Maybe it’s just power of suggestion, but I feel like they do affect persuasion outcomes relative to known traits - and it’s only the listed success odds that are wrong / not working. E.g my character has calculating, and I swear that I get successful outcomes when I choose talking points with calculating way more often then the listed odds would suggest. In 1.4.
@Badcritter @somethinglightthat
It's unfortunate that there's no way to technically prove such a hunch.
My assumption has always been that traits which light up green should increase the stated (and actual) chance of success for that option, and red ones should reduce it. Since I've been on 1.3.1 for a long time, and I've logged more hours on the forum than in the game for the past two weeks, I finally bit the bullet and moved to 1.4.1 to try and get some data points on this (since trait impact on persuasion is supposed to be fixed in 1.4.1).
Before I upgraded I checked Family Feud persuasion chances on 1.3.1. They are always the same regardless of traits, Charm skill etc:
Option 1: 74% success
Option 2: 9% fail, 9% critical success (increases to 10% with Meaningful Favors perk)
Option 3: 9% fail, 59% success
On 1.4.1, where traits are supposed to have been fixed for persuasion -
I created a character with the Mercy trait, which is one of the influencing traits for option 3. All other traits and skills were grey. Chances were:
Option 1: 71% success
Option 2: 14% fail, 5% critical success
Option 3: 14% fail, 56% success (Mercy green)
I created a character with the Valor trait, which is one of the influencing traits for option 2. All other traits and skills were grey. Chance were:
Option 1: 71% success
Option 2: 14% fail, 5% critical success (Valor green)
Option 3: 14% fail, 56% success
Conclusion:
Even on a reportedly "fixed" version, traits do not change the stated chances of success.
Either my assumption about how it should work is wrong, or the fix failed. Traits providing some kind of bonus or additional chance, the magnitude of which is not visible, is possible. It would take a massive sample size of well documented actual results to confirm or disprove that beyond reasonable doubt.