Opposed roll clarification

[Pentallion]

Quote

A tie (where both participants succeed but achieve the same quality of result)

In spirit combat page 368 the rules for opposed rolls are worded differently.  What is "quality of result?"  Page 144 says:

Quote

 A tie (where both participants achieve the same type of success but roll the same number)

The example doesn't cover what happens when both succeed but don't roll the same number.  So both have the same quality of result (success)?  but what if they don't have the same number rolled?  I know from owning Mythras what happens in that game, but RQG seems to take that knowledge for granted and doesn't expressly explain how opposed rolls are won or lost.  Page 368 infers if both succeed it's a tie.  Page 144 makes it seem more ambiguous.

Also needing clarification in spirit combat is if both succeed do both take damage?  The only example given is for crits.  Do both only take damage in ties when both crit?

Ties are not explained very well and I'm purposefully ignoring what RQ 6 rules were because one shouldn't have to know RQ 6 rules when reading these.

Edited June 2, 2018 by Pentallion

[smiorgan]

It seems to me that the correct version is that on page 368, because the wording of page 144 would leave a lot of cases unaccounted for.  Apparently, the rule works as in Elric! / BRP/ Magic World (tie is a tie) and not as in RQ6 (higher roll wins in ties). 

[jajagappa]

43 minutes ago, Pentallion said:

The example doesn't cover what happens when both succeed but don't roll the same number. 

p.368: "If both combatants succeed, the winner is whichever combatant achieved the better result."  I take this to mean there is a winner and a loser, unless they both rolled the same.

44 minutes ago, Pentallion said:

in spirit combat is if both succeed do both take damage? 

No, based on the above, at the same level of success, there is a winner.  "The winner succeeds and the loser fails. The winner does spirit combat damage to the loser."

Only in the case of both achieving a critical success do both take damage.

[Pentallion]

That's great jajagappa but define what is meant by "same level of success" and "better result".  It doesn't say who rolled the highest, which to most RQ players is unintuitive and confuses players.  It says level of success which infers simple success, special success and critical success.  So if we both normal succeed, it is a tie, no winner and loser.

[BWP]

In the absence of any official wording saying anything different, my attitude would be whoever achieved their success by the bigger margin would be the winner.  If my skill is 30% and your skill is 35%, and I roll "29" and you roll "10", I've rolled higher but only with a margin of 1%, whereas you've achieved a margin of 25% -- in my view, a more substantial victory.  A rule that says "highest roll breaks ties" is advantageous for people with higher skills (in general) but disadvantageous for people who actually roll better at the time the contest takes place -- and isn't that the point of the test in the first place?  Otherwise you may as well just give the win to whoever has the highest skill level and not bother rolling dice at all.

 

[jajagappa]

1 hour ago, Pentallion said:

"same level of success" and "better result"

To me, same level means:  crit/crit, special/special, or success/success.  This seems a reasonable interpretation since it subsequently calls out: special/success and crit/special.  And on p.143 it notes that the tie is: "where both participants achieve the same type of success but roll the same number".  Crit/crit is then noted as a special case of a tie for both opposed rolls and spirit combat (though in the latter, both do damage to the other).

[Furry Fella]

Agree solidly with Jajagappa on the level of success as this is consistent through out. It does though raise a set of issues that should be clarified by tidying up the multiple wordings And an index reference that covers this simply and clearly - an area where I've become an increasing fan appendices that include a section on system mechanics. There are not many conventions in RQG (so far in my reading) that this would be complex or burden some.

[Mugen]

Quickstart was clear that "the better result" means "the highest die". 

Page 6 : "Opposed rolls are made when one character’s ability is pitched against an opponent’s ability. A simple success may not be enough to overcome the opponent. Opposed rolls are not used to resolve melee combat.
To make an opposed roll, both participants roll their respective abilities. If both participants succeed, the winner is whoever rolled higher. If one participant got a special success and the other a normal success, the special success is the winner. If one participant got a critical success and the other a normal or special success, the critical success is the winner.
An opposed resolution may thus result in a winner and a loser, a tie, or two losers. 
• Winner and a Loser: The winner succeeds and the loser fails.
• Tie: A tie (where both participants succeed but roll the same number) means the situation is temporarily unresolved. If both participants rolled a critical success, the result is a tie.
• Two Losers: Both participants fail their rolls. Stalemate, no success. "

I really think the "two loser" part should not be, and in case both fail their roll, the highest roll wins, even not as clearly as if he had won. This would make opposed rolls more similar to resistance table, where chance of success depends on the difference between characteristics, and not absolute values.

Edited June 3, 2018 by Mugen

[PhilHibbs]

A tie should be broken by the highest die roll winning,. This gives the edge in both cases to the higher skill.

Edited June 3, 2018 by PhilHibbs

[jongjom]

1 hour ago, PhilHibbs said:

A tie should be broken by the highest die roll winning,. This gives the edge in both cases to the higher skill.

This is what should be written, or maybe it was, kind of. It would be good to have clarity on this from Chaosium. 

[simonh]

It’s possible there was a highest roll wins tie breaking rule in there at the time of the QuickStart, but it’s got removed in favour of a strict level of success ladder and now there are some residual references hinting at the old method in the text. Looks like we need clarification.

[smiorgan]

10 minutes ago, simonh said:

It’s possible there was a highest roll wins tie breaking rule in there at the time of the QuickStart, but it’s got removed in favour of a strict level of success ladder and now there are some residual references hinting at the old method in the text. Looks like we need clarification.

It looks like that. Botched editing away of a "highest roll wins" rule. They need to look into it.

 

[styopa]

8 hours ago, PhilHibbs said:

A tie should be broken by the highest die roll winning,. This gives the edge in both cases to the higher skill.

Exactly, simply correct.

This is how I've always seen opposed rolls resolved.

I know in beta versions of the rules there was a lot of back and forth about whether opposed rolls would even be included, so I'm unsurprised that the text reflects some confusion.

[Mugen]

If you only consider quality of success, there will be a lot of ties.

And in cases where there is an active protagonist and a passive one, a tie is the same as a victory for the passive one, as his goal is to prevent the active character from reaching his goal.

For instance, if a gard is searching for a hidden character and both have 60% in their appropriate skill, the cases in which the guard finds the other character are when :

-He rolls between 13 and 60 and the other player rolls 61+

-He rolls between 4 and 12 and the other player rolls 13+

-He rolls between 1 and 3 and the other player rolls 4+

For a total of 24.35% chance of success, from which we can substract the 1% chance they get the same result on the d100.

[hanataka]

It should be
”Tie: A tie (where both participants achieve the same type of success and roll the same number) means ..."

[simonh]

10 hours ago, Mugen said:

If you only consider quality of success, there will be a lot of ties.

And in cases where there is an active protagonist and a passive one, a tie is the same as a victory for the passive one, as his goal is to prevent the active character from reaching his goal.

For instance, if a gard is searching for a hidden character and both have 60% in their appropriate skill, the cases in which the guard finds the other character are when :

-He rolls between 13 and 60 and the other player rolls 61+

-He rolls between 4 and 12 and the other player rolls 13+

-He rolls between 1 and 3 and the other player rolls 4+

For a total of 24.35% chance of success, from which we can substract the 1% chance they get the same result on the d100.

This stuff is complicated. I’m not sure that’s the correct way to combine the different probabilities.

EDITED: Oops, found a silly bug. Updated below

I just worked up a Python script to go through all the possible combinations of guard and player rolls and count up the successs. It came up with 2943 winning combinations for the guard out of 10000 possibilities, or a little over 29%. I didn’t consider fumbles though. I used ‘elif’ to avoid double-counting.

# Chance of success for 60% skill
# Remember range() starts at 0

# g is the guard, p is the player

s = 0 # Success counter

for g in range(99):
    for p in range(99):
        if g <= 2 and p >= 3:
            s += 1
        elif g <= 11 and p >= 12:
            s += 1
        elif g <= 59 and p >= 60:
            s += 1

print s

 

Simon Hibbs

Edited June 4, 2018 by simonh

[Mugen]

1 hour ago, simonh said:

This stuff is complicated. I’m not sure that’s the correct way to combine the different probabilities.

I actually did some errors in my computations, and the right chance of success is 29.03 %, counting 1% chance of tie.

[soltakss]

19 hours ago, simonh said:

This stuff is complicated. I’m not sure that’s the correct way to combine the different probabilities.

EDITED: Oops, found a silly bug. Updated below

I just worked up a Python script to go through all the possible combinations of guard and player rolls and count up the successs. It came up with 2943 winning combinations for the guard out of 10000 possibilities, or a little over 29%. I didn’t consider fumbles though. I used ‘elif’ to avoid double-counting.

Although Probability Theory is fine, I find that this is the best way of working out exactly what the chances are. I did the same thing in Progress 4GL/OpenEdge to work out success chances for RQ6. 

[PhilHibbs]

33 minutes ago, soltakss said:

Although Probability Theory is fine, I find that this is the best way of working out exactly what the chances are. I did the same thing in Progress 4GL/OpenEdge to work out success chances for RQ6. 

I agree, for the basic reason that probability and statistics are hard to get right. Crunching a million sets of numbers is easier, at least for my particular skill set.

Edited June 5, 2018 by PhilHibbs

[Psullie]

guys we all know that that dice gods ignore probability or statistics - at least that's when my games feel like