Halving skills example

nennafir

Mongoose
With apologies if there have already been too many of these. Lets do an example of an opposed skill check using some non-combat skills (since combat skills are supposed to not be halved.)

We have a trollkin and a human using the skill.

Warning: I did this using the windows calculator which is kinda sucky and it's 9:00 pm at night. So although I think it is correct I might have made an error using that silly calculator.

Also note that there is some ambiguity in the rules on page 20. I assume, for example, that if both fail and get the same roll so that neither is "lowest" they just have to reroll. The rules do not make it clear, at least in that section of the rulebook.

(A) Scenario 1. Human has skill at 100 and trollskin has skill at 30. Then the trollskin wins the skill test if:
trollkin succeeds&human fails+trollkin succeeds&human succeeds but worse than trollkin+trollkin fails&human fails worse than trollkin
=(.30)(.05)+(.01)(.01+.02+...+.29)+(.01)(65(.05)+.04+.03+.02+.01)
=(.30)(.05)+(.01)(.01)(1+2+...29)+(.01)(.01)(69+70+....+100)
=(.30)(.05)+(.01)(.01)(29)(30)(.5)+(.01)(40(.30)+(29)(30)(.5)(.01))
=.015+.0435+.0335=.092
In other words, 9.2% of the time.

(B) Scenario 2: Human has skill at 120 and trollkin still has skill at 30. So we halve things to get new human skill 60 and new trollkin skill 15.
Along the same lines as above, the trollkin wins the skill test if:
=(.15)(.4)+(.01)(.01+.02+...+.15)+(.01)(45(.4)+.39+.38+...+.01)
=.06+.012+.258=.33
In other words, 33% of the time.

Again, with apologies if this has been done too much. I just wanted to give one concrete example at least.
 
nennafir said:
(A) Scenario 1. Human has skill at 100 and trollskin has skill at 30. Then the trollskin wins the skill test if:
trollkin succeeds&human fails+trollkin succeeds&human succeeds but worse than trollkin+trollkin fails&human fails worse than trollkin
=(.30)(.05)+(.01)(.01+.02+...+.29)+(.01)(65(.05)+.04+.03+.02+.01)
=(.30)(.05)+(.01)(.01)(1+2+...29)+(.01)(.01)(69+70+....+100)
=(.30)(.05)+(.01)(.01)(29)(30)(.5)+(.01)(40(.30)+(29)(30)(.5)(.01))
=.015+.0435+.0335=.092
In other words, 9.2% of the time.


(B) Scenario 2: Human has skill at 120 and trollkin still has skill at 30. So we halve things to get new human skill 60 and new trollkin skill 15.
Along the same lines as above, the trollkin wins the skill test if:
=(.15)(.4)+(.01)(.01+.02+...+.15)+(.01)(45(.4)+.39+.38+...+.01)
=.06+.012+.258=.33
In other words, 33% of the time.

In Scenario 1, you don't half. You only half a skill above 100.

Scendario 2 is correct. There is an anomaly that will continue any time your cross a hundred percent barrier (200 to 201, 300 to 301, etc.)

Doc
 
??

I did not half things in example 1 Doc. That was the whole point of the example. In the first case, where things weren't halved, the trollkin only had aroun d a 9% chance of winning. But when the human increased his skill and the trollkins stayed the same, the trollkin's chance of winning went up (dramatically, in fact.)

I am sure that this has been stated before, but concrete examples that really spelled everything out seemed absent. So I offered one. The whole halving mechanic is obviously hopelessly broken.
 
Note also that my example assumed that the text on page 20 is correct when it says "Both characters fail--Whoever rolled the lowest in their skills test wins the opposed test."

This rule is also completely bass-ackward I think. It means that if a highly skilled person fails their attempt, they really botched it and are doomed. I think a much more realistic way would be to say that if a hgihly skilled person failed then they probably didn't fail too badly. In other words, the rule SHOULD be:
Both characters fail--Whoever rolled the highest in their skills test wins the opposed test."

Note that this is a separate issue from the halving though. I think they should implement the above fix and figure out some entirely new method aside from halving for that problem.

The whole thing leaves me feeling that MRQ was not really game-tested at all, or only game tested by non-critical people. This is really pretty inexcusable. I'm going to be starting up an MRQ campaign and will give them the benefit of the doubt, but they really need to release some errate promptly. The whole problem could have been avoided, of course, if they had adequately testing things.
 
Nennafir,

You can check out my Probability Calculator (see my signature) if you wish.

You are right though, this has been done to death already.

cheers
 
bluejay said:
Well criticals don't come into standard Opposed Skill tests.

Oh, I didn't know that.

Was it so in RQ3 as well? We have always played so that with critical skill roll, you got really good results - why it is not the case anymore?

I would like a idea about using criticals which are calculated before halving - it would work well and will give that >100% guy some more reasons to improve skills.

EDIT: And of course I would then use criticals on standard opposite skill tests as well. AND Fumbles
 
nennafir said:
??

I did not half things in example 1 Doc. That was the whole point of the example. In the first case, where things weren't halved, the trollkin only had aroun d a 9% chance of winning. But when the human increased his skill and the trollkins stayed the same, the trollkin's chance of winning went up (dramatically, in fact.)

I am sure that this has been stated before, but concrete examples that really spelled everything out seemed absent. So I offered one. The whole halving mechanic is obviously hopelessly broken.

DOH! Many apologies. Another reason after a long, hard week, I should've just went to bed.

Doc
 
I think accepting that opposed roles are not halfed in combat is dangerous. when both opponents are over 100% the combat will just revolve around who has the most combat actions. I propose two things:- 1. Ignoring the 'official' explanation which clearly contradicts the book and;
2. when rolling opposed test, if one or both>100%, then letting the highest % choose whether to half or not. Actually this makes sense for non combat as well.
 
Back
Top