itself. Specifically, retirement or loss of character individually and as a whole).
First, one criticism of the experience model in early versions of D&D was that the gaining of "points" (xp) often failed to correspond to successes of the characters. For example, if their goal was to protect the local village from the depredations of nearby monster tribes, they were rewarded for killing and looting the monsters. But they could also earn both xp and gold for killing the villagers and looting their village. They would not, however, be rewarded as well for driving off the monsters, and not at all for negotiating a peace treaty between the village and the monsters which could allow both to become far more prosperous.
Where are these defined "endgame conditions"? Many games, especially from OD&D to AD&D 2e, had no plan for an "end" to the game. Characters continued adventuring more or less in perpetuity, gaining higher and higher levels. At about 9 - 11 level (dependent on the class), advancement became extremely slow. 3e accelerated level gains, and also capped them, setting a target for character retirement, but not a campaign goal, however by that point, I suggest the game had evolved from the roots you are focused on.
Finally, from the player's perspective "dead character" and "retired character" have pretty similar outcomes - the player creates and brings in a new character. We certainly don't tally up victory points and declare a player "the winner" at the end of the typical campaign, even when it has "an end" rather than just dying out.
If you defined games as exclusively needing a winner determined, then perhaps Name Level could be assigned for each player. I think this could be detrimental to cooperative (not collaborative) games. D&D is after all not about winning, but about succeeding. About players actually improving themselves according to the test which is the game design. IMnshO, D&D's initial design is at the heart of what games are. Quite unlike group storytelling.
Emphasis added - this is commonly presented as a reason why D&D (and other RPG's) differ markedly from games people are more familiar with - they lack a defined endpoint, defined conditions of victory and a defined winner.
Neither of us could know, right? Because they still so wanted to play a game, even a broken one? Because no one knew how to fix it without ruining the game, (i.e. the mechanical balance)?
So your interpretation, then, is that this group continued for months despite the fact no one was having any fun. That seems highly unlikely to me. I would also note that the detractors of each and every edition of D&D typically point to some element which they refer to as "broken". Presumably, then, the game you are playing is just as objectively "broken"* as the game played by those in the anecdote.
* Which is to say, not at all. That people are still playing and enjoying it means, at least to me, that it cannot objectively be "broken".
You're questions make no sense. You're asking where are the rules for "this behavior" in a game that doesn't have those game elements as having that behavior?
[grammarnazi]Might we apply another "non-game's" rules? "You're" means "You are". I believe you are seeking a possessive which would be "your".[/grammarnazi]
To the reality, you have claimed the DM must adjudicate solely from the rules. The examples I provided can all be easily envisioned as arising in a game, and have no defined rules for adjudication in D&D, although greater structure has evolved over time. The jump to social skills in 3e is, IMO, the clearest example of a jump in evolution, but still leaves considerable judgment required by the DM. The only game I can think of where there is a rule given for dealing with matters which are not covered by the rules is Toon, which quite simply states there are two possibilities - a thing will happen, or it will not, so flip a coin. Every other game instructs the GM to exercise best judgment in assigning probabilities to results and adjudicating the outcome.
Judgement refers to assessing the state of the game board according to the rule pattern preset. New DM material comes from die roll generation to determine outcomes. All new player material need only be defined as needed to deal with the current action, with all the rest abstracted.
My hobgoblin example provided the state of the game board, saw the players take what I think most of us would consider a reasonable, permissible action and asked how you would apply the rules to set the die roll required and the possible outcomes. No such rules exist. GM improvisation is used to resolve such situations, with greater or lesser rules guidance depending on the rules set in use.
#1. Of course, the pattern behind the screen can be drawn out in front if players can receive all that information under the rules. This is well known. Players can attempt to take actions like movement, but they still require the referee to determine if such attempts are possible. That the player moves the piece is partly like players rolling the dice. But minis also enables players to more fine tune their description attempts for how they want to move.
You said the players cannot move the tokens, by the rules. Now you are saying they can, but the action might be interrupted by factors unknown to them. It seems like the rules are much less clear-cut than you initially asserted.
#2This is most blatantly obvious when we recognize the rest of the world understands game walkthroughs to beat computers games as cheating rather than attempting to play the game for one's self.
Gygax's introduction to the 1e DMG, if read literally, indicates that no one who DM's a game (and thus needs to read the DMG) may ever play in a game again, as he has seen the DMG. I believe Gary himself played in others' games, and that some of his players ran their own games. In any event, it was in response to my request you cite a rule requiring a screen, not a rule requiring the DM have knowledge the players lack (whether rules knowledge, a precept of that 1e intro that many gamers dispute, or campaign knowledge such as the location, strength and other details of adversaries).
I believe 2e finally included in the statblocks cultural behavior like alignment for reaction checks, ability scores, organization, % in lair, environment, morale, and loyalty adjustments. Monster Manual 1977 included all kinds of social organizational designs in the descriptions. Those are suggestions for a DM to use in their design. Many are very good.
The fact my hobgoblin scenario (and those raised earlier) cannot be easily resolved by that statblock seems to pretty clearly indicate that this is not the statblock you previously described, which would permit easy adjudication/mechanical resolution of any interaction with the creature described in that statblock, sticking entirely to the rules with no improvisation required of the DM.
Maybe you don't know about roleplaying in the 50s-70s after the war, but it wasn't about fictional personas. D&D is the iconic RPG as the term roleplaying was used in army wargame simulations. They taught soldiers their role. D&D is a game where players improve their ability to perform their role (class) by mastering the game system it refers to. They can prove this and increase needed class abilities to more easily overcome and accomplish higher level challenges and objectives by scoring points relating to their roleplaying.
Wikipedia said:Role-playing refers to the changing of one's behaviour to assume a role, either unconsciously to fill a social role, or consciously to act out an adopted role. While the Oxford English Dictionary offers a definition of role-playing as "the changing of one's behaviour to fulfill a social role", in the field of psychology, the term is used more loosely in four senses:
To refer to the playing of roles generally such as in a theatre, or educational setting;
To refer to taking a role of an existing character or person and acting it out with a partner taking someone else's role, often involving different genres of practice;
To refer to a wide range of games including role-playing video game, play-by-mail games and more;
To refer specifically to role-playing games.
What you are describing strikes me as the first definition, which is role playing, but is not a game. D&D evolved from wargames (where players direct tactical units) to a game where the player controlled a single individual, which better fits the second definition of taking on the role of another person, in this case a fictional person created by the player. Such a fictional person, properly role played, will not be a pawn or playing piece, but will come to life through the role playing of the player. To me, at least, that is the "RP" in "RPG".
I can easily play Talisman, Dungeon or the D&D board games like Wrath of Ashardalon as a Game (no RP required or desired). At least to me, a true RPG goes beyond that "G" to add "RP". If we extract the "G", we remove the classical "game" element of rules as mechanics for adjudicating success and failure, and move to a pure collaborative theatre exercise (which can still be a "game" in many senses of the word), a very basic example being the "Bakery" game referred to some ways above.
Improvising a personality wasn't part of roleplaying in the RPG community until the 80s. Personality stuff was also fun to do, but like in any game it can interfere with a person trying to play a game.
I think the personality became a matter of greater focus as time wore on (let's remember that the first edition of D&D was published in 1974 and, if not the first RPG, is generally considered the first commercially available RPG. The first widely available edition was published in 1977, which was the first opportunity for the game to move outside the circle of wargamers. There is not a lot of pre-'80's RPG history to address, just its ancestors and historical roots in wargaming). The old Rogues Gallery book published some characters with CHARACTER. That was back in the Holmes Basic/1st print AD&D books, published in 1980 for AD&D. Perhaps the characters presented therein, with their somewhat limited personality sketches, may have lead to the gamers with whom you were familiar back in the '70s adopting personalities for their characters for the first time (speculation just because of your and the book's timing), but that in no way indicates that other groups did not play characters with CHARACTER in the '70s. It does indicate that the founders of the game were playing characters with personalities, as that is where these characters were drawn and adapted from.
Being one of those who bought The Rogues Gallery off the shelf back in 1980, I have my own history with the hobby to draw on. It bears noting, however, that I came to the hobby with no wargaming roots and preconceptions.
Why we need DMs to play D&D
"Although it has been possible for enthusiasts to play solo games of DUNGEONS & DRAGONS by means of 'Wilderness Adventures', there has been no uniform method of dungeon exploring, for the campaign referee has heretofor been required to design dungeon levels. Through the following series of tables (and considerable dice rolling) it is now possible [1] to adventure alone through endless series of dungeon mazes! After a time I am certain that there will be some sameness to this however, and for this reason a system of exchange of sealed envelopes for special rooms and tricks/traps is urged. [2] These envelopes can come from any other player and contain monsters and treasure, a whole complex of rooms (unfolded a bit at a time), ancient artifacts, and so forth. All the envelope should say is for what level the contents are for and for what location, i.e. a chamber, room, 20' wide corridor, etc. Now break out your copy of D&D [3], your dice, and plenty of graph paper and have fun!"
--The Strategic Review Vol. 1 No. 1
Emphasis and references added. I would note:
1. The reference to a large volume of dice rolling indicates this is a change from the manner in which a living DM would have generated a dungeon. This implies an expectation that DM's typically used their own improvisation and judgment in designing and stocking dungeons, even back in 1975 when this article would have been published. This is wholly inconsistent with your random table model - the tables here are clearly perceived as differing from the activity of a DM, even in the days of the game's infancy.
2. This "sealed envelope" mechanism reinforces the role of the DM in creating interesting and challenging encounters which simply cannot be simulated by any random roll mechanism. It also indicates that players were expected to possess the skills, and the rules knowledge, of a DM. It is players exchanging these sealed envelopes, but DMs creating their contents, so members of these various groups are acting as DM's for other groups while acting as players in their own groups. Of course, this was in 1975 - the DMG was not published until 1979, so the article writer could not incorporate Gary's intro.
3. This again suggests that the players have access to the full rules, including those relevant to the DM. Again, about four years before Gary's intro to the DMG saw print.
Overall, this also reinforces the need for a DM who does more than just roll dice and consult random tables, as even this simulation is expected to fail without adding the improvisation of a DM. That it is rarely if ever used (certainly from the late 1970's to now, in my experience) further demonstrates the failure of simple random tables to emulate a "true D&D experience". This is, as you say, why the game requires a DM, not just players.
howandwhy99 said:Well that's a hell of a claim you couldn't possibly back up. How do you prove by reference every player playing in every D&D game ever? I mean, that is stuff even people do when playing videogames, primary stuff. Like recognizin the patterns in levels of Pac-Man or Tetris or BBEGs to beat them. But you just the need to make sweeping generalizations about the D&D population for 40 years? What's your purpose for starting this thread?
Pemerton's claim matches my experience as well. I invite anyone who has ever extrapolated the DM's monster placement and/or dungeon design random table from game play to post their contrary experience. @pemerton also acknowledges there may be some statistical error in his sample. If your model is widespread, it seems like examples of such extrapolations should be similarly widespread.
I tend to find this rather judgemental, especially from someone who has fairly recently posted criticism of someone else in this thread for "one true way-ism".
It does depend largely on definitions and every table will vary in its desired balance of "RP" to "G". However, in my view, there is a point where the RP faces into such obscurity that we have only "G", a mechanistic and tactical exercise. One example is the suggested use of random charts for every aspect of dungeon design. Another is where the characters are mere ciphers - pawns lacking personality - and not Characters. At the other end of the continuum, the "G" fades away and we move to live theater. One example of this is the DM who ignores any mechanics of social skills (limited in older editions to reaction rolls; 3e added skills like Diplomacy, Intimidate and Bluff) in favour of pure role playing of such interactions. That impacts only one element. A broader example would return us to the resolution mechanics of "let's pretend", where there are no dice, only player consensus as to the results of any given effort.
I prefer a game in which drawbacks for the character aren't drawbacks for the player but, rather, opportunities for the player.
If they are drawbacks for the player in the sense that they reduce the fun, I agree. However, often the fun is in playing the drawbacks of the character, not just the strengths. I enjoy playing my impatient and impulsive character because, while often less than tactically perfect, his impetuous tactics are fun. Ditto my berserker highlander who stares that Umber Hulk right in the eyes, and does not know the Monster Manual stats rhyme and verse, even if I myself am 100% conversant with the monster's stats. A steady stream of "I avert my eyes in the manner that provides the greatest save bonus with the least to hit penalties and pick the best tactic every time because I am a robot lacking any personality" is not nearly as memorable, and (perhaps because) it is not nearly as fun.
Roger Musson had begun to work out this approach in 1981, when he wrote "I believe that the restrictions on some character classes, though they might be viewed as disadvantages, are more the reverse. Restrictions make it easier to play "in character" by dictating necessary attitudes. A paladin should be noted by his largesse and flamboyant acts of charity; these make him more interesting than a stereotyped fighting man."
This is quite true. It requires the DM view those restrictions as challenges, not as impediments. If, instead, we present the Paladin with no-win situations (either he compromises his principals or he/the party dies; refusal to act dishonourably guarantees defeat; Paladin's Dilemma where every choice violates the code and causes a fall from grace) or just sweep his beliefs under the rug when they become inconvenient (routine "the Paladin leaves the room, whereupon we begin torturing the prisoner"; the Evil characters have lead shields so the Paladin never notices they are evil; "my code doesn't require me to do anything about my friends being completely dishonourable"), then there is no point playing such a character.
Superman doesn't kill. Instead, he finds alternatives to succeed without killing, even when they are difficult or inobvious. If we instead write Superman into a scenario where he either kills or loses ("well, since you refused to kill Lex Luthor, Earth is destroyed"), we violate the character. This is no different from designing a game world where the Paladin can never succeed. If you don't want LG characters, run a world where LG behaviour equals losing. If you want to encourage less bloodthirsty characters, they must achieve positive results from less bloodthirsty solutions (not "every prisoner we take betrays us; every enemy we spare comes back to attack us again; every helpless prisoner we rescue backstabs us"). I'm always amazed by GM's who will complain bitterly that their players don't follow genre tropes, when they ensure that every instance of such tropes results in adverse consequences (not only no benefits, but a penalty).
Flanking provides a bonus, so players try to flank. Using a weapon with which one is not proficient imposes a penalty, so players avoid using such weapons. If honourable behaviour provides bonuses, players will be inclined to play honourable characters. If honourable behaviour virtually always results in the character being taken advantage of, expect a lot more backstabbers and murderhobos in your PC group.
If you roleplaying really bumps into or cuts across your "roll playing" then, to me, that tends to suggest that your mechanics aren't really doing their job of facilitating the desired play.
I would say if role playing results in challenges (it doesn't have to result in bonuses), it makes the game more fun. That, in itself, is a reward - bonuses not required. But if it results in failure, plain and simple, then role playing will simply disappear. Like the dinosaurs, it simply could not survive in that environment.