Informal game theory and GURPS Social engagements

This post looks at tactics for social engagements using the second draft of the rules.  People playing in an engagement won't go to this length to understand their options: they should focus on having fun. However,  a little head-scratching is in order before the next playtest.

Test Case 1 : Minimal example

In this minimalistic example, two characters are seeking consensus on a policy issue. PC has the initiative.



Let's look at possible options:

Option group 1 - Attacking the issues

PC can either make a supporting argument for a funding cut, or an attack on the "raise tax" option. Both would aim to weaken the gap between the two issues and ease a future sway action. It appears to be preferable for PC to uses his stronger Political skill in the leftmost state. However, NPC will be able to oppose this move using his equally good Politics-14. Let's consider the two possibilities

• Politics-14 vs Politics-14 on "cut funding": 36% chance of transition down by 2pt, 45% of transition down by pt, 8.4% blockback in composure.
• Politics-12 (-2 plane mismatch) vs Politics-12 (-2 plane mismatch) on "Raise tax" : 35% transition down by pt, 45% transition down by 1pt, 19% blowback on composure. 20% re-defining "Raise tax" as a political issue.
• Politics-12 (-2 plane mismatch) vs Economics-10 (-2 skill mismatch) on "Raise Tax": 50% transition down by 2 pts, 57% transition down by 1pt, 13% of blowback on composure. 35% re-defining "Raise tax" as a political issue.

At equal adjusted skills, it is better to go with the better skill to minimize the chance of blowback on an action. NPC quite clearly should try to fight this exchange in the Political sphere to match the initiative launched by PC. Redefining the issue would be useful for PC to re-define the "Raise tax" state with a skill that he is good at and NPC is not.

Option group 2 - Attacking the individual

PC may decide that it may be better to attack NPC in order to sway him more easily in the following round. Politics could be used to explore a credibility shortfall for NPC. In this case, the distance causes a -2 penalty and yields Politics-12. This would be outmatched by NPC's rational counter attack using Politics-14.   Politics-12 vs Politics-14 means 27% chance or causing 1+ point of composure, and 23% of losing at least one.

Option group 3 - Swaying NPC into "Cut funding"

The third option group is for PC to dislodge NPC through the transition by outsting him from his position. PC will have to fight an Economic issue, either with Economics-12 or Politics-12 (-2 for mismatch). To this, PC has to contend with a -2 distance penalty as NPC has his guard up due to the disagreement. This means either Politics-10 or Economics-10. Either way, NPC will defend at 12. Here is the option matrix:

PC / NPC ->	Politics	Economics
Politics	Politics-10 vs  Politics-12 (plane mismatch)	Politics-10 vs  Economics-10 (skill mismatch)
Economics	Economics-10 vs  Politics-10 (skill, plane mismatch)	Economics-10 vs  Economics-12

There is no better rational choice for PC as NPC may make a choice in both cases to maximizes its chances to resist. The sway has 18% chance of crossing the 2pt transition, and 37% chances of causing at least 1 pt of lost composure. Of this 37%, some 5% of the skill contest would lead to a temporary loss of over 5 points in composure. 

From these results, it looks like a sway action is similar to a personal attack: the distance penalty has a significant impact on the odds of success, and of a possible blowback to PC. 

Conclusion

From all this, it is possible to conclude that when two evenly matched opponents are struggling over a simple two-state engagement map, that the safest tactic is to soften-up the opponent's position to negate the distance between the positions. This tactic is safer, but not the most sound. It is a double-edge sword, however, as the transition cost is applied in both direction. To create assymetry, a personal attack has a better balance of success vs loss than a sway action. A rational player controlling PC should then realize that the most promising option is to keep the initiative and say something nasty to NPC to cow him before it gets done to himself. A barely biased flip of a coin, Fighting one's mirror image is never a good idea! 

There is a glimmer of hope: if PC goes into a great tirade and the GM likes what he hears, he could get a +2 for his roleplay, and shift the odds into a  to Politics-14 vs Politics-14: 36% chance of transition down by 2pt, 45% of transition down by pt, 8.4% blockback in composure. Let the political venom roll off your tongue: it's a roleplayer's game after all!