I got it !

Opposed rolls:

Compare the sum of CHAR + SKILL + SPEC of the parties involved.

The base difficulty for the roll si based on the active players' sum. Divide the sum by two. Whole numbers are Purple dice and any .5 is a Black die.

Remove or add a Purple die for every "increment" away from equal that you are.

EX. A Thief with Ag 4 and Stealth trained once (SUM=5) tries to sneak by a guard with Int 3 and no skill.

Base difficulty is 2.5 ( 2 Purple and one Black ) and you remove one purple because his SUM of 5 is higher that the guard's SUM of 3.

Result is 1 Purple 1 Black, which means 82 % chance of a success if 2 Green dice are used.

Being weaker and unskilled is bad for you.

Now let's try again with the same Thief but the Guard Int 3 is skilled and specialised in keeping watch (SUM 5).

BAse diff 2.5, unmodified because the SUMS are now equal, add a Black for skill and a Black for spec. So the final pool is 2 Blue + 2 Green + 1 Yellow + 2 Purple and one Black (base diff) + 1 Black for skill and 1 Black for spec = 50% chance of success !

Being weaker but very skilled makes you dangerous !

This rule seems to work... It gives more importance to skill and spec and also scales for very skilled characters.

Example very skilled characters.

Thief 6 Ag, 2 skilled, 1 spec against Guard 6 Int 2 skilled 1 spec.

Base diff 4 Purple + 1 Black (9/2= 4.5)

No Purple removed since SUMS are equal.

+ 2 Black for the guards' skill

+1 Black for the guards' spec

= 56% of one success !

Equal talent, equal chances. More or less. It still works !

Jericho said:

"The base difficulty for the roll si based on the active players' sum. Divide the sum by two. Whole numbers are Purple dice and any .5 is a Black die.

Remove or add a Purple die for every "increment" away from equal that you are."

Interesting but for me it does not work. The maths does it, but its not very logical.
Why does the base difficulty increase when your base stats and skills are higher. Of course there is the incrementation that lowers difficulty if the adversary is of lower skill.
Could we not simply look at the non active opponent character skill for base difficulty and divide it by 2. And as you proposed for .5 use black dices.
And also use black dices for trained skills and specialisation.

Jericho said:

You cannot use stance dice in story mode (and if it's an encounter using a progress tracker things can work very differently depending on how the progress is measured), that means that all your estimated probabilites are too high. Yes, GM can let one stance die be used, but I would restrict that severely (only allow it when RP calls for it).

But I agree that it could be a problem (the non-scaling of opposed tests). Comparing your last (very skilled) example with equal characteristics to an example with low skill adversaries I get (now using RAW, exactly the same training, no stance dice):
Chance of at least 1 success (using http://jaj22.org.uk/wfrp/diceprob.html )
Very skilled (i.e. 6vs6+2 skill, spec): 76%
Low skill (3vs3, 1 skill, no spec): 45%
So, yes, that might indeed be a problem. Of course one could use an opposed test, but I think it's a bit weird that a game mechanic that works for beginning chars have to scrapped at higher ranks. It should be scalable.

Your system might work with a little bit different scaling (without stance dice your success probabilities become a bit too small in my opinion), maybe removing one <P> from the base diff could work. It feels a bit complicated though and a bit weird that the base diff goes up as you become better.

I've just been chatting about this on RPGNet, and I think if I run into this situation, I'm going to make this adjustment:

Opponent's characteristic is 6 or 7: +1 challenge dice

Opponent's characteristic is 8 or 9: +2 challenging dice

Opponent's characteristic is 10 or 11: +3 challenging dice

Etc.

If you want that kind of opposed rolls it's easier to use the competitive roles from the core rules. Simply have the player and NPC roll their dice pool and then count final result in each dice pool. Lets say the player gets three successes and 2 boons. The npc gets 2 successes and 1 boon. That would mean the player wins with one success and one boon. If it's something that takes place over a span of time more competiitive rolls can be rolled and tracked with the tracker. Sneaking past a guard for example.

Other than that I have my general rule for combat where your difficulty is the opponents rank (or threat level minus 1 for monsters), but you can never have a higher difficulty than your own rank. This balance things out perfectly and is simple. But I have a thread on it "Various combat rules" in the house rules forum.

geekoo said:

Probably works that way too, I will have to check. It would be faster in game too. I agree that it feels weird that difficulty becomes harder as you improve, but that is what scaling is all about ! Remember, what I call base difficulty is actually the difficulty of an opposed roll against an opponent of same strength. So if you are better, the opponent is too if he equals you, hence more difficulty.

This said, your way at looking at it might work also and is much easier, SIMPLER, and that is what this thread is about.

I'm looking for simple tweaks, not rules overhaul.

Just made a few tests. Unfortunately, my convoluted method seems to work best at acheiving a 50-50 chance for equal sum characters, regardless of their skill level. The other way around (using half the SUM of the non-active character as base difficulty) seems to give an advantage to the active player and to raw characteristics vs skills, which I don't like.

I'll continue testing...

As for the Rank for Difficulty, my personal distaste of levels and the lack of realism it often creates (why is the rank 3 scribe so hard to hit ?) forbids me to use that. The characteristics and skills should define the character's abilities and capacities, not an abstract notion of rank. (and would that houserule be also used in social conflict ? how then explain the strange guile of the dumb rank 3 pit fighter ? ... doesn't work for me)

Jericho said:

If you use competitive rolls (that require no tinkering with the rules) you get perfectly balanced rolls based on the relative strenght of the involved parties. It can't be done more fair or precise that that.

Gallows said:

If you use competitive rolls (that require no tinkering with the rules) you get perfectly balanced rolls based on the relative strenght of the involved parties. It can't be done more fair or precise that that.

Two players with the same dice pool will have exactly 50/50 chance of winning. Just one single extra fortune dice in one of the pools will affect the outcome precisely. Look at competitive rolls on page 43 of the core rulebook .

True, it's probably the right thing to do in most circumstances.

But for those who want to use opposed rolls as a basis for difficulty in combat (or to cater to those Observation vs Target Guile) kinds of cards, I thought it nice to have an opposed system that would actually work, and be scalable. This said, the straitforwardness of competitive tests makes them easy to use and they can't be more precise !

Jericho said:

"Remember, what I call base difficulty is actually the difficulty of an opposed roll against an opponent of same strength. So if you are better, the opponent is too if he equals you, hence more difficulty."

Technicaly also if you have the same stats as your opponent you should have 50%/50% chance succeeding a task but I feel that as players are interpreting special characters, therefore I do not mind a slightly higher % of them suceeding tasks. But slighly only . I dont want to make it fair really.
if its 55% fine for me .

I tried a a few rolls tests and looked on the online dice roller + added some cards to have a clearer idea.
Seems to work fine...
Did quite a few simulations in tables on Excel and looks not bad.

so in résumé:

opposed rolls:

- charac as usual
- Stances. Here is the slight modification I would perhaps use. the idea is that u can change characteristic dices with stances but not more than half of main stat (If allowed in your career).
ex: STR 3 = 2 blue dices + 1 red/green
STR 4 = 2 blue dices + 2 red/green
STR 5 = 3 blue dices + 2 red/green
- yellow dices
- white dices
- action cards.

DIFFICULTY
- opponent charac /2. the .5 results are converted in black dices. the full integer numbers in purple. so ex: AG 3 = 1.5 = 1 purple + 1 black dice
- the opponents trained skills, specialisations, tactical disaventages... = black dices
- adding action cards
- NO DEFENCE bonus. I am not using the defence only the soaking value. This defence system for me seems weird. I prefer using the AG of opponent.


if a chaos stars appears perhaps would assume attack was missed like proposed by Gallows.

At the creation I would restrict humans to max charac of 4. And in later game a maximum of 5.
For Elves and Dwarves 5 max at creation and max 6 for the racial bonus charac later in character evolution. Or simply not allow Elves... God I hate them...

Still balacing the creation points for a lower start (need to run tests):

I think the human around 20 or 15 creation points.
Elves and Dwarves around 15 or 10 creation points.
I would not restrict Stance points at the creation (simply like in rules). The only thing I would balance is the use like mentioned prior. You cant use more than half your base characteristic in stance points. Now this is a suggestion only. Stats of course are a little better with the normal use of stances.

voilà c'est tout pour le moment

I found that with 3+ purple the chance for one chaos star is too great. Instead I am thinking about making some better defensive cards for rank 2+ using eon. It's easy to create new defensive cards with +2d or more. Even cards that remove dice from the attackers pool... many ways of creating defensive cards that make defending viable.

I hear you Gallows.

But since advanced defense gives purple dice... if you make them better... how will you without upping the occurence of Chaos Stars.

This said, as per RAW, CS aren't that bad... Only in certain situations/cards do they truly have an effect, and then only a fumble type effect, but nothing catastrophic. ie, in the ruins, they cause 1 wound.

Jericho said:

I simply discarded the whole chaos star = miss idea, because with 4 challenge dice means over 40% chance for a chaos star.

Currently I am looking at it all again. For example someone with the following:

3 Blue dice (str)

3 green dice (str)

3 yellow (ws)

2 white (spec + str die)

7 purple

Chance of one success = 68%, Chance of three successes = 32%

The main issue is the purple dice I think. They are too weak in terms of the actual challenge they provide. Only 50% chance for one or two challenge symbols on one die, whereas a green die has 70% chance for one success. In fact one green against 1 purple give a success rate of 35%. This is where the issue is, so I am toying with an idea of making the purple more powerful in terms of giving challenges, since making the purple die stronger would also make scaling a hell of a lot easier. Currently I am calculating some success rates based on the following rules for the purple die:

One challenge = one challenge

Two challenges = two challenges

One bane = one bane + one challenge

Two banes = two banes + two challenges

Chaos star = chaos star + three challenges.

That will create a much stronger purple die and make scaling a lot easier without having to add 10+ purple dice to a dice pool. Hopefully FFG will create another kind of challenge die in a supplement that is a lot stronger. But for now I will tinker with this route and see how it works out from a newly created chatacter to a rank 3+ with a characteristic of 6, a skill og 3, one furtune die in the characteristic, specialization and 3 dice converted to stance dice.

Now if wendolis could help out with some nice visual statistics I wouldn't mind

Gallows said:

;-) if I find the time ... But started to take up the SE Plugin again ...

vendolis said:

Hehe yeah that's important stuff. While some of us still have the "banned because of copurights" version it's really great if everyone in the community can create and share spells, actions, careers and such. SE is so easy to work with that anyone can make new cards in a few minutes if they have an idea. I'm just using your die roller now to find some averages for the dice (just 1000 rolls since I can't make it accept more than 100 at a time).

Gallows said:

Just to clarify, it is not my dice roller. I took it from the site and modded it for myself to roll faster and more. I have no idea who created the roller since there is no claim in the roller itself or any hin on who programmed the tool.

This is very rough calculations, but just trying to get a rough estimate if this route is viable.

1b = 0.5 average Successes on one die

1g = 0.7 average Successes on one die

1r = 0.7 average Successes on one die

1y = 0.6 average Successes on one die

1w = 0.3 average Successes on one die

1b = 0.3 average challenges on one die

1p = 1.5 average challenges on one die (versus 0.75 average according to core rules)

With the normal rules for a purple die the average number of challenges on one die is just 0.75 and compared to one green it's almost 50-50 in terms of breaking even (failing of course with 0 successes). But the value of the purple die is way too low. With this new rule the average of the purple die has doubled, so that two stance dice have almost a 50-50 chance of breaking even.

I think this will balance out a lot better across the ranks if better active defence cards are produced as well. It decreased the chance of success for starting players as well of course, but mostly it will hurt those who have 2 dice in a stat and no skill.

2b+2g+1y+1w = Around 3 average successes

1p = around 1.5 average challenges

2p = around 3 average challenges

I think that looks pretty fair for a starting character and is more scalable than using opposed rolls for combat, since starting at 2p makes active defence relatively weaker. And since these rules affect everyone, npc and players it's not a relative weakening of the players. In fact with their active defences it might actually make them slightly stronger.

Chaos stars being 3 challenges is slightly too much. Chaos star being 2 challenges is better.

That's 1.375 challenges average on one purple die. Also it doesn't give quite as ugly spikes with three challenges on one die.

vendolis said:

Thanks for the link to your version of the die roller... very nice. Doing 1000+ rolls is easy now. For some reason I can't send you a PM.

Making a fast run with 10000 Rolls on the modified purple die with the following setting:

{B} = {BC}
{BB} = {BBCC}
{S} = {SCC}

And a dice pool of:

<PPPPP><BBB>(GGG)[YYY][WW]

Has the following estimated success/fail percentage:

Fail: 80%
Success: 20% (of which 5% are Tripple Success)

There are near to non (blow 1%) Tripple successes with double banes. And very little (1.4%) Tripple successes with a Star. This effectively eliminates the "Pyrrhic victory" effect from the results.

Going down to <PPPP> changes it to about 60% fail / 40% success ... <PPP> then goes to about 35% fail, 65% success (with 29% Triple Success).

My feeling is the die is too extrem an takes the very interesting part of loosing wins out of the account.

p.s. Yeah .. no idea.. PMing does not work for me either..(even with friends)... Just contakt me on the SE plugin forums: www.vendolis.org/forum

Can you write that {B}={BC} stuff somewhere in the utility so I can test our a few different settings?

Gallows said:

But you'll be rolling twice as many dice. The beauty of the new system is that you only roll dice once, and you can see success and failure in a single glance.

The system really needs a better system for opposed rolls.

Personally, I think skill and specialisation should count for a lot. Especially in combat. I don't like it at all that combat abilty is now mostly based on Strength. Coordination is a much bigger factor in trying to hit someone (or avoid being hit), except maybe for the most untrained of fighters. I want each point of skill in the opponent to be worth at least a challenge die.

I think the biggest trap that the opposed roll system in RAW fell in, is to use relative ability as absolute difficulty. This is something an improved system needs to avoid. One solution is to make everything relative, which is what the original proposal from this thread does, another is to make everything absolute, which I think I'd prefer. With the differences in the dice, you're never going to get it quite as perfectly balanced as a competitive roll, but most rolls aren't perfectly competitive anyway. We just need something where the math works out reasonably well.

As an aside, you all do realize that the down side to rolling stance dice are the penalties involved on those dice? Reckless has banes as well as Fatigue/Stress. Conservative has delays. This is to 'offset' the increased chance of success on the stance dice.

I would suggest keeping whatever method you decide on pretty simple. Consider that the active player is already taking into consideration his own abilities and training by adding positive dice to his dice pool. All you are doing at this point, is trying to adjust the challenge for the opponent. Why not use just the opponent's abilities as a base? For example, the opponent's characteristic/2 (round up) in <P>, with an additional <P> for training and for specialization.

1-2 => <P>

3-4 => <PP>

5-6 => <PPP>

7-8 => <PPPP>

9-10 => <PPPPP>

So, the challenge is directly based on the stat of the target's ability, while the user's abilities are reflected in the stance/characteristic/training dice being rolled.

Yes but adding more dice as I was testing with my initial rules, rendered the active defences more and more useless. From doubling the number of challenge dice at rank one to adding just 25% higher up. Modifying the die itself lets it scale. But well, perhaps more dice will be released. The last diary on the dice certainly hinted at that.