The fallacy of the quantum ogre argument

Strictly random or railroaded worlds are devoid of player agency

Schrodinger's Ogre

Schrodinger’s Quantum Ogre Quest

There’s an interesting discussion going around the RPG blogosphere about the nature of illusion of control and how a GM can rob fun by not roiling dice with the Universe.

I thought perhaps I might examine this concept and see where it leads us.

Edited to add these relevant links to the discussion:

On How an Illusion Can Rob Your Game of Fun

Player Agency

Let’s start with getting everyone on the same page for this discussion…

Some definitions:

Quantum Event – Any event (encounter or otherwise) that is presented in the game despite any player agency or choice.

In reality, the Quantum descriptor is a misnomer since in reality what we’re discussing here is the illusion, despite any player interactivity, of a collapse of probability options into a given, known result decided before the event determination should arguably be decided. In essence the Quantum Event (say a Quantum Ogre Encounter) is actually nothing more than a predetermined Railroad Event.

From a Quantum Mechanics viewpoint, the Quantum Event (Ogre Encounter) is in violation of a fundamental law of nature since at least one Observer (the GM) knows the outcome of an interaction (is the Ogre encounter here?) before it is even measured (all roads lead to the Ogre.)

Player Agency – Narrative control granted to players that allows them to shape the game much like the GM. It is the granted ability to make choices and impact the game world with these choices and decisions.

It is player agency that separates interactive entertainment such as games from narrative entertainment such as books movies. In fact, player agency is the key factor that moves narrative entertainment into the realm of interactivity.

Predetermination – Deciding something before the question arises.

Randomization – While not an exact descriptor, for our purposes in this discussion we will define it as: Deciding something at the moment the question arises.

While some predetermination is necessary for a game to have any sort of structure, some randomization is necessary for a game to have drama. Varying levels of each are embraced and enjoyed by different gaming groups.

It should be noted that while predetermination and randomization may appear to be mutually exclusive, this is not necessarily the case. In fact, one can oftentimes be the driver of the other; a random selection of predetermined treasure possibilities for example.

Meaningful Choice – Any choice or decision guided by key indicators as to the probable outcome of said choice.

You have only one choice

You have only one choice…

False Choice – Any decision that is offered in which the decision doesn’t really matter and will have no effect on the situation.

Player agency in the absence of meaningful choices yields false choices. However, it should be noted that both require knowledge that the decision does not impart any effect on the outcome of the situation.

As a result, player agency is only measurable in the light of foreknowledge of possible alternate outcomes and is directly measured by how significant the meaningful choice is factored as a deterministic means to (or away from) a specific outcome.

Palette Shifting – The method by which a GM forces a quantum event to occur regardless of player agency. Palette shifting is the end result of false choice in player agency.

Palette shifting is shaping the storyline by simply moving predetermined features and events to match with the choices of player agency. In this regard, player agency is a farce since no actual decision altering future events occurs as all relevant choices become false.

What is interesting about palette shifting is that, like time, it is relative, requiring player knowledge of it’s application to exist. Remove the knowledge of it’s application and the methodology actually disappears from the game.

The quandary of deterministic balance.

Chaos robs choice of meaning in determination:

When a GM allows chaos to determine the game, player agency is also ignored because no meaning can shape the future outcome of an event.

Arguments that predetermination does not allow the game to generate awesome and unexpected results fail to rationalize that the game cannot play itself without any input from the players.

Railroading robs choice of meaning in scope and application:

When a GM railroads the game (by ignoring player agency) into a quantum event, deterministic foreknowledge of probabilities is also ignored because the outcome was determined before the moment of observation.

Arguments that the story is lost when randomization is given center stage fail to take into account that it is this randomization that separates games from stories.

An example of the relativistic nature of Palette Shifting:

The characters are adventuring in the woods. The track before them branches into three directions.

Options abound. The players can choose to have their characters…

  • Go down path A, B or C
  • Go back down the path they just traveled
  • Stop here and make no choice of direction
  • Leave the path entirely
  • Split up sending each character back to the decision point for a multitude of choices
  • Some other decision no GM would ever consider but the players will decide is perfect and logical

While the GM knows that no matter which direction/decision the players decide to take, they will encounter an ogre, from a player perspective, with no foreknowledge of the possibility that they should be able to avoid the ogre encounter, the players have no measure against which to determine that their decision was a false offering.

Since the players have no knowledge that their decision had no impact on the future event, no palette shifting has occurred from their relative perspective.

So is predetermination as expressed in a palette shift a violation of player agency?

From a QM viewpoint, predetermination isn’t a violation of the laws of nature per se, it is the deterministic foreknowledge that is the violation. Either the cat is alive, or it is dead; determination of the state of the cat is collapsed from the superposition of both possible outcomes at the moment of observation. What’s notable is that the possible outcomes are selected from a predetermined subset of possible outcomes.

In an odd way, it is predetermination that is a requirement for meaningful choice because randomization robs choice of meaning. In a randomized setting, no foreknowledge can be used to measure the deterministic outcome of choice and all choices become false in the chaos. Since structure is required for meaning, it is predetermination that empowers player agency. However, if the outcome of the event is also predetermined from a choice of one, then choice also has no meaning.

The answer to this question is yes and no… it’s relative since the palette shift only really occurs if it is a known factor.

The real question is if ignoring player agency robs the game of fun.

Does it truly rob the game of fun?

I say not necessarily. No more than ignoring the dice (or even yourself as the GM) of agency robs the game of fun.

Consider another example:

A character is hard pressed in battle against the ogre. The battle has gone on long into the night, she is nearly out of health, her resources are low, and her companions are all either unconscious or immobilized and unable to assist in the struggle. Things are looking quite bleak.

The players all believe a TPK is on the horizon. Everyone knows this won’t turn out well… everyone but you.

As the GM, you’ve already predetermined that the ogre will be defeated. Player and dice agency are not in play for this decision point.

It is the ogre’s opportunity to dispatch with this final obstacle and feast upon the fallen party. One last hit will spell certain defeat and disaster…

You roll the dice for the ogre’s attack and (pretending to interpret the result but actually ignoring it*) describe how the creature strikes at the character but thanks to cat-like reflexes borne of desperation, misses her by mere inches. The ogre strikes a large hardwood tree splintering the trunk into a thousand small pieces of wood.

Time stands still as the ogre and the character both glance skyward to view the massive tree sway against the starry sky before toppling right onto the hapless ogre.

Pinned beneath the weight of the tree, the ogre is no longer a threat and will be easily dispatched. The day is won, the party will survive.

Were they robbed of fun? Doubtful.

And only if the players are aware of the palette shifting. So long as nobody ever knows that the ogre so easily could have overcome the party could your predetermination in ignoring the agency of the dice be called into question.

Remove the foreknowledge of the possible outcomes and you’ve removed all knowledge of the palette shift… and with removal of that knowledge, the palette shift itself is actually removed because the knowledge is a requirement of the shift by definition.

What robs the game of fun is when the palette shift is known and is in direct opposition to the meaningful choice expectation of the player. Failing that, there is no measure against which to determine how the choice strayed from meaningful to false.

And that is the true fallacy of the Quantum Ogre argument, that it is relativistic and requires player foreknowledge to categorize it as a false choice offering.

Eliminate that foreknowledge and there is no Quantum Ogre.

*This is the very reason why the GM is a referee and not an adversary. The GM’s goal should be fun, not strict adherence to results dictated by the agency of anyone at the talbe.
Posted on
Last modified on:
This entry was posted in Roleplaying game (RPG).
Tagged , , .
Bookmark the permalink.

39 Responses to The fallacy of the quantum ogre argument

  1. WB says:

    Interesting argument, but with a predetermined result one player will know and depending what floats their boat that player’s “fun” will be affected: mine as the GM. I play to find out what happens, not to detail my own story.

  2. Kevin says:

    Perhaps. But as a referee, it is the GMs role in the game to interpret the results of agency and palette shift the fun elements into the forefront, not to sit idly by and just see what happens.

    A good GM knows when to cede to agency to the players, when to cede it to the dice, and when to ignore both in preference to his or her own agency.

  3. WB says:

    I agree that the GM should not simply be standing idly by, but rather their role is to drive play to situations which challenge the players and/or characters. I also agree that without foreknowledge the illusion is unlikely to affect the (non-GM) players fun.

    My concern with your otherwise very interesting analysis is that:
    - You seem to ignore that the GM is a player and this approach *can* impact on their fun.
    - The example of predetermining that they will defeat the Ogre makes it (without further context) seem like the situation (encounter) is pointless to challenging the players/characters. In which case why include it? But this is not necessarily a problem with your overall argument, as I can think of examples that fit your argument that don’t have this aspect.

    Anyway, very interesting and thought provoking overall.

  4. -C says:

    *sigh*

    This post perpetrates the worst kind of enormity on the role-playing community.

    If I were at the table, and at the end of the fight, you fudge some dice and had a tree fall on the ogre, I would get up and walk out and never come back – and this is why.

    I chose to engage the ogre. That is a choice I made. If I and the party were unable to defeat it, then that is the consequence of my choice.

    If I do not have the ability to fail – if negative consequences cannot result from my decisions – then I am not truly making meaningful choices at all, simply sitting there, spinning my wheels until I win.

    By trying to ‘make sure I have fun’ you have in fact removed the only thing gaming gives me that I can’t get from ever other aspect of my life. The ability to make a decision that carries meaningful consequences.

    I see you’ve read my article, and are still not understanding the enormous value in what we are discussing versus games that lack it – here is an article that I did not write that deals with the issue you raise in greater depth. I’m curious if after you read it you will still hold the same viewpoint that it’s ok to rob players of their agency as long as they don’t know about it, or if it’s in their favor.
    http://www.escapistmagazine.com/articles/view/columns/checkfortraps/7485-Judging-the-Game

    Thanks for taking the time to read my comment.

  5. Kevin says:

    *huzzah*

    With a single post about how the Quantum Ogre is relativistic and not the hobgoblin of player agency it is purported to be, I’ve apparently perpetrated the worst of offenses in gaming discussion to date. Sometimes I amaze even myself. I’ll have to add that to my trophy cabinet.

    On to the response:
    It seems to me that you missed the entire point of my post and instead are continuing to argue against the relative nature of the quantum event by granting the players knowledge they don’t necessarily have.
    In my example, the player doesn’t know the dice were ignored. That’s the relativistic nature of the quantum encounter or event. Your arguments seem to want to presume that players know when the GM has palette shifted an event or fudged a dice roll. I say that’s not always likely. Removing this presumption makes your argument less valid.

    Event A happens with Consequences B
    From a player standpoint, there is no way to know if Choices C led to Event A or actually prevented Event D.

    You may presume or surmise that I ignored the dice, and you may subsequently leave the table forever, that’s your choice as a player, but you would be doing so over an assumption, not concrete knowledge. However, to remove my role as interpreter of the results of the situation and the dice is to remove my role as a judge of the world and instead replace it with my role as little more than mouthpiece of the rules and the dice. If that’s the nature of the game you’re interested in playing, then so be it. But it strikes me that to do so puts core of the game back in the stimulation-mode of war gaming.

    Should a GM constantly palette shift? No.
    Should a GM ignore all the dice rolls? Of course not.
    Should a GM routinely remove meaning from choice? Nobody’s advocating that (as far as I can tell.)
    Should players die from stupid choices? Yes… most of the time.
    Should players be given opportunities to have foreknowledge and make plans? Of course they should… if they choose to do so.

    But should the GM be bound to adhere to the almighty results of dice and never place their agency above the players? I say no… so long as:

    The players are never quite sure it is being done
    It is not done every time
    it is never done to punish a player

    As for the escapist article:
    GMs rule on the scope and nature of player agency all day long. GMs interpret the results of dice and move drama and adjust hazards on the fly in order to challenge as well as reward in the name of enjoyment at every single session. GMs fudge dice, bend rules, make judgements all game long.
    All in an effort to foster fun.
    And while creating fun isn’t the job of the GM, as the player at the table with the greatest authority, the GM must always be mindful of what fun means to the rest of the participants and vigilant to make an environment where the framework for fun exists. Doing anything otherwise is to find yourself railroading scenes in violation of player agency of another kind.

    Let’s take another (probably more common) example:

    Players are adventuring in a module and break into a room pre-stocked with an ogre and half a dozen goblins.
    The characters were pushing into a new region of the map in search of some treasure, got turned around and were attempting to find a safe place to hold up while they rest.
    Realizing that the combined might of the ogre plus the goblins will likely be way more than the characters can handle, the GM has already decided to downgrade the ogre to a hobgoblin and reduce the goblin numbers.
    Has the GM violated player agency by not letting their decision to enter the room lead to a TPK?

    See, in my mind, fostering fun is a balance of interpretation between what the players desire, what the dice indicate, what the rules direct, and what the GM feels is reasonable.
    Player agency is no more the ultimate ruler of the game any more than the dictates of the dice are. Were that the case, we’d just remove the GM and dice and rules and just let all the players decide the outcome of the situations they created.

    Thanks for taking the time to comment and join the discussion here as well.

  6. Kevin says:

    @WB:

    Addressing your concerns:
    – I left the GM out of the discussion since it is this player’s methods that introduces the genesis of the possible quantum ogre. By definition the GM scopes the environment and is reasonably aware and in control of all Quantum Events. To state that one specific representation of such an occurrence will necessarily rob this player of fun is a bit overstated (IMO.)
    – I’ll grant that the example doesn’t address the point of the challenge if it is easily (assuredly) defeated. I appreciate that you can see past such an omission to the implication of the example.

    I hold to the core post response that the Quantum Ogre by definition is relativistic in that if the players aren’t aware of the possibility that it might not exist, it isn’t by definition quanta in their view.

  7. -C says:

    Has the GM violated player agency by not letting their decision to enter the room lead to a TPK?

    I would say yes. I would describe the choices that the characters made to get into this situation poor (got turned around – how did that happen?).

    If you always do this – always change the TPK around so that it never happens, then the players will never encounter a fight they cannot win. And ergo, never approach situations with caution.

    The error is in your basic thesis, which is that, even after discovering the ogre and goblins that they should continue to engage in combat against overwhelming odds leading to a TPK.

    If the game were just an endless series of encounters that I could handle, week after week, where’s the freedom to fail?

    I exerted my agency when I put the ogres there during design. I poured all the cool and awesome out onto the page. Once play begins, it’s time for me to stop changing things around and let the players be responsible for their own success or failure. Winning in that situation is fun.

  8. Kevin says:

    But I never indicated that this is done all the time. In fact, I’ll reiterate what I posted earlier in my response to you (this time with emphasis):

    Should a GM constantly palette shift? No.
    Should a GM ignore all the dice rolls? Of course not.
    Should a GM routinely remove meaning from choice? Nobody’s advocating that (as far as I can tell.)
    Should players die from stupid choices? Yes… most of the time.
    Should players be given opportunities to have foreknowledge and make plans? Of course they should… if they choose to do so.

    But should the GM be bound to adhere to the almighty results of dice and never place their agency above the players? I say no.

    Nowhere did I indicate that this should be the modus operendi for every encounter.
    Nor did I indicate that the game should be a string of endless encounters that the characters can handle.
    In fact, I’m on the record expressly against such a methodology. See post referencing my annoyance with putting your game on training wheels here.

    You appear to want to argue the assumptions you want to bring to the argument while either missing, or purposefully ignoring, the core argument that from the player side of the screen, no palette shift can occur without knowledge of an alternate outcome.

    I’ll state it again: The Fallacy of the Quantum Ogre is inherent in it’s Relativistic Nature.
    If there is no knowledge that it occurred, one cannot claim that it even exists.
    Quite simply this is a classic example of perception being reality.

  9. WB says:

    Kevin,
    I much prefer your alternative example.
    Understood. My broader concern on not mentioning the GM is that often people forget to consider them as “players” of the game whose “fun” is also important. Putting this in terms your argument, that because one “player” (the GM) DOES always know then sometimes (and only sometimes) this means your argument does not follow if you consider this broader context. I’d personally be bored if I ran a one hour combat that had no point, even if the other players did not know. Anyway, its a fine point. I can see why you skipped over it and I think readers can determine from these posts if they are interested, so I’m happy to let it rest there.

    Perhaps I can present a third example that might help people see more readily where you are coming from:
    The group prefers balanced encounters and the game rules they are using support this. The GM sets up an encounter with an Ogre and some Goblins, which is balanced to the party which the party chooses to go and combat a group of goblins lead by “a large humanoid”. Immediately before the encounter two players are unavailable but the other players are keen to play on. The GM shifts the Ogre to a Hobgoblin without the other player’s knowledge so the encounter is now balanced.

    In this scenario the GM is shifting without the players knowledge, but clearly moving to something they would be in favour of had they known. (It also doesn’t affect the GMs fun in all likelihood.)

    WB

  10. Kevin says:

    That’s a much better example. Much appreciated. If the hobgoblin + goblins is the equivalent challenge for the new party as the ogre + goblins was for the original, then the concept of shifting to a known outcome is eliminated from the discussion as well.

    However, I still hold that the Quantum Ogre is a relativistic issue at best.

  11. -C says:

    I didn’t miss those comments, I just think it’s a case of slippery slope.

    Let’s say we’re playing chess. It’s a game with rules, no? And lets say I want to make sure you have fun, so I decide I often won’t make a good move. You don’t know when I’m not making the best move I can.

    Do you think you have more fun beating me at chess that way, then if I play my best?

    Maybe if you’re seven.

    I view D&D the same way. They are playing a game, and if I cheat, fudge or change things whether they know or not then eventually (because I play with smart people) they will figure it out and have less fun.

    As a player, my gut reaction this GM agency during play is training wheels and irks me to no end. As a DM, I really think the most fun my players can have is sink or swim, because it’s not that hard to swim, and when they do it – they enjoy it that much more because they are right near the edge of failure, and it’s their responsibility to avoid it. i.e. they are invested in their success, because they have agency in it.

    If you change the rules where do you stop? How do you know each time you are making a decision to save a players life that you are doing it because of bad luck, and not because they lacked the skill to avoid a situation where bad luck could kill them? At some point it falls down to the whims of what the DM thinks – and truthfully, at least for me, I’d rather just leave it up to the play of the game and the dice.

  12. Kevin says:

    Again with your analogy, you’re making the assumption that you know I’m not playing to win. While that may be apparent in the chess discussion (though it may not since it may simply be the case I didn’t see the move or perhaps have a different strategy in mind) in the analogy of the Quantum Ogre argument as presented, there’s no possible means to know if the decision to enter wood A led to the ogre encounter or not (or interestingly if some prior decision seemingly unrelated did so instead.)

    That simple change in the assumptions of the argument illustrates the relativistic nature of the event. From the player perspective the ogre is and was the outcome of the decision to enter wood A. Since we can’t go back and re-run the decision, there’s no way to claim that the decision didn’t lead to the outcome.

    From the player reference, each wood is the superposition of possible ogre and possible not ogre. Only when they enter the wood does the superposition condense to one or the other regardless of whether the GM has palette shifted the ogre’s location because the players don’t know the answer to the question (ogre or not?) until the decision is made (enter wood A.)

    As for “if you change the rules where do you stop?”
    Simple: where the group feels it’s appropriate and fosters the most fun.

  13. -C says:

    I am enjoying this discussion, and I think you bring up good points.

    How can I not know you’re throwing the game?

    I guess our experiences are different. I have often been a player and become aware of the fact that the DM was fudging the dice. I work in psych, and frankly people are very bad at lying. What’s more, RPG’s tend to pick up intelligent types who are better than average at picking up lying (at least, in my experience, with my friends).

    Now does lying work in games? Certainly, when the burden of telling the lie is removed from the participant, such as in balderdash. But in general, I disagree that the knowledge can stay hidden. How can I not know, (or at least rapidly figure out) that you aren’t throwing the game? And what’s more, as a player if I find out, I’d be doubly mad – first that you were tricking me, and second that you were removing my agency. ;-) I personally don’t think this information can remain hidden for long, if at all.

    As to the second, if you are restricting player knowledge of when you are changing the rules – how can the group provide input on when it feels appropriate?

  14. Kevin says:

    I too am enjoying the discussions. Hopefully I don’t step on anyone’s toes with my current post.

    You may presume I’m throwing the game, but then you may also consider that perhaps I’m a lesser player than indicated or that I’ve seen something you did not consider. The point is, you can’t know unless I divulge the information. That’s the crux of the argument. Devoid of that info, you’re left with no other reasonable result than to assume that Choice A led to Event B. Any other result requires you to make assumptions.

    If the information that I’m keeping hidden is as simple as where the ogre is (or isn’t) or what the dice indicated, then these are easily withheld.
    And while I’ll agree that perhaps some may be unable to hide such things, just as some will never be able to figure it out, in general, small lies (either outright or ones of omission) are kept from the players as a matter of normalcy in the game when you’re the GM – it comes with the territory. Good GMs know when to tip their hand and let such a lie be known, and when to play the situation close to their chest… forever if need be.

    But without that info being passed to the player, there can be no Quantum Ogre.

  15. You say that palette shifting is only recognizable as such when it is observed. Thus, if it cannot be observed, a fun encounter with an ogre happens without players feeling robbed of their agency.

    This is true. But when I read the Hack & Slash post, I felt there was an additional point to be made: I’d have more fun learning something about the surrounding area and the whereabouts of ogres and then deciding for myself whether I wanted to engage it or not – even if the encounter with the ogre itself is very entertaining. That, I feel, is empowering. That, I feel, makes me aware of the fact that I’m not being railroaded. My choice matters.

    Railroading happens when players can only choose to go forward and meet the ogre. This can be very entertaining none the less because the encounter with the ogre has the potential of great fun being had by all.

    Being able to make a random choice without information to base it on and then meeting the ogre anyway is just as good. Players have the illusion of choice and encounter the ogre. The encounter with the ogre has the potential of great fun being had by all. But as a player, I can’t tell whether I’ve been railroaded. I made some choice and somehow I ended up in an encounter with an ogre. I don’t know how to feel about that.

    Awesome player agency happens if players learn that ogres are in one direction and no ogres are in the other direction. Then they can choose to encounter the ogre, or they can choose to go the other way. Even though the encounter with the ogre has the potential of great fun being had by all, offering the choice promises an independent and different kind of joy to some players (like me).

  16. Lior Wehrli says:

    Lets assume the players do not notice that the encounter with the Ogre was forced by the GM and they did not notice that the GM fudged the dice during the encounter and lets in fact assume that they had great fun playing out the scene. Great! Good for everybody.

    Now, what about all those encounters where the players did notice that the GM set them up because its one of those Important Plot Moments again? What about all the time the players did notice that the GM fudged the dice to let the bad guy get away As Required By The Plot? And what did that do to the fun at the table?

    Its moot to discuss whether a little cheating is OK or not when the only example stipulates that everything worked out fine. Its just a theoretical question of Ethics then.

    But we all know that in a lot of cases the players notice when the GM uses force to save The Plot from dissolving. Some groups are OK with that in a sort of “That is how you play D&D” way, at least up to a certain point. But you can GM a game differently once you drop the notion of a prepped plot or of prepped scenes. Sandboxes are one way to do it.

    So: Better not to put yourself into a position where you have to force the ogre on the PCs by planning a particular plot or a particular Grand Moment. Not because its theoretically evil to do it but because it does not work well (they often notice) and because it disturbs the player’s investment in the fictional world (when they notice it makes them doubt the causality of the game world). Much better to use sandbox techniques, impro, random encounters, kickers & bangs, conflict webs – whatever floats your boat.

  17. Kevin says:

    Let’s take the example of the ogre in the wood and give it some life.

    Players are once again embarking into the woods.
    Being smart and cautious players who know that information is the real currency of the game, and details help to prevent a GM from palette shifting, their characters speak to an individual who they presume (on their own without being influenced by the GM) is knowledgeable of the surrounding area.
    Said individual informs them that deep in the woods, past three creeks and the gladed hollow the track will branch into three directions.
    The NPC further claims he’s heard tell of a giant man, wild and ravenous who lurks in the western-most portion of the trail.
    The players smile knowingly and prepare their characters to meet (or avoid) an encounter with an ogre-like creature.

    Here we have a myriad of possibilities that can occur:

    The informant may be lying for one reason or another.
    The informant’s knowledge of the woods could be incorrect.
    The description of the creature could be wrong.
    There may be more than one creature.
    The creature may be sleeping when the characters arrive.
    The creature may live in a cave in the deep woods.
    The creature may be friendly.
    The creature may not even exist.
    The creature may be the pet or minion of something else.
    The creature may be an trick to keep people away.
    The characters may face dangers before reaching the creature and have to turn back.
    The characters may choose to avoid the creature altogether.
    etc, and the list goes on and on.

    At the very moment before the characters enter the woods and meet the creature, from the player perspective, the quantum event is the superposition of all possible outcomes.
    The players have no way of knowing how the agency of the GM, combined with that of the dice as an outcome of their own choices will manifest. At this point, each possibility is probable.

    Now let’s presume that weeks ago, when designing the woods, the GM exerted agency and decided the following:

    For simplicity sake lets say:
    Bob the NPC isn’t lying.
    Bob correctly identifies the creature as an ogre-like giant man.
    The creature is in fact an Ogre.
    The creature is not friendly.
    The creature is a solitary beast.

    However, having creatures who just sit in their den waiting for characters to come along, kill them and take their stuff is so 80′s, so the GM further decides to give the Ogre life:
    The Ogre as an encounter is time specific.
    During the middle parts of the day, the Ogre spends time sleeping in a small cave at the base of a muddy ravine.
    The entrance to the cave is hidden from view, but the stench from years of an Ogre’s activities make finding it challenging but not too difficult if someone is in the area.
    During the cool of the day (twilight or dusk) the Ogre will 50% of the time be found at the cave (either preparing to go/returning from a hunting trip) and 50% of the time wandering the woods seeking food.
    Should the Ogre be wandering the woods in the cool of the day, it will only be encountered as a wandering monster.
    During the night the Ogre will be hunting in the woods and will be encountered 25% of the time no matter which track the characters have taken.
    As the Ogre is quite knowledgeable about the inhabitants of the woods, each subsequent day the character spend in the woods adds 5% to the chances of meeting it at night.

    Now from the GM perspective, these decisions have collapsed (and even eliminated) some of the possible super-positional possibilities from the quantum event while giving framework for the collapse of others.

    Here’s how the night of play is spent:

    The players spend a few days seeking out the Ogre but the dice never seem to be in their favor.
    Even with the increasing likelihood of a meeting, the agency of the dice are dictating that the Ogre is either unaware of the characters, or prefers to leave them alone in the hopes they’ll travel on.
    Each night’s roll indicates the Ogre must have been hunting in a different portion of the woods and each twilight and dusk rolls yield no wandering foe.
    The players never think to have their characters go find the Ogre’s den where waiting will assuredly bring the Ogre encounter to fruition.
    The players are beginning to show signs of frustration, they really want to face the Ogre and in essence are starting to feel railroaded by the dice and the situation.
    You honestly had no idea that encountering the Ogre would become such a white whale for the group.
    Fun is slipping away as tedium creeps in…

    Ignoring that eventually the Ogre will come (since 15 days of wandering will give the Ogre encounter a 100% chance of occurring at night) what do you do to reinvigorate the fun and how do your actions impact agency around the table?

    If you sit back and let the agencies at the table do their magic (since you put all your agency into the encounter at time of creation) then you’re violating what I contend is the ultimate role of the GM: to foster enjoyment.

    If fostering that enjoyment means you have to nudge and tweak the situation from time-to-time, then that’s the nature of the beast. And if the players never know it happened, then there is no Quantum Ogre.

  18. I’d say: If the players want to meet the ogre, tell them “you search for the oger and after a few days, you find it. ROLL FOR INITIATIVE!” or something along those lines. I don’t care enough about the simulation of the world. Some people might. That’s why the rolling is unnecessary as long as the players a. learned where the ogre is and b. did what was necessary to meet it (or avoid it).

    In addition to that, if information is provided by an informant, it to proves to be incorrect (informant is lying, doesn’t know, etc), then you’ve introduced the possibility of your players no longer bothering to ask NPCs for information. Unless finding the knowledgeable person or unmasking the ogre’s friend in the village is a plot you had in mind, I’d suggest avoiding situations where the players can’t depend on their senses. I think it trains players to mistrust the DM and threatens to break the game. It would for me, at least.

    If all is said and done, however, and you still find yourself in the situation you described, I think you should absolutely nudge and tweak the situation. Just don’t make it impossible for players to make meaningful decisions.

    I really like the Judging the Game blog post by Alexander Macris blog post. There, he says: “change the rules in advance to prevent that sort of situation from happening.” Don’t use rules that require players to spend up to 15 days searching the forest if that’s not an outcome that you like. (At the same time, don’t just provide a single outcome that you do like. Provide more outcomes and meaningful choice.)

  19. @Alex

    Here’s the troubling part (for me) about arguments like yours. For me, it’s not really about simulation but situation. So here’s what I’m going with when I run a game… everything is connected — especially because the PCs are the agents of change in the game-world.

    So — to take you situation — if informants are always honest, then yes, the PCs will always talk to them and learn something useful. However — what happens when an informant is dishonest (or the PCs just picked a crappy informant)? To take the first instance, most parties I’ve played with would head back to town and have an “interview” with the informant in question to wonder why he lied to them. Characters in stories lie… it happens. This is the PCs story and whether you tell it by fudging, letting the dice determine everything, or just “telling the story” people still lie. Heck, maybe the guy was just jealous of the PCs and wanted to see them fail because he used to be the BMOC around this village (which explains why he knows so much about the monster haunted woods)…

    I don’t run games where every NPC is trustworthy because that is just as dishonest as a game where they always lie. It’s a balance. Sometimes PCs need a little prompting. Sometimes they don’t. In the end, if we have fun, so what?

    PS — I’m just curious, at your table (and I really really don’t mean this to be snarky, I’m really serious and curious) is it okay for the GM to have the party ambushed while traveling somewhere or is that a violation of agency?

  20. Things have changed over time at my table. These days, if an information is dishonest, I will make it obvious in my role-play. I’ll add things like “we tries to dodge the subject but finally he mutters…” or “As you leave, you can’t shake the feeling that he seemed happier than he ought have…” Thus, while some players will have their characters walk into the trap anyway because that’s what they enjoy, others might ask, “so the guy’s a liar?” and I might say “well, you don’t know, do you? But he certainly didn’t seem very honest to you.” I use vague terms, but everybody at the table knows that the informant was dishonest.

    What I’m aiming for is what -C is working on with regards to traps. Walking down the corridor and falling into a pit trap is boring. Walking down the corridor and realizing that the monster tracks suddenly disappear right there and wondering whether to turn back, search for traps, jump, fetch wooden planks or pour water is what I enjoy. Similarly, preparing against an ambush and then being ambushed anyway is boring. Knowing that you have been lied to, and wondering whether this is connected to the new found love interest (and he was there yesterday at the inn, now that you think of it), whether he’s jealous of your success (and he did look hurt when you dropped all that gold on his table), and deciding to go anyway but be extra careful — that is what I enjoy.

    I enjoy the premonition of betrayal and the walking into a trap with fair warnings more than the simple story of the betrayal itself.

    That doesn’t mean that characters never get ambushed. At my table, players offended a bandit queen and parted on bad terms. I determined ahead of time that four weeks of in-game time in the future, she’d know where they were and if possible, she’d send dudes to poison her enemy (one of the player characters). There was no informant. Here, player agency was very limited. They had chosen to make an enemy and the enemy started moving against them. The poison also wasn’t lethal and worked as a plot hook for the next adventure. This was a short moment of having no agency at the table.

    I didn’t feel bad because essentially it worked similarly to hearing a rumor. “You have powerful enemies out there.” Player agency returned in that there are ways to discover where the bandit queen is hiding. Players are also free to ignore the threats and move elsewhere. Players are also free to tell me how they will defend against future poison attempts (and they’ll work).

    I appreciated your curiosity. I find it’s more productive to explain how things work at our respective tables than presenting claims and counter claims.

    You also say: “In the end, if we have fun, so what?” I’m not sure what to say. If we both agree that we’re both having fun and so what, the discussion ends. But I’m trying to learn and teach, to exchange ideas in order to have even more fun. I don’t doubt everybody is having fun but as far as I am concerned, the question is: “In the end, if we can have even more fun, why not?”

  21. Kevin says:

    @Alex:

    “I find it’s more productive to explain how things work at our respective tables than presenting claims and counter claims.”

    This. Exactly.
    This is why I decided to respond to the original argument in the first place.
    How I do things isn’t the right way and more than how you do things.

    That said, I’m still convinced the Quantum Ogre is a molehill being perpetrated to be Mount Doom and I’m waiting for the argument that demonstrates to me how I’m wrong.

  22. That doesn’t mean that characters never get ambushed. At my table, players offended a bandit queen and parted on bad terms. I determined ahead of time that four weeks of in-game time in the future, she’d know where they were and if possible, she’d send dudes to poison her enemy (one of the player characters). There was no informant. “Here, player agency was very limited. They had chosen to make an enemy and the enemy started moving against them. The poison also wasn’t lethal and worked as a plot hook for the next adventure. This was a short moment of having no agency at the table.

    I didn’t feel bad because essentially it worked similarly to hearing a rumor. “You have powerful enemies out there.” Player agency returned in that there are ways to discover where the bandit queen is hiding. Players are also free to ignore the threats and move elsewhere. Players are also free to tell me how they will defend against future poison attempts (and they’ll work).”

    –And for me this is the crux of it. I wouldn’t, in my opinion, even say that agency had been compromised at all. The PCs made an enemy. That’s what happens. You decided (as GM) that at some point in the future there would be consequences to their original decision. No foul here — the PCs got to make decisions and NPCs acted (reacted) in appropriate ways.

    Similarly, you say that you make it clearly apparent when an NPC is lying, when the PCs are being betrayed… because you appreciate the premonition of betrayal. Okay, that’s cool. But should they always know? I suppose that begins to enter the territory of the player skill vs character skill debate — but should they always know? I don’t always know when I’m being lied to, do you?

    And don’t get me wrong… I agree with you about learning, it’s why I asked the question. And why I was trying so hard to be sure that it didn’t come across “internet snarky” because I really wanted to know.

  23. Kevin says:

    I think the original Quantum Ogre argument would clearly indicate that the bandit revenge event is in violation of player agency because there was no meaningful choice presented as a way to avoid it.

    However, in my view, since by your own definition the players had no way to know if their actions directly or indirectly led to the event occurring, there was no way to claim their agency had been impacted negatively… even if that’s the “real” case from the GMs viewpoint.

    And this is where the relativistic nature of the discussion makes for some fascinating conundrums that are incongruous yet perfectly valid since from the player’s perspective, reality shows that their choices led to X while from the GMs view X is independent of these choices.

    But the deeper question as to if the bandit revenge event robbed the players of fun is the real crux of the argument.

  24. You are right in that the bandit revenge event is in violation of player agency because there was no meaningful choice presented as a way to avoid it. Then again, the event was really short compared to a fancy ogre encounter. It depends on whether we are talking two minutes or thirty minutes. The more time at the table an unavoidable event takes, the more its inevitability irks me. Now that I think of it, that’s my main problem.

    I think that two minutes of rail-roading are not bad if it introduces a nemesis that enables players to make a lot of choices for many sessions to come. On the other hand, I think that an entire hour (or an entire session!) of rail-roading (such as unavoidable ogre encounters) are bad. In recent years, I think my patience in these things has dropped. I no longer want long expositions, long character backgrounds, long fights and long stretches of time where I can’t make meaningful choices as a player.

    Perhaps this personal preference of mine explains why I keep returning to the discussion. Right now I feel like playing some games and seeing whether I can bring this stuff to the table! :)

  25. Lior Wehrli says:

    Ignoring that eventually the Ogre will come (since 15 days of wandering will give the Ogre encounter a 100% chance of occurring at night) what do you do to reinvigorate the fun and how do your actions impact agency around the table?

    I would never for the life of me set my game up like that. Basically, in your example, you set yourself up for failure. I would spend infinitely less time on a the ogre’s life schedule (I don’t really care when he does his pedicure or what sort of tea he buys at the grocery) and I would spend way more time thinking about what would make the ogre interesting to meet and whether in the end he is worth the bother of centering a game evening around him or even of having him in the game world at all.

    If fostering that enjoyment means you have to nudge and tweak the situation from time-to-time, then that’s the nature of the beast. And if the players never know it happened, then there is no Quantum Ogre.

    Here’s my point: avoid even going in his direction, so you never get into the situation where you have to nudge or tweak. That may not be possible to 100% in D&D… Still, prepped encounters should be designed not to require bending the rules. For your example: If the ogre is really worth it, make clear to your players that they can find him (you don’t have to spell it out bluntly, your NPC is more than enough). Make it so that if they want to then its all about *how* they will meet him.

    How will that “how” be decided? You should have no trouble improvising a reaction to whatever outrageous plan the players come up with based on your encounter prepping – as long as your material is rich enough to drive your imagination. Voila: you have your prepped ogre, the players have their tactical planning and a lot less dice had to be rolled than in your example. (This solution may not be to the liking of -C who might prefer a fixed location where the ogre is always to be found in one of several states)

    The important thing here is that even if you only have the ogre on offer (the players have no other choice but encountering him) the way that encounter goes is decisively negotiated between the GM and the players. The players have agency and it matters a lot.

    @Rhetorical Gamer: “it’s not really about simulation but situation”, “the PCs are the agents of change in the game-world”… I hear you and I’m with you on this. You and me, we are Story Now guys. But this is essentially a Game On Up vs Right To Dream discussion so we are out of place in this. Or at least that is my take of it. ;-)

  26. Kevin says:

    I would never for the life of me set my game up like that. Basically, in your example, you set yourself up for failure.

    You’re avoiding the question posed. I know I’ve set myself up for failure in the example, that’s the point.
    But who among us hasn’t sat behind the screen knowing things are heading down the tedium path…

    The question is, should you find yourself at the table in such a circumstance what can you do but violate player agency or fudge a roll?

    Of course I’ll admit it’s rhetorical (and probably why you tried to skirt it) since there’s no other real option if you want to keep the game fun.

    In that example, the GM is forced to violate agency in order to preserve fun, not rob the game of it.

  27. Lior Wehrli says:

    But the deeper question as to if the bandit revenge event robbed the players of fun is the real crux of the argument.

    I am just guessing here…

    That depends on what the players have fun with. Do they enjoy the tactics of a big fight, hatching and executing a plan? If so, then they are likely to complain about not having had the chance to use their perception skills to out-smart the ambushers. If on the other hand they enjoy playing through the consequences of their dissing the bandit queen they might accept the unavoidable ambush because it makes sense. And if they like the drama of almost dying because of rashly made arrogant remarks their PCs made weeks ago? Those players could love it.

  28. Lior Wehrli says:

    Of course I’ll admit it’s rhetorical (and probably why you tried to skirt it) since there’s no other real option if you want to keep the game fun.

    This question is not only rhetorical, its also boring and that is why I assumed you are not asking that question. More interesting is the question of how not to get to this position in the first place.

  29. Kevin says:

    I suppose I could create another example eliminating the design flaw of the situation if you’d prefer, but I’ll admit to growing weary of examples.

    The question illustrates that sometimes palette shifting is required in order to foster fun even at the expense of player agency since palette shifting (moving encounters around, adjusting creature strength in battle, fudging dice, etc.) is simply a tool the GM uses in order to keep the game enjoyable.

    And if that shift occurs without player knowledge, then once again… there is no quantum ogre because the players don’t know their choices were meaningless.

  30. -C says:

    The players are beginning to show signs of frustration, they really want to face the Ogre and in essence are starting to feel railroaded by the dice and the situation.

    Several times I’ve seen this assumption in these examples where this situation is set up and something bad happens due to the DM allowing player agency and the players using it in such a way that it makes them unhappy due to the dice. First with the ‘we don’t pay attention and get lost and get into a fight we can’t win’, second with ‘we fight the fight we can’t win to the death’, and third with ‘we rely on the dice to accomplish our task’

    The key here is, if you consistently give them agency, they will become skilled at avoiding letting the dice decide things that are crucial. Player agency means that they have the freedom to be boring or fail. That is ok. They will be happier for it.

    This makes me so angry!
    palette shifting is required in order to foster fun even at the expense of player agency since palette shifting (moving encounters around, adjusting creature strength in battle, fudging dice, etc.) is simply a tool the GM uses in order to keep the game enjoyable.

    The problem comes in this!

    Losing a character makes me sad! But knowing I can lose a PC makes the game matter!

    When the GM fudges to keep the game fun, they are trading a feeling of actual value as an activity that truly matters, authentic, for more one of just going through the motions. Something that’s fun for the moment, but in the end is no different then playing a video game – nothing that really matters. *note: edited by Kevin to keep the blog kid-friendly*

    As far as the bandit queen encounter is concerned, the players had agency when they made an enemy.

  31. Kevin says:

    This discussion has ensorcelled additional bloggers in our field and I wanted those coming here to be able to reference/join in on the continuing conversations there…

    http://monstersandmanuals.blogspot.com/2011/09/faking-it-or-youd-better-be-al-pacino.html

    and

    http://rolesrules.blogspot.com/2011/09/grand-inquisitors-roleplaying-game.html

    and

    http://thegrandtapestry.blogspot.com/2011/09/fallacy-of-quantum-ogre-argument-korpg.html

    If anyone finds others, please bring them to our attention.

  32. Matthew Miller says:

    Your argument seems to boil down to: what the players don’t know can’t hurt them (or rather, can’t rob them of their fun).

    I would argue that players are generally more aware of when dice rolls are being ignored than you give them credit for. I can nearly always tell when I’m dealing with a judge who is routinely fudging dice rolls, etc. Maybe an otherwise “honest” judge can get away with this once in a while, but that’s a dangerous road they’re heading down. If they persist in the habit (“just one more drink… won’t make me an alcoholic…”) they run the risk of eroding trust when the players begin to suspect what’s going on. Better to “Just Say No” from the beginning.

  33. Kevin says:

    I suppose that’s a fairly accurate restatement of my argument.
    Though I would prefer the more precise: If the players don’t know their choices were false, then with no evidence to the contrary they must presume their choices were meaningful.

    And while we’ve wandered the terrain of “how much a player will know”, the point of the fallacy is that if they don’t, then the illusion of meaningful choice (false choice) becomes the reality of meaningful choice from their perspective.

  34. @Kevin: “Perhaps. But as a referee, it is the GMs role in the game to interpret the results of agency and palette shift the fun elements into the forefront, not to sit idly by and just see what happens.”

    You’re begging the question pretty hard there.

    If you start with the assumption that the GM’s job is to palette shift, then it’s really unsurprising when you conclude that the GM is justified in palette shifting.

    More generally, I think your example of combat with the ogre is conflating illusionism (“I’m going to present you with a false choice; no matter what you choose, it will result in the outcome I predetermined”), dice-fudging (“I’m going to modify the mechanical outcome of this action resolution in order to change the outcome to something I think will be preferable”), and script immunity (“you can’t die, no matter what you choose or how the dice roll”). While people with a particular taste for “let’s see what happens, good or bad” are going to be unified in their dislike for all three of these things, they’re actually very different things. And attempting to conflate all three of them into being a single thing is, IMO, needlessly muddying the waters of understanding.

    In specific regards to the quantum ogre/illusionism: The argument that it doesn’t matter whether or not a choice is predetermined if the players have no meaningful context for making the choice is probably true. It’s also irrelevant on any practical level because (a) such choices will rarely occur in any sort of meaningful game and (b) even in the absence of relevant information, the decision not to seek out such information is an inherently meaningful choice (unless, of course, the outcome is predetermined).

    For example, consider the Caves of Chaos from B2 Keep on the Borderlands. The argument can be made that choosing which cave to enter first is not a meaningful choice because the players have no idea what lies within each cave. But, in actual practice, is quickly falls apart: The players could stake out the valley and watch to see which monsters come in and out of which caves. And even if they don’t do that and just pick a cave at random, this first choice will quickly result in them learning more and more about the other caves (as reinforcements are sought or prisoners are questioned or whatever). In short, when it comes to actual gameplay, the majority of choices made by players will not be completely blind.

    In other words, this hypothetical scenario in which the quantum ogre isn’t significant because nobody knows (or can know) what’s happening inside the boxes that may (or may not) be holding the quantum ogre is only plausible because it IS a hypothetical scenario.

  35. Kevin says:

    Justin:

    Great comment. Thanks for taking the time.

    Like a great moment of serendipity, I had just concluded my Quantum Ogre test on Saturday night.
    I’d like everyone to take a look at the results of my experiment.
    Summoning the Quantum Ogre

    As a result of my findings and my previous position, I find myself both agreeing with you on your premise, and still disagreeing on the outcome a bit.

    I never presumed that all player decisions are blind and devoid of consequence. As a result I’ll agree that’s a terrible metric by which to test the scenario. If I gave the indication that’s the case, then my bad.

    However, in all cases where choice is made, player knowledge is by nature less than complete (and always somewhat blind.) That’s the nature of the game we play and and that’s where the Quantum Ogre rears it’s relativistic head. And while the consequence of player decision helps to shape the probabilities available to future events by collapsing the probability envelope of those events, players also can’t know the scope of future events from which a a GM is working.

    As a result, the only material the player has to base their point of reference as to the outcome of their decisions is what they decided and what resulted.

    Where your argument against the scenario being only valid in a hypothetical situation is that you’re bringing in external information to the table and claiming it helps you determine when the illusion has been cast.

    I’ll say again that devoid of any information to the contrary a player cannot know that you’ve cast an illusion.

  36. Ozymandias says:

    “I’ll say again that devoid of any information to the contrary a player cannot know that you’ve cast an illusion.”

    A situation that only happens in a vacuum. Unless you are playing a game with people who have no familiarity with the DM at all, it is highly unlikely that a player will completely overlook the clues that give away the illusion.,

    Consider an experiment: set up a video camera and tape yourself telling a series of lies to someone close to you. Someone important. Someone who doesn’t know that you’re taping your lies in the moment. And make sure that the lies you’re telling have real consequences. Then, after you’ve apologized and promised to make amends, watch the tape together. Have the other person show you the moment where they knew you were lying.

    You might be surprised.

    It is not possible – barring the extreme example of a new gaming group where not one player is familiar with the DM – to present a situation where each player is denied the sort of interpersonal cues we are inundated with every day of our lives.

    Someone at the table will know you fudged those dice…

  37. Kevin says:

    That situation happens even outside the vacuum. DMs hide and obfuscate information all the time in their general role as DM.

    And while I’ll grant if some single bit of illusion and pallet shifting is done repeatedly and/or any generic shifting is done all the time, the gig is up and you’re likely to draw the ire of all your players (see The inherent problem with making your fantasy world one Big ol’ Super Random, Character Killing, mondo, MEGA-DUNGEON of DOOM!) But when done well, even with the inclusion of a tell, the reason behind the tell won’t immediately be readily identifiable as specific to a shift.

    Without direct knowledge, players cannot know you’ve cast the illusion… they may suspect, but they can’t know. That’s what keeps the ogre relativistic in nature.

  38. zuv says:

    OK. Now take a deep breath and go back and read Hack and Slash again. You clearly did not understand any of it.

  39. Heh, that’s funny… You do realize that if you just disagree (or think -C has the right of it and I don’t) you can state your case and I’ll be willing to have a good detailed discussion where my opinion on how RPGs play should be structured vs. how those who adhere to the Quantum Ogre belief obviously seem to feel the play should be structured.

    I believe I’ve demonstrated that is fair fashion here and in other posts.

    Now, while its altogether possible I’ve not understood any of the issue, I find that simplistic (and somewhat baiting) premise hard to accept given that -C was the one over here discussing and arguing the situation with me. Clearly the ultimate separation between the two of us is whether a GM cheats the players if he or she alters the dictates of the rules, dice, choices, or events for an alternate outcome.

    Its my belief that certain alteration is good and necessary for fostering a more enjoyable game.
    Those who fear the Quantum Ogre naturally disagree.

    So tell me, where exactly have I “clearly not understood” any of this?

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>