How can I calculate the probability of being able to purchase a card in my custom die system? has had an interesting life so far:

  • closed (3 votes, incl. a diamond mod) as unclear
  • bunch of comments asking for clarification on the die-system being described (now moved to chat)
  • edits to better explain the system
  • (automatically) placed in queue for reopen review, which looks like it was rejected
  • migration to B&CG suggested (which I declined, not being sure it'd be on-topic there)
  • further editing for readability
  • more and more comments about whether it's on- or off-topic
  • comment-discussion about how to program it in anydice, moved to chat
  • locked
  • unlocked
  • reopened by 5 regular users
  • closed by me, because it sure looks off-topic to me and I'd like to hear why it's on-topic

[this narrative assumes I'm interpreting the "timeline" (extremely obscure tool) correctly, which may also be a problem--others please feel free to correct my read!]

So, have at it: on-topic, or off?

[update] I've reopened the question to favor a "default: leave it open" policy. Too much waffling--someone should take my hands off the mod-stick!

  • \$\begingroup\$ @doppelgreener--good edit. I certainly wasn't intending a class-wide ruling, just this particular question. \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 10, 2017 at 1:05
  • 2
    \$\begingroup\$ The lock-unlock can probably be disregarded: I was isolating certain comments for moving to chat and that was just a precaution while I was doing deleted-undelete gymnastics with the other comments, so it's not really the question's doing. \$\endgroup\$ Commented Jul 10, 2017 at 6:11
  • \$\begingroup\$ Note: the timeline isn't really mod-only as I understand it, it's just that normal users have to use a userscript or manual link-hackery to get to the link. (The first reopen review item was invalidated by SSD's lock, by the timestamps.) \$\endgroup\$
    – user17995
    Commented Jul 11, 2017 at 3:15
  • 2
    \$\begingroup\$ (That is, rpg.stackexchange.com/posts/103090/timeline shows me a summary of everything pretty much as described.) \$\endgroup\$
    – user17995
    Commented Jul 11, 2017 at 3:25
  • \$\begingroup\$ @TuggyNE iiiiiiinteresting--good to know. Will edit accordingly =) \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 11, 2017 at 3:36

3 Answers 3


This question has precedent

BESW asked a question about probabilities with coins, for the purpose of game design, with 33 upvotes and 6 answers. That Q received no VTC's or controversy, but it is effectively the same type of question as this Q under fire: I am designing a game with randomness built in, what are the probabilities for this aspect of it?

WrongOnTheInternet has asked about a purely statistics-type question concerning exploding dice, receiving 11 upvotes and 3 highly technical answers, but is again on the same lines: I am designing a system, but how do I calculate the probabilities?

HeyICanChan asked about calculating the probabilities with Anydice on a dice-based system based on multiple types of dice, which is a Q along similar lines: here is a system, how can I calculate the probabilities?

A new user has also asked for a mostly similar Q about formulas for permutations, another new user asked for help about using Anydice, another new user asks about exploding dice probabilities. All three of these questions have in common that they are not attached to any specific gaming system, and are actually purely statistics-based questions.

If this one Q is off-topic, the the six listed above should also be closed, for consistency's sake.

These types of Q's are on-topic

Leaning on historical precedent, there has been no push back for the above six questions in terms of VTC's. Nothing has changed about the sites policies about these types of statistics Q's in the intervening time, and so since those 6 Q's did not receive the VTC or on/off-topic controversy, this Q shouldn't be off-topic either.

While the Q is not about a problem in the sense of "X happened, what do I do?" or a rules clarification for an RPG system (the typical questions), it is still a Q related to an RPG -- just an unpublished, untested one.

Finally, the presence of the statistics tag allows these types of questions to be asked. If a policy change happens, then we will have to move the statistics tag to be off-topic as well, in the same way that shopping questions were made off-topic.

  • \$\begingroup\$ So I'm with you that there's precedent, but it seems like there's a little dodging of the actual question: yes, if those are on topic than this one is, but do you have a feeling/argument for those/this being on topic or not? \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 10, 2017 at 12:43
  • \$\begingroup\$ @nitsua60 I think it's on topic, and updated my A to reflect that \$\endgroup\$
    – user27327
    Commented Jul 10, 2017 at 12:59
  • 1
    \$\begingroup\$ For obvious reasons, I agree and upvoted this answer, but I think it's fair to say that taking such questions to Math SE or elsewhere seems a heavy-handed intervention that unhooks the unasked question — Is this a reasonable RPG mechanic? — from this site's users area of expertise. An answer that addresses How can I compute this? and also explains what impact those probabilities have on the RPG should be found superior to those that don't address the latter. (That's what makes this answer to my question particularly good, for instance.) \$\endgroup\$ Commented Jul 10, 2017 at 20:19
  • \$\begingroup\$ Mere statistics and dice probabilities questions are still off topic — we don't want to become a magnet for stats/dice questions with nothing to do with RPGs — but I would agree that the (clarified) RPG context of the question redeems it just enough to belong here. \$\endgroup\$ Commented Jul 10, 2017 at 22:26
  • \$\begingroup\$ @SevenSidedDie That would make the last 3 example Q's off-topic. Two of them are a few years old already though, and one was recently asked \$\endgroup\$
    – user27327
    Commented Jul 11, 2017 at 0:21
  • \$\begingroup\$ @markovchain Would it? It seems that the emergent intelligence of collective voting hasn't detected a topic problem with them. \$\endgroup\$ Commented Jul 13, 2017 at 16:30
  • \$\begingroup\$ @SevenSidedDie The last 3 Q's are purely stats questions, with no ties to any RPG system. The emergent intelligence of collective voting hasn't detected a problem with them, which means they are fine. Therefore, questions that are purely about stats, with no ties to any RPG questions, are fine. I am contradicting what you said, specifically: "Mere statistics and dice probabilities questions are still off topic" because if you are correct, then you must also declare the last 3 Q's off-topic and claim the emergent intelligence of collective voting had it wrong, thrice. \$\endgroup\$
    – user27327
    Commented Jul 13, 2017 at 16:41
  • \$\begingroup\$ The emergent intelligence isn't perfect and I'm not claiming it is. What I'm saying is that you can't blatantly contradict our help pages about topicality in order to construct your argument. If there are unclosed questions that you think violate our topic rules, then there is an excluded middle error in your reasoning, because you're overlooking part of the system or oversimplifying the model we actually use. Do not use them to construct arguments that satisfy simplistic first-order logic while violating higher-order pragmatics. (And if you think they're off topic — vote so.) \$\endgroup\$ Commented Jul 13, 2017 at 16:48
  • \$\begingroup\$ @SevenSidedDie I am pointing out the error in your reasoning. On the one hand, you said that class of questions are off-topic. On the other, you (not I) endorsed the emergent intelligence, which says there was no problem with those last three Q's. Those two positions are incompatible. I happen to think they are on-topic. As the one between us who believes those last three Q's, which fall under the category of "mere statistics and dice probabilities questions," as being off topic, I believe it is you who must consider your vote. \$\endgroup\$
    – user27327
    Commented Jul 13, 2017 at 17:03
  • \$\begingroup\$ @markovchain Excluded middle error. You are arguing that either “mere statistics and dice probabilities questions”, or that those questions should be forcefully closed. You're ignoring several possibilities that allow both things to be compatible in the moment, such as these two: that open/close state is changeable and rarely/never 100% aligned with what the open/close status of the whole site's questions should be if the site were static; that your own judgement is not ruling on their close state, and you might be wrong (your judgement is only ruling on your decision to close vote). \$\endgroup\$ Commented Jul 13, 2017 at 17:28
  • \$\begingroup\$ Consider this: it is axiomatic that mere statistics and dice probabilities questions are off topic. If you think that those questions are on-topic but violate that axiom, then either your understanding of our topic scope is in error, or you need to revise your judgement of their on-topic-ness and vote them closed. \$\endgroup\$ Commented Jul 13, 2017 at 17:30
  • \$\begingroup\$ @SevenSidedDie I couldn't clearly understand what you were saying on your lengthier reply. I am saying that these two are mutually exclusive: "the last three Q's are mere stats questions, and therefore off topic" and "the collective intelligence has not seen a problem with the last three Q's, and thus sees them as on-topic." Either the last three Q's are off topic, or the collective intelligence was wrong. You seemed to endorse both of these mutually exclusive ideas. Let me ground this in something concrete. Do you agree that the last three Q's are mere stats questions? \$\endgroup\$
    – user27327
    Commented Jul 13, 2017 at 17:39
  • \$\begingroup\$ @markovchain This isn't binary; I see them as troubling borderline questions that cause problems for out sense of topicality, which I expect will take a while to shake out decisively. (The collective intelligence is not always fast.) This situation is too complex to apply simple propositional logic to, as it is insufficiently expressive to capture the state of the site and policy. (Properly modelling it would require jumping several orders of logic up to pragmatics, but we don't have time to actually model it rigorously. Just: propositional logic is too weak here.) \$\endgroup\$ Commented Jul 13, 2017 at 17:45
  • \$\begingroup\$ @SevenSidedDie Let me try evidence-based logic. The oldest of those three Q's is almost 3 years old. You seem to be implying the collective intelligence has not yet decided on the validity of those questions. Consider that almost 3 years ago, a question of a certain family was asked. Two months ago, another question under the same family was asked. Both of them were treated the same way each time: no votes for closure. Is this not a long enough time span? Is the collective intelligence still deciding? Is the collective decision to not close indicative of how it views this family of questions? \$\endgroup\$
    – user27327
    Commented Jul 13, 2017 at 17:56
  • \$\begingroup\$ @markovchain That's not out of the norm for a low-traffic subtopic on the site. Being old is actually not correlated with being on-topic either: check some of our oldest questions, and you'll see that many were closed years later. We refine our understanding of the site's topic scope constantly, and get closer to best practices by doing so. If dice/stats questions have now hit a critical mass that many users are questioning them, then what's likely (from experience) is that a community re-evaluation of existing and future dice/stats questions is approaching. \$\endgroup\$ Commented Jul 13, 2017 at 18:00

When I edited the question to be readable there was no indication that the system had any significant relation to an RPG at all. The OP said in the post at that point:

OK working on a dice game as part of an in game system where players can play a dice Game that requires a players roll specific results in order to gain a card,

Ok its a game within a game

This plus the content of the game made it appear that the game mechanics asked about had no bearing on the RPG they were being played in at all. The question, as it appeared, was "How can I get probabilities for this card game I'm developing?" with the only RPG context being that the OP was planning on having the people playing an RPG with them play this card game, too. The question did not and does not invite commentary on how to integrate such a game or anything like that, it is and was solidly about the probabilities of a card game.

This differentiated it from arbitrary questions about statistics, in my mind. Were the mechanics intended for use in an RPG (as the OP's answer here indicates) our expertise would be applicable, and we could talk about the stochastic characteristics of this method in the context of its use in RPG conflict resolution (well, somebody could. I have trouble doing that with non-math terms).

When we ask a question about arbitrary statistics on this site, it is understood that we are asking about those statistics in an RPG context. That context is necessary for the questions to be on-topic, and the principle of good faith tells us we should assume that's the context the question is in unless there is strong evidence to the contrary. If somebody asks "Are sequential coin flips good random bytestring generators?" here, it's appropriate to respond with "No, because that takes way too much table time", because we're answering from an RPG context.

Now, questions about using tools primarily used for RPGs are, in my opinion, also on-topic here. We have the experts to be able to answer such questions to a high standard, because the experts in those tools have that expertise as an extension of their RPG expertise. Dice are a tool we use extensively, but dice are also used extensively outside RPGs. Asking about dice in general should be on-topic, I think, since they are ubiquitous and quintessential RPG tools, but asking about dice in the specific context of a non-RPG (like, 'what are the odds of me being able to take this space in Risk') is not on topic. This is kinda weird because the underlying question of 'what are the odds that A beats B with this weird dice pool mechanic?' would be on-topic, but somehow because that pool is called 'playing the board game Risk' it becomes off topic. I'm not sure that this division is fair but it is the division that seems most appropriate to me.

Questions about a Board or Card game with no significant RPG context shouldn't be on topic here, even if they are just a question of statistics, even though general questions about statistics with no context are on topic.

That said, the OP made it clear here that these mechanics are intended to be used in the actual game system as well, so I'll edit the question to mention that (since the OP still hasn't), and it should be open.


I am using the same mechanics for both systems one for an custom RPG and the Card game for an in game game in the RPG, as the mechanics for the card game are a simpler function of a more complex system, using them to work-up with the game. Just one is not intending to ask how to fully develop the dice mechanics on the full rpg as that misses the point of asking for help to understand the basics so that can upscale with main game. The Question originally did explain that it was for a both an In game RPG Card game, but was edited down and removed due to the question being to verbose. However it does raise an interesting point about having the 'Anydice' tag and covering Anydice issues, If it is a matter of moving away, then having a Stack Exchange dedicated to AnyDice might work better.

  • 5
    \$\begingroup\$ Yeah, given this explanation I think it's on topic, if not the best fit. This is a pure ol' math/probability question, and you'll get a better answer to it on a more relevant Stack, but being on topic on another stack doesn't make it off topic here (neither does something not being on topic on any other Stack make it on topic here). I'd reopen it. \$\endgroup\$
    – mxyzplk
    Commented Jul 10, 2017 at 1:37
  • \$\begingroup\$ @mxyzplk I wonder how this compares with our "campaign research" question policy? I mean, "pure ol' math/probability" doesn't seem to require RPG expertise, either? However, I know I would like questions like these to be on-topic, so perhaps its time for a class-wide meta on statistics/probability topicality? \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 10, 2017 at 12:45
  • \$\begingroup\$ Well, that's a good point. While e.g. desktop publishing is on topic inasumuch as it applies to RPGs, there's a Venn diagram of whether something really benefits from RPG expertise or not - this one certainly doesn't. \$\endgroup\$
    – mxyzplk
    Commented Jul 10, 2017 at 12:58
  • \$\begingroup\$ In any case, I think this meta about a specific question isn't the right place to hash out the entire class--I've reopened the specific question and have it on my scratch-list to work on a meta that discusses the whole class. Personally, I'm happy having the questions on-site, but I think it'd be good for lots of eyes to see arguments for why these might be on-topic when others are off-. \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 10, 2017 at 13:08
  • \$\begingroup\$ @nitsua60 You probably already know this, but this issue has been discussed twice before. \$\endgroup\$
    – Miniman
    Commented Jul 11, 2017 at 3:59

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .