How do I ask statistics questions on this site?

Question

I recently asked this question after asking this question for two reasons.

1. To get a better idea how drowning works in Call of Cthulhu 7th edition
2. To use the answer to decide how to run my next game. If drowning chance was too high, I would find a way around it, and if drowning chance was too low I would bump up the difficulty.

What I got was mostly people telling me that the probability didn't matter, which didn't help much. I was trying to tune the probability of drowning to have a 1/8 risk of getting drowned by sentient water, to give stakes but not a guaranteed death. But I mostly got no answer which wasn't helping me adjust dice rolls. There were some answers that gave good math and I think accurate rulings, which is enough for me to move forward.

How should I ask questions relating to probability in the future to help people understand the questions?

Answer 1

When you ask how best to use a hammer to drive a screw, you have to expect some number of people to suggest using a different tool.

You outlined your problem well, and it was clear what sort of answer you were looking for. I've never played the game and I understood the question perfectly well. But one of the neat things about asking questions to a community of experts is that we often see that the best solutions to our problems don't even fit within the framing of our questions. When this happens, we call it a frame challenge:

A frame challenge is where an author answers a question in a wholly different way the querent never asked for, or potentially expressly forbade, but in a way the author feels will actually solve the problem. (Or otherwise improve the querent's life quality or prevent them from making some terrible mistake.)

Reading through the answers to your question, it looks like you got a couple answers working within the scope you asked for, and a couple answers challenging your framing based on their experience with the game. This is the system working as intended. We can't forbid frame challenge answers, some of our best answers are exactly that. Sometimes a frame challenge changes the way you think about a problem and provides satisfying resolution; sometimes it doesn't. That's what votes are for.

In your particular case, it seems your question framing was quite ripe for being challenged. If I'm understanding Glazius' answer correctly, some of what you are trying to do is somewhat at odds with how the game was intended to be run by its authors. To be clear, this is okay. There is no badwrongfun. But this is where the hammer and screw analogy comes in. There is nothing inherently wrong with driving a screw with a hammer, generally speaking (building codes notwithstanding). But the screw was not manufactured with the intent to be driven by a hammer. So naturally, when you ask for help, we must expect that someone is going to suggest that we are going about it the wrong way, that the screw was not made to be driven by a hammer. If you're happy with your hammer and don't care to use a screwdriver, just say thanks and move on, someone will probably come along and help you use the hammer, which is what happened to your question.

Answer 2

Please add details and clarify the problem being solved.

(This is an injoke about some site boilerplate text. Hang around a while, you'll see it eventually if you haven't already.)

One of the functions of the domain knowledge that an expert should possess is to be able to distinguish the meaningful and meaningless elements of the relationships between things in that domain.

This is why the question page asks you to be specific and provide details. If things have come to the point that you hope anonymous strangers on the Internet will be able to solve your problem, you should be honest with yourself that you are, at least situationally, not an expert, and you do not know how meaningful some of the details of your specific situation are -- so provide as many as you can. Be open, be conversational, tell your story! (Well, not the personally identifiable details of your story. This is the Internet, after all.) If some of it turns out not to matter, or to be a distraction from the more relevant elements of your question, you can edit it out later -- regardless of site reputation you can always edit your own stuff freely.

I understand the impulse to pull back into generalities to hope for something useful by chance, but if you need specific help, asking more and more general questions trying to find it is only likely to get you more and more general answers.

Especially in this case, wow! I mean -- you probably already know this, but I'm writing this just as much for a general audience -- Call of Cthulhu has been going for over 40 years now, and over most of that time it's built up something of a reputation as a real investigator meatgrinder. The seventh edition of Call of Cthulhu (CoC7E) is written in what I would, uncharitably, describe as a defensive crouch against its own reputation. "Don't kill the investigators! Don't kill the investigators!" it's screaming, left and right.

Don't kill the investigators... in general.

Where possible try to avoid an outcome that will end the game (unless you wish to, of course).

-- "Failed Dice Rolls and Sudden Endings", CoC7E Keeper Rulebook, p. 86

It's just fine to kill them in specific. Like, when the story's reached its climax and even their deaths would still be interesting.

And hey, while I'm copping to being uncharitable! In what an outside observer would probably call a thoroughly predictable plot element, as this whole affair has dragged on I spent way too much time in close reading of that eldritch tome and it messed with my head a little. I've done jack all to help you get to this point and my responses to your more and more general queries have made the whole thing worse for you.

So: thanks for putting yourself through the process and opening up about what you were trying to do. I'm sorry for the confusion you've been through, both personally induced and in general. If you want to head back to your original question and edit in all these details you've shared - the final confrontation, the dire threat, the desperate escape - I can absolutely get you a better answer.

Answer 3

You can do so by asking a question only about [statistics]

In earlier editions of Call of Cthulhu, there were none of these modern ideas of failing a skill check not leading directly to failure. In those old editions, when you failed a swim check, you started to drown, period. And if there was no land or boat nearby to drag yourself up to, you would die. It sounds like that is how you want to play, only with skill values that calibrate the probability to a level that you feel is the right one.

In CoC 7e, the probability resolution mechanism still works pretty much in the same way, except for the addition of pushed rolls. But: the rules about resolving statistical probabilities are not the only rules of the game. It does not end there.

The rules about how to deal with the outcome of such rolls are just as much rules of the game, and they they have changed a lot from earlier editions and now loudly and clearly tell you to not let the investigators die if it does not make narrative sense. You prefer to ignore that part of the rules. You prefer to stick with the old-school approach, and let the die fall as they may, consequences be damned.

And you can, but then you are overriding the game's rules. You either can ask a question about probabilities, which is hard to do in a vacuum outside the other game rules, or you can ask a question about how deadly the game rules make swimming. These are very different things.

If you just want a question about a statistical probablity calculation: the way you could ask it is by providing all the assumptions and conditions, as you did with the swimming skill value, and the process that the sequences of rolls are to follow, for a hypothetical homebrew game you are considering. This game may be similar to CoC 7e, but it is not CoC 7e, as it ignores several of the core rules. Then, someone interested can calculate what the probabilities will work out to under the given assumptions and procedure.

However, as soon as you combine this with how this will work in 7e, people will tell you: in a way that makes the statistical rolls really not that important.