A recent question has led to me seeking clarification as to what exactly counts as being 'on topic' for questions asking about probability.

The question I am referring to is Help Needed With Probability Math for 2d10.

The reason for my confusion is that our policy with this type of question seems confused at best, and undefined at worst. For example, Is 3d6 the same as 1d18? is another recent question that instead of being closed, received one of the highest vote counts in recent months.

What are the criteria that define whether a question talking about dice or game statistic probability is on or off-topic?


4 Answers 4


I think on RPG.SE there is place for that type of questions. You don't have to be a mathematician to know the mean result of 2d10 as you don't need to be a psychologist to know how to handle your game group.

Simple questions about simple probability calculations behind RPG systems are on topic IMHO. If not, we wouldn't have a probability tag, as absolutely any probability question is better answered by a mathematician.


If a regular roleplayer wants to know something about dices, where he probably is going to look, in a mathematician site or in a RPG site?

Don't forget this is a knowledge base. Maybe mathematicians are more prepared to answer those questions, but sure they will be far less interested than the people here, in answering or in reading the answers.

  • 1
    \$\begingroup\$ +1 remembering that the site must be useful as knowledge base. A roleplayer will surely look here for that kind of problem than on the math site. \$\endgroup\$
    – Flamma
    Apr 1, 2014 at 12:00

I think that we have two slightly conflicting elements to these kind of questions:

  • Complex maths is a minority interest amongst RPG players.

  • Dice games, and especially game design, often rely on probabilities working within certain bounds.

So we might have some questions which tread an awkward boundary, generally of the form I am considering using rule X in my game, but wondering if it is mechanically fair, or what the consequences were?

Could we extend this policy, similar to Stack Overflow's for recommendation questions: I.e. Instead of asking directly "are guns allowed in 1300 Europe" or "what is probability of having character with 3 18's under this system", please explain your related RPG/gaming problem and any approaches to solving it that you have considered or tried.

For the referenced question, this would then be fair game for probability-based answers, and other approaches. The question would still be closed as it stands on the grounds that we don't know what the OP's problem really is (not knowing the relevant maths is not necessarily an RPG problem). If the OP wants to know if the stat increase might affect game balance, or errs on side of too weak/boring, then that can lead to other answers exploring similar house rules.


The criteria is this, featured in our What topics can I ask about here? help article (the only one we get to customise):

Question about a general real-world topic such as history, geography or economics, whilst relevant to RPGs, may be off topic if they are not RPG-centric (or better belong on another Stack Exchange site, such as History). A good rule of thumb is to ask yourself …

Would an RPG expert give me a better/different/more specific answer to this question than a Historian, Geographer, etc?

If yes, then feel free to ask it here.

That's our rule of thumb for real-world topics. We arrived at a decision to use this back in 2012 when we couldn't iron out where to draw the line on campaign research questions, but it is not limited to just campaign research.

For example, Is 3d6 the same as 1d18? is another recent question that instead of being closed, received one of the highest vote counts in recent months.

It received one of the highest vote counts because it was in the hot questions list, and that's what happens to some questions that are in the hot questions list. People came, learned, registered, and voted. You might notice that one of the answerers, Eric Lippert, renowned for such things as his work on the C# programming language, has never used our site prior.

Why's it open though? I have no idea, and I voted to close it when it arrived.

It doesn't pass our rule of thumb. It involves dice, and dice probabilities are relevant to us, but answering that question requires no RPG expertise, and RPG experts have not given a better, different, or more specific answer than something a mathematician would.

  • 4
    \$\begingroup\$ Key differences between probability and campaign research (to me): RPG probability is a narrow subset of overall probability; campaign research is pretty much everything historical. RPG probability benefits greatly from RPG jargon (i.e. you can ask "what is the average damage of a brutal weapon?" without getting into the specifics of crit mechanics). Probability is a core RPG design skill, and an important skill for players / DMs. Finally: history can be fudged at the table; probability can not. \$\endgroup\$
    – AceCalhoon
    Mar 31, 2014 at 15:25
  • 4
    \$\begingroup\$ The difference to me is that 3d6 vs 1d18 has a practical RPG application. There's cases where you can choose one big die vs more smaller ones in some games, and knowing what that difference means matters. In the case of this question, it's more a straight probability question ("how many rolls on average to get to 100?"). The line is pretty fuzzy, but it seems that people see one. \$\endgroup\$
    – Tridus
    Mar 31, 2014 at 15:30
  • \$\begingroup\$ If that criteria is going to be used so strictly, I think the probability tag should be erased. All probability questions are best answered by a mathematician. \$\endgroup\$
    – Flamma
    Mar 31, 2014 at 17:07
  • 2
    \$\begingroup\$ @Flamma Not necessarily. Lots of probability questions are going to be asked by people who can describe the game rules involved, but don't have enough maths background to transform that into a pure-math question a mathematician can actually help with. For those, being able to describe the rules involved without understanding them and asking how the probability works does require RPG experts. \$\endgroup\$ Mar 31, 2014 at 17:58

You must log in to answer this question.

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