Questions regarding interpretation of the rules - Skill levels and specialties.

[Heavy Arms]

Firstly, I can only work with the numbers you gave me,...

Well, clearly you can work with other numbers, because I gave you a situation where I rolled a crit. That is, I rolled two 10s, for 4 successes. A 3/1 split is impossible in this situation, but you've talked about a 3/1 split anyway.

...and I'm not turning it into a failure, granted it's less of a success narratively.

You are, but you're obfuscating it with this 3/1 split talk.

Basically, you are advocating for a system where, instead of rolling Attribute + Skill to beat a Difficulty, you roll Attribute to beat a Difficulty, and Skill to beat the same Difficulty, and the practical results are dictated by the lower roll.

So, in my setup, where the options are 4/0, 2/2, or 0/4, against a Difficulty of 1, I fail two of the three potential outcomes because I got zero on the worst roll. You want to bring up a 3/1 or 1/3 split because those are still successes at Difficulty of 1 rather than overt failures; thus you don't have to address my actual example.

simply because one roll shouldn't dictate the result of something that has taking many years to complete.

Why not? You know there are games out there where you never roll more than once for anything?

And we're talking V5 specifically, which directly advocates for not turning something into multiple rolls unless there's a compelling reason such the difficulty changing over the course of the action. Unless there's an important reason to extend the roll into multiple rolls, how long it takes doesn't matter. Just writing a term paper should be a simple roll because it's a single action despite the months spent on it. There are certainly plot ways you can spice it up and justify an extended roll here, but those are plot reasons, not time.

Adding rolls feels like a transparent way you're trying to mitigate issues with your ideas. If you roll a lot, on average, successes based on individual traits instead of full dice pools will map to your ratings in those traits. If you roll a little, there's an increased chance for highly incongruous results like my above example where my Int 3/Academics 3 character doesn't roll evenly between the two and thus doesn't actually feel like a balanced character that succeeds through equal applications of two traits.

That really depends on interpretation,...

What passage, exactly, is up for interpretation from the book?

Even if looking at it statically, the paper would only reflect Intelligence 3 and Academics 1

Why is my Int 3/Academics 3 character putting out an Int 3/Academics 1 paper with 4 successes? How is that a "correct" outcome?

The whole idea behind interpreting the result of the roll based where the success / failures come from, would be to make a more correct narrative interpretation of what's actually happening.

Sorry but, "something I like more," doesn't make it correct.

The closest to an objective "correct" regarding the interpretation of dice in an RPG would be: "reflects the way the RPG uses words and numbers to define the character while still presenting the results in a fashion that aligns with their impact."

Dividing Attribute and Skills successes is not a "more correct" method than leaving them combined and using the static ratings on the sheet to guide narration instead.

You have yet to make any sort of case for what the benefits of the added work of tracking different types of dice, redoing the math of the system, rolling more often, and so on actually are compared to the simpler option of using the existing ratings on the sheet.

[Kharnov]

I think the argument that has been made previously, that you should generally just intuit the narrative result from the ratings in Attribute and Skill, rather than care about which dice produced which successes, is the one that makes the most sense. If you really want that level of "granularity" though, it does seem like there's a relatively simple solution: split each dice pool into its Attribute and Skill components, roll them separately, and then add up the successes separately. I tend to think more about narrative than mechanics and statistical probabilities so maybe I'm missing something, but it doesn't seem like it would produce a statistically significant difference in the results; at the end of the day, you're rolling the same number of dice against the same Difficulty, just taking the time to note which successes came from where.

But that's the big downside: Time.

Handling things this way turns pretty much every dice roll into two dice rolls. It's not a huge difference in time, but it'll add up over the course of an hours-long session. And "too much time spent rolling dice" is one of the more common criticisms that can be directed at TTRPG systems when it gets in the way of letting the narrative flow. This is why just making an estimate based on the ratings of Attributes and Skills is better, it saves time that can then be spent on more fun rather than being nitpicky about the narrative needing to precisely reflect every single roll of the dice.

And you definitely shouldn't be making someone fail an Academics roll just because all the successes came from Intelligence. Have it alter the description of the outcome, sure; the paper with Intelligence successes gets more points for generally solid reasoning, the one with more Academics successes gets points for technical proficiency and information specific to the field. But if 4 successes is a grade of 80%, then you get 80% regardless of whether the successes came from Intelligence or Academics. If you have any rating in Academics, that should cover the basics to ensure you aren't losing points for bad formatting, accidental (or obvious) plagiarism, etc. And if you don't have any rating in Academics, then you are just writing a paper based on Intelligence and it should probably fail, or at least be criticized, in an academic setting, but should still be rhetorically persuasive based on solid reasoning and general information.