Probability of Success, and Critical Success

Howdy all,

I've encountered a contradiction in the GURPS Fourth Edition roles
text, and I hope someone can point me to something official that
resolves it.


Page B345: "You may not attempt a success roll if your effective
skill is less than 3, unless you are attempting a defense roll …".
So, for any non-defense success roll, any effective skill lower than 3
cannot succeed, thus no roll and no chance of a critical success or
failure.

Page B347: "A roll of 3 or 4 is *always* a critical success.".

So, for any effective skill that warrants a roll, a 3 or a 4 is
*always* a critical success. Even if effective skill is 3, a roll of 3
or 4 still succeeds. This makes effective skill of 3 or 4 *identical*
as far as probability of success: 4 ways in 216, or about 1.9%.

Page B348: "A roll of 18 is *always* a critical failure. A roll of 17
is a critical failure *if your effective skill is 15 or less*;
otherwise, it is an ordinary failure.".

So, for any effective skill level that warrants a roll, a roll of 17
or 18 (also 4 ways in 216) is *never* a success. Any effective skill
of 16 or higher has the identical 98.1% (212 in 216 ways) probability
of success.

Page B556 repeats the same definitions as the above references.


Page B171: The "Probability of Success" table shows that an
effective skill of 3 has only 0.5% probability of success (i.e. 1 way
in 216), and 4 has 1.9% (i.e. 4 ways in 216). This flatly contradicts
the above references, which imply that the probability of success for
effective skill of 3 and 4 is identical, at 1.9%.

The same table *concurs* with the definition above, though, for
effective skill 16 or higher: always the same probability, 98.1%.

The GURPS GM Screen also shows the same table: separate probabilities
for effective skill 3 and 4 contradicting the definitions; concurrent
listing for effective skill 16 or higher.


I can't find any official GURPS material that addresses this
contradiction. (I could have missed seeing it, of course.) I could
guess at the intent of the designers, and I suspect that's what many
people do when encountering the same thing, but such a guess could
easily be wrong. I'd much rather see a positive statement from the
GURPS team resolving the contradiction.

Can anyone point me to such an answer from SJ Games?

--
\ "I went to a general store. They wouldn't let me buy anything |
`\ specifically." —Steven Wright |
_o__) |
Ben Finney

Ben Finney wrote:
> Howdy all,
>
> I've encountered a contradiction in the GURPS Fourth Edition roles
> text, and I hope someone can point me to something official that
> resolves it.
>
This is the same as it was in 3rd edition. Although there was more of an
explanation than in 4th edition.

>
> Page B345: "You may not attempt a success roll if your effective
> skill is less than 3, unless you are attempting a defense roll …".
> So, for any non-defense success roll, any effective skill lower than 3
> cannot succeed, thus no roll and no chance of a critical success or
> failure.
>
A success roll must be equal to or greater than the target number.

> Page B347: "A roll of 3 or 4 is *always* a critical success.".


> So, for any effective skill that warrants a roll, a 3 or a 4 is
> *always* a critical success. Even if effective skill is 3, a roll of 3
> or 4 still succeeds. This makes effective skill of 3 or 4 *identical*
> as far as probability of success: 4 ways in 216, or about 1.9%.

No. You're neglecting the point that the roll must be EQUAL to or UNDER
the target number, therefore if skill level is 3, a roll of 4 is NOT
under the skill level and is therefore NOT a success.

(Info in previous paragraph B347.) " Extremely High or low rolls have a
special effects -beyond those of for normal success or failure -
regardless of your exact margin of success or failure."

You first must roll a success.


>
> Page B348: "A roll of 18 is *always* a critical failure. A roll of 17
> is a critical failure *if your effective skill is 15 or less*;
> otherwise, it is an ordinary failure.".
>
> So, for any effective skill level that warrants a roll, a roll of 17
> or 18 (also 4 ways in 216) is *never* a success. Any effective skill
> of 16 or higher has the identical 98.1% (212 in 216 ways) probability
> of success.
>
> Page B556 repeats the same definitions as the above references.
>
>
> Page B171: The "Probability of Success" table shows that an
> effective skill of 3 has only 0.5% probability of success (i.e. 1 way
> in 216), and 4 has 1.9% (i.e. 4 ways in 216). This flatly contradicts
> the above references, which imply that the probability of success for
> effective skill of 3 and 4 is identical, at 1.9%.

No, it doesn't. You have a .5% probability of rolling that 3 for that
skill level of 3, and for any defense roll under 3. You have a 1.9%
probability of rolling a 3 OR 4 for a skill level of 4.

>
> The same table *concurs* with the definition above, though, for
> effective skill 16 or higher: always the same probability, 98.1%.

Any roll of 17 or 18 (a 1.9% probability) is a failure. That means that
a) on ordinary rolls even with skill levels of higher than 16, you still
only have a 98% chance of success.

b) where minuses are applied, having a skill higher than 16 (17+ range)
still allows the roll to be at a 98% success probability if the
effective skill level is 16 or greater.

c) when doing quick contests, reducing high skill to 16, and subtracting
the same difference from the other character, makes a playable (if not
necessarily more realistic) comparison of skills, and still allows that
1.9% chance of failure.
>
> The GURPS GM Screen also shows the same table: separate probabilities
> for effective skill 3 and 4 contradicting the definitions; concurrent
> listing for effective skill 16 or higher.

Again, the point that 3 & 4 are critical successes ONLY apply if the
target number is equal to or higher than the roll (ie 3 for a roll of 3,
or 4 for a roll of 3 or 4).


>
> I can't find any official GURPS material that addresses this
> contradiction. (I could have missed seeing it, of course.) I could
> guess at the intent of the designers, and I suspect that's what many
> people do when encountering the same thing, but such a guess could
> easily be wrong. I'd much rather see a positive statement from the
> GURPS team resolving the contradiction.
>
> Can anyone point me to such an answer from SJ Games?
>
I don't see this as a contradiction. And the same rules have been in
place since version 1, actually. You first MUST roll equal to or under
the target number in order to succeed. Then, if your roll is a 3 or 4 it
is a critical success. And a 3 is a critical success if rolling a
defense where your defense is less than 3.

By the same token, rolling over the target number is a failure, and
rolling 17 or 18 are always failures. A roll of 18 is always a critical
failure and depending on the target number, a 17 may be a critical failure.

Not an official SJGames ruling, but the way I read and applied the rules
ever since I began with the boxed set in version 1.

Regards,
Roger

PS- House rules, however vary. WE play that a 3 is always a critical (ie
spectacular) success regardless of target number. An 18 is always a
critical (ie spectacular) failure, regardless of target number. A 17 is
a failure, but never critical. Other numbers are merely ordinary success
or failures. (We did away with the 10 over or under skill level rules in
3e.)



---
avast! Antivirus: Outbound message clean.
Virus Database (VPS): 081231-1, 12/31/2008
Tested on: 12/31/2008 6:18:05 PM
avast! - copyright (c) 1988-2008 ALWIL Software.
http://www.avast.com