Slayer mechanics question

[ThreeGee]

Hey all,

I am revising Slayer of Dragon RPG with the intention of finishing it in the for-real, no-kidding sort of way. After six years, the mechanics are showing their age, and I am planning to use a new core mechanic, but it is pretty funky, so I want to run it past the gear-heads first before committing myself to something I will hate.

Attributes range from 0 (no ability in this area) to 5+ (inhuman level of ability), as do skills (called Techniques). The attributes are called Earth (Body), Air (Speed), Fire (Skill), Water (Chi), and (Void). Void is not a usable attribute and represents that ability which the player brings into the game. Each level of attribute has an associated die type: 1=d12, 2=d10, 3=d8, 4=d6, 5=d4, etc.

To resolve a challenge, the appropriate two attributes (for example, when attacking, use Earth+Fire) are rolled and the difference is subtracted from the relevent skill. If this results in a positive number, the challenge succeeds with a degree of success equal to this final sum.

If the action is opposed (for example, by the defender dodging: Air+Water), the opposing player makes a similar check and generates successes similarly. By comparing successes, the final result is determined.

This is all pretty abstract, and I will be detailing the crunchy bits in the actual game, but I am hoping the basic resolution system does not lead to wailing and the gnashing of teeth. Please let me know if there are potential pitfalls that I seem to have overlooked, or even if this system makes you go "Oooh!"

Thanks,
Grant

[ThreeGee]

Hey all,

The other mechanic I am playing with is this: reversing the die types and keeping the low die of the roll. This would be added to the skill.

However, the problem that I see with this method is that characters become less predictable with experience, rather than moreso. I would prefer a mechanic that allows for wildly poor rolls for young characters, predictably good rolls for characters in their prime, and wildly good rolls for characters past their prime.

Later,
Grant

[szilard]

Hmmm...

I haven't figured the probabilities here, but it strikes me that, under this system, it won't always be in your best interest to have a higher score. Say you are rolling Earth+Fire, and you have a d12 and a d4 in them, respectively. Rolling a 12 and a 4 wouldn't be so great. Since you want the numbers to be closer together, might you not be better off rolling a d12 and a d6 under this system?

One way to fix that is to add the Attributes together to get a score from 2-10. Then assign dice to that:
2 : d20
3-4 : d12
5-6 : d8
7-8 : d6
9-10 : d4

and subtract the result from the skill.

Stuart

[Christoffer Lernö]

Grant, with the caveat that I never really played your original system (only read it) here are my comments:

* Won't this mean all probabilities and stuff might totally clash with the balance in the original Slayer of Dragon? Meaning you have to do it all again (the balancing of powers and stuff)

* I'm always weary of people introducing metaphysical elements and other clever-looking devices for stats, skills and stuff. Just make sure this doesn't mean you have to make compromises later on.

* I thought your original was pretty neat and simple as it stood. The only things lacking was some more streamlined magic (are the "new" stats a way to unify these things perhaps) and doing away with having to average over stats all the time.
I'd rather see an evolution of the system rather than a near total rewrite of it. Besides, you already know it's strength and weaknesses - wouldn't it be easier to do something solid by iterating on what you have instead of rewriting it from scratch? But maybe you feel your system is getting too restrictive?

* As for the mechanics, I'm concerned about the subtraction. There are some older threads on this kind of handling time which you can look up. But the short take is that it will take longer to do subtractions that it'll take to do additions, and adding large numbers will take more time than adding small numbers. The ideal would be to be able to read the degree of success directly off the dice.
[/list:u]

[ThreeGee]

Hey Stuart, Chris,

Stuart, having attributes closer together is definitely a good idea. I forgot to mention a higher die type can always be chosen. Your proposed mechanic is a perfectly good one as mechanics go, but it does not fit the bill. If I am going to be weird for the sake of being weird, I am going to be weird for the sake of color.

Chris, I need to go through the move lists anyway to clean up the math, so I might as well do other changes I have been thinking about. The problem with the existing system is that the level of the attributes do not really mean anything, making adjudication difficult for someone who does not know the system well, and the higher the stats, the more the system breaks down in terms of relative ability. I would prefer to crunch everything down into a coarser-grained scale.

As for subtraction, I really do not think it is a problem in this case. Finding the difference between two small numbers is something most people can do without thinking. If I have five apples and take two away, are you thinking of the answer before I finish this sentence? Besides, the existing system involves division on the fly. That is ugly, even if I personally can do it quickly.

What actually worries me is the high potential for whiff-factor. The mechanic is non-obvious in terms of probability, and while the system takes into account that most rolls will fail at lower levels, I wonder if people will be frustrated by the fact that everything is negative.

Later,
Grant

[Mike Holmes]

As for subtraction, I really do not think it is a problem in this case. Finding the difference between two small numbers is something most people can do without thinking. If I have five apples and take two away, are you thinking of the answer before I finish this sentence? Besides, the existing system involves division on the fly. That is ugly, even if I personally can do it quickly.

What actually worries me is the high potential for whiff-factor. The mechanic is non-obvious in terms of probability, and while the system takes into account that most rolls will fail at lower levels, I wonder if people will be frustrated by the fact that everything is negative.

Huh, this is an odd system. Get this, no matter what dice you roll in terms of the lower die, the odds remain nigh identical of getting a particular result. In fact for - say - a d4 vs a d8 the odds of getting any difference 0 to 7 is exactly the same as if you'd rolled d8 vs d8. d8 vs d6 is actually better by a couple of percentage points, but not worth metioning.

Just ran figures for d12. Same phenomenon where odds are the same with d6 vs d12 as with d12 vs d12. d4 vs d12, the difference in expected value goes from only 3.97 to 4.42. d8 OTOH, is slightly better with a 3.76 difference being deducted from skill, and d10 is, oddly, the same result.

Maybe most interesting is that in all these cases, the odds of getting the best result, matching dice, never changes no matter what the smaller die is.

Tres bizzarre.

This is a problem; basically players will either have to guess what's the best die, or memorize which is best when (OK, gotta rememeber to roll a d10 or a d8 whenever I'm also rolling a d12). In any case, the lower stat will rarely be the die chosen, and never because of any intrinsic value to the low-ness of the die type.

Yes, it does make people with lower die types better, and on the "Wild end", yes, the bigger die types do allow for some bad results. That part seems to work about right.



Using system 2, you get a more stable result. Lower die types always result in lower results, so you'll always want to roll the largest available. d12 vs d12 halves the chance to get a one as compared to d12 vs d4. But in all cases rolling a one is the most common result. This sounds very contradictory to what you want.

Hmm. How about using the system 1 stats (low die is good), and subtracting the lower die roll result? That means you'll always want to roll the dice you are given, and the worse stats are wilder. In system 1, the most common result is a 1 dfference (second best to a zero result). In the system I've delineated, the absolute best result, a one, is the most common. So they're almost idenitcal in that way.

How's that sound?

Mike

[ThreeGee]

Hey Mike,

Hmm... I knew there was some weirdness in the math, but I had not realized it is that bad. System 1 is officially scrapped.

Your system has definite potential. Let's see, it still generates low numbers, so that is good for initiative, which I would like to make rolling. Still involves subtraction, but only once, not twice, and no division in sight. It loses the color of unity being key to success, however. How weird would it be to make doubles a critical success (no modification to skill), regardless of the roll?

I should probably 'fess up to being inspired by this article in Issue #9 of Places to Go, People to Be.

Later,
Grant

[Mike Holmes]

The article is interesting in some ways, but I'm not seeing the part about why it's important to have matching abilities. That must be something setting specific?

It loses the color of unity being key to success, however. How weird would it be to make doubles a critical success (no modification to skill), regardless of the roll?

Well, I'll assume you have some good reason for wanting this effect. What you have does not work as once again, the chance of a tie is solely predicated on the size of the larger die. Think of it like this, you roll the smaller die. This comes up with one number, doesn't matter which. The odds of the larger die matching the first are 1/d where d is the number of sides on the die. So matching types does not make it any more likely to tie, only lowering the larger die does.

The best way I know to get the effect you seem to be looking for is to make the results a cross product of some sort. That is, essentially do something like multiplying them together. The expected value for d6 x d6 is 12.25 while d4 x d8 is only 11.25. Now multiplying is cumbersome, but the same effect can be had with target numhbers. So, if you have to roll , say the larger die type to get under the level of the stat (1-5) of the lower, that works to produce the same effect.

Unfortunately, that's exactly what the system I suggested does do. The expected value of 6d and 6d is 2.52 while 4d x 8d is 2.18. So dichotomy is producing less subtraction.

OK, there are 8 combinations of possible rule sets here. I ran a logic table, and found the correct one. Lower die types are better, when you roll, you take the higher of the two dice, and subtract that from skill. The best result comes from low dice in general, but 6d and 6d is better than 4d and 8d, though 4d and 6d is better than either. So, basically, if it's a choice between total of sides, go with the even split, but whenever posible lower the total number of sides. Which means that it's never good to raise your die, so you don't need that rule. And poor stats are wilder than good ones.

That ought to cover it. It does mean that you are still subtracting which as you point out seems counterintuitive. But since your hurt more by your poor stats more than the good ones there's still a bit of intuitive matter there.

Here's the expected values for the system for each combination of stats:

Code Select Expand


      4      6      8     10     12
4   3.13            
6   3.92   4.47        
8   4.81   5.22   5.81      
10  5.75   6.08   6.55   7.15  
12  6.71   6.99   7.38   7.88   8.49

Mike

[ThreeGee]

Hey Mike,

I have been crunching the curves and there are some irregularities, so I am leaving the idea on the cutting-room floor. Oh well. At least I know not to try something of the sort in the future.

Today, we have all learned the value of doing one's homework. Premise and artistic concerns are great, but if the basic mechanic does not work to specification, the everything else becomes hopelessly derailed.

Edit: I missed part of your last post. Hmm... I appreciate the help, but I think the ultimate answer is to avoid gimmicks and just go with what works. 2d6 and target number, just like the current version, but with less math.

Later,
Grant

[Christoffer Lernö]

If you go with the old system... Didn't I already post you a lot of suggestions on how to improve it? I don't know if any of those were useful but I just wanted to remind you about it.

I'm still a little unclear on how your ideal system would look like though. Could you sketch a "wish list" so I see what you're working towards?

[ThreeGee]

Hey Chris,

Everyone's suggestions so far have been helpful, and that definitely includes yours. I have decided the current incarnation needs a serious going-over to bring it into line with my design standards. The moves are not balanced (at least not when you actually crunch the numbers), the math is ugly (divide by two every attack, divide by six every time you roll initiative, add and subtract long strings of numbers), and the numbers just "float", by which I mean there is no way of telling whether a given character is good or bad without knowing the system as well as I do.

The problem, I think, is that the design has drifted over the years from a relatively sim game based on fantasy arena battles to an overtly gamist game based on martial arts in general. The game is six years old, and its predecessor was designed around ten years ago. I was only half-way through high school at that point and was still learning the difference between roleplaying games and the more traditional games I had been designing before that.

Ideally, I would like a coarsely-grained, d6, roll-high, low-handling time, relatively transparent, easily learned system. However, I am willing to make sacrifices for flavor and I saw a lot of potential in the first mechanic I proposed. Unfortunately, it turns out to be horribly wonky, as are any similar dice mechanics.

My plan at this stage is to do what I originally intended: to make the attributes on a scale of 0-5, matching the technique levels, and to set target numbers for a 2d6 roll based on flat comparison. Not a sliding scale as in the current version, but flat numbers written down in a happy chart on the character sheet.

-3 FAIL
-2 = 11+
-1 = 9+ / CRIT 12+
0 = 7+ / CRIT 11+
+1 = 5+ / CRIT 10+
+2 = 3+ / CRIT 9+
+3 SUCCEED

For example, Skill 3 hits Dex 2 on a 5+ and does extra damage on 10+.

I still like the idea of rolling (continuous--it happens to also be random) initiative, so I will probably retain the yin-yang dice, using two dice of the same type.

Here is the link to the first thread.

On another note, I want a damage system that is quick and brutal for mooks, but allows for more detailed results against worthy opponents. I was thinking of something along the lines of "Star Power", which is added to Str/Soak to prevent damage and/or knock-out rolls. Player characters automatically get the maximum amount and faceless ninjas get zero.

Later,
Grant

[Christoffer Lernö]

The problem, I think, is that the design has drifted over the years from a relatively sim game based on fantasy arena battles to an overtly gamist game based on martial arts in general.

So what is your target now? You say "it's gamist based on m-a" right now - what do you want it to be?

My plan at this stage is to do what I originally intended: to make the attributes on a scale of 0-5, matching the technique levels, and to set target numbers for a 2d6 roll based on flat comparison. Not a sliding scale as in the current version, but flat numbers written down in a happy chart on the character sheet.

-3 FAIL
-2 = 11+
-1 = 9+ / CRIT 12+
0 = 7+ / CRIT 11+
+1 = 5+ / CRIT 10+
+2 = 3+ / CRIT 9+
+3 SUCCEED

I think this sounds great but then again Yggdrasil pretty much works like that :)

On another note, I want a damage system that is quick and brutal for mooks, but allows for more detailed results against worthy opponents. I was thinking of something along the lines of "Star Power", which is added to Str/Soak to prevent damage and/or knock-out rolls. Player characters automatically get the maximum amount and faceless ninjas get zero.

This was something I was considering for Yggdrasil as well. Maybe not exactly done this way, but basically heroes and master villains get bonus points somehow. I didn't quite manage to sew it together though, but I was working on it.
Anyway, I think it's a great idea. Ideally (I think) it should be kind of graded so that you have this overpowering fiend that is pretty much immune to your attacks and vice versa against unimportant enemies WITHOUT seriously screwing up the randomness in the combat system in the process.

I'm not sure how to implement it the right way, but the right FEELING of it should be like in the videogames when you're like level 4 going against this level 20 foe you can't TOUCH (or if you do you don't do damage) who kills you 10 times over with a single attack.
That feeling but without the obvious imbalancing built into those systems.

[ThreeGee]

Hey Chris,

That's exactly it: I want Slayer to be a fun game about martial arts. There is a whole genre of films about martial arts. Sure, some of them are modern action films and some of them are romantic period dramas, but the cool thing in these movies is the martial arts theme. They usually present overwhelming oppresion that one man can defeat through sheer focus and superior ability to kick ass.

The early editions of the game had lots of detail and rules to cover any situation, but they lacked the essential element of fun. When I ran games, the rules inevitably went out the window. If I, the designer, cannot play my game by the rules, something is wrong. Hence, a rules revision.

The color is all there, but the system gets in the way. I want a system that can be explained in less than ten minutes and a game that can be started in less than thirty minutes after the session first begins. That means simple character creation, which already exists, but also simple conflict resolution, which is lacking.

I do, in fact, have the revision mostly ready to go. Once I get it typed up and pretty, I will be presenting the next edition of Slayer for everyone's perusal.

Later,
Grant

[Mike Holmes]

The early editions of the game had lots of detail and rules to cover any situation, but they lacked the essential element of fun. When I ran games, the rules inevitably went out the window.

Here's a question that I think is key. The rules went out the window, and were replaced with what in terms of decision making? Who made the decisions? Based on what criteria? And don't say "whatever was the most fun". What made the decisions fun?

Mike

[ThreeGee]

Hey Mike,

Perhaps I misspoke. I do not mean I took my copy of the rules and lit it on fire before the players. I simply mean that there are times when applying the rules exactly takes longer than just saying the player accomplishes whatever they had in mind. Want to rip the mook's spine out and wear it as a hat? You do. Want to punch your hand through the door and yank the mook back through? You do. Why go through all the trouble of rolling initiative and calculating bonuses when there is no reason the player doesn't succeed at pulling off his crazy stunt?

What I have always loved about Slayer is that every time I have run it, the players have surprised me with increasingly crazy ideas. Players have used beds as weapons. Players have used other players as weapons. Players have done the nuttiest things I've ever seen in many long years of roleplaying, and that is what I love. As I have said before, the game started out with relatively detailed rules to encourage a strongly tactical game, but that is not the part that grabs people, so the rules have drifted over time.

I wrote DRAGON SUN : JAZZ NINJA to test the idea of a completely player-driven game based around the same ideas. I like DS:JN, but the crunchy bits are sorely missed. Having learning something in the process of creating DS:JN, I feel I am ready to finally create the definitive edition of Slayer.

I hope this answers your question. If not, can you be more specific? I feel you are applying one of your standard rants without regard to what has already been said.

Later,
Grant