d100 Roll Mechanic - Gradients of Success (With a Graph!)

[Maerok]

I'm working on experimenting with some different mechanics behind rolls. This has often been an intriguing part of games for me, from White Wolf's d10 to DnD's d20 and even diceless - how success and failure are quantified.

---

Something I've been toying around with now is a d100/d% style system. I'm still trying to find a master rolling mechanic for a project I'm working on.

How it works is that a character has each skill rated between 1 and 100 (or 0 and 99, whatever). There aren't really attributes but you could group the skills as such (and I might, later, for the purpose of traits *cough* feats). When you make a skill check, roll d%. If your result is lower than or equal to your rating, you got a success in one try. This rates as an excellent success (nicknamed an A). If you get it on the second try, it's an improved success (B). In three tries is a normal success (C). Four, a reduced success/minor failure (D). And after four, it's a failure (F). An S-rank might be if you roll a 1 on your first try (essentially turning a combat check into a death strike).

Excellent Success (A) - You performed the task as well as it could be done. You have definitely improved your situation.
Improved Success (B) - You succeeded with room to spare, and you've most likely taken a big step forward.
Normal Success (C) - You did it, with all your limbs intact!
Reduced Success/Minor Failure (D) - You kind of messed up, but you managed to not impact your situation negatively.
Failure (F) - You botched it. This will most likely come back to bite you or hurt your situation.

I guess the overall effect is that there is a built-in gradient of successes and it isn't necessarily pass or fail (rolling a 39 on a DC 40 Climb check leaves you falling into the abyss rather than securing a foothold but unable to proceed further). There is also a Conviction/Luck skill which can be sacrificed to increase your level in a skill for one encounter, so the player can somewhat actively affect the outcome and make up for shortcomings. I think Luck is applied before the roll and Conviction can be applied afterward.

Overall, the project is a modern Noir/Dark Fantasy set-up, with a diminished role of combat and more toward paranormal investigation. Kind of like a mix-up of Hellboy, Constantine, and The Brothers Grimm.

Applied over 4000 skill checks, the results look like this (there are some anomalies)...

http://i49.photobucket.com/albums/f286/pwned101/d100success.png

From these statistics, I can determine the starting point for all skills (maybe at 10 or 15) and then give a price to improving skills (I think it'll be more expensive to buy a skill point within a range of 20 to 29 than 10 to 19, and so on). Character points can be applied to increasing skills, health points (I'll sort of be like life hearts in Zelda games, but any damage not sufficient to empty an entire heart will quickly heal), and traits. Spellcasting/rituals (called Pacts) will also be factored into this via skill checks and expending entire health points.

Where the system comes under question is how to treat opposed rolls. Character A and B each have their own 5 possible outcomes but should that necessarily have 25 total outcomes when they go head to head? What happens when both people roll an A, when the unstoppable force meets and unmovable barrier? Should the degree of successes be subtracted? I am thinking that attrition and the so-called spiral of death might play well into the setting of the project, and this will decide the final answer to this issue.

---

Other systems I have been investigating are d20 (but using a d8 and d12, or "d8+12"; I like the sort of curve it makes) and some d6/d8 stuff.

[Flickerdart]

This is interesting...though you need a large investment of skill to not fail disastrously. No real way to set difficulty for a task, is there though?

With the opposed checks, you could have a "who rolled lower" to determine the winner, where the unstoppable force is just a tad more unstoppable than the immovable object. You could lower the result's effectiveness to C-level: sure, you managed to break his super-shield of lance-breaking, but you didn't really hurt the guy himself at all.

Equal rolls could be solved by who has the highest modifier. If they both have the same modifier, both checks fail.

[Maerok]

The option to have skill check modifiers is always a possibility, but that was an issue that I haven't been able to fully address yet. At low skill levels, there is a very big chance of failure. I'll see what I can do to exaggerate the ranges and make it harder to fail.

EDIT: How about starting off at 30 for each skill? This seems to have more agreeable ranges of success and failure. Reducing the value of these starting skills could be used to improve other ones.

[Flickerdart]

You can always have 50 be the baseline: you are half-likely to succeed at a task of average complexity (such as scaling a rough wall with a number of handholds) on your first try. There is an 11ish percent chance that you'll do poorly, though, if you have no training in climbing.

Now, assume that this character has gained a few levels and put 10 points into Athletics. He has a 60% chance of A-level success, and less than 9 percent of failure.

However, secretly, the handholds were just an illusion, created by an evil wizard because he was bored. The wall is, in reality, completely vertical and flat, and climbing it is impossible for your average Joe (-50% for a 0% chance). Our hero's meagre skill gives him a fair ~25% chance of eventually getting up that wall, but a huge chance of failing and plummeting to his doom. More than fair.

[Maerok]

Yep. That works pretty well. I'll most likely set some kind of max skill level that mortals can attain. After that, only supernatural creatures can fully master a skill.

[Flickerdart]

Traditionally, 95% has been the success cap: that natural 1 on the d20 is the bane of all but the most powerful even in games that don't use a d20. Limiting people to 70% rank (with a 2% chance of failure) is a lot kinder. Impossible (-50%) tasks still have a 40% fail chance, but a master of his or her craft should be able to succeed half the time.

Alternately, you could just make it really hard to get up that high, say by redoubling the costs for every 10 ranks. Sure, rank 70-79 represents the pinnacle of mortal power, but it's going to cost you as much as getting a new skill from 1 to 69.

[Maerok]

For opposed checks, half of the defender's opposing skill value could be applied as a modifier.

Deceiving someone with a Scrutiny of 100% would be at -50% (Impossible). On average (the defender has 50%), it would be a -25%. The deceiver would then have an effective skill of 25% which will get a Success or better about 60% of the time.