Diplomacy core reaction – why so favourable?

[sargon0]

The MOO AI can be difficult to deal with but certainly the reaction to your diplomatic proposals can be more favourable than intentioned leading to various exploits. The guide describes the AI’s ‘core reaction’ to your efforts as fairly stable subject to certain temporary (patience) modifiers and the formula presented suggest it may take several attempts to be successful.

Core Reaction theory
The guide covers the conditions for making and accepting each of the possible diplomatic actions.
Option…..Minimum.Accept..Counter..Reject. Formula
NAP…........+15(Rela)…...75+..74 to 50…...49-….core+temp(treaty)+d100
Alliance.....+50(Affa)…..125+..124-100…...99-….core+temp(treaty)+d100
dow…….....+19(Amia).…150+..149-125…..124-...core+temp(treaty)+d100
dow(pact).+19(Amia).…250+..249-225…..224-...core+temp(treaty)+d100
break ally.+19(Amia)…..200+..199-175…..174-.…core+temp(treaty)+d100
Trade…....-100(Feud)….100+..99 to 75.….74-…..core+temp(trade)+d100
Exchange.-100(Feud)….100+..99 to 75…..74-…..core+temp(exchange)+UTA+relations+d100
Peace…...-100(Feud)…….0+..-1 to -25..<-25…...core+temp(peace)+damage+d100
Threaten..-100(Feud)..100+(Appease)..99-75(Cower)..74-25(Ignore)..<25(dow)…personality modifier+relative power

Option ‘dow’ is requesting AI to declare war on another AI, ‘dow(pact)’ if an alliance or nap is in force between the AI’s. Minimum is the minimum current relations (as shown on the Races screen) needed for AI to consider offer. ‘damage’ is positive damage inflicted by you on AI and negative damage inflicted by AI on you (pop=5, mb=3, fact=1, ship=1/2/3/4 based on size). UTA is the ‘Unfair Trade Advantage’ is a modifier of up to +100 if your proposed tech (selected by AI) is twice the tech level of the AI tech in exchange. Relative Power is 0 if you have equal or less total power on the Status screen, up to +100 if you have twice or more total power. To threaten option, Appease is gift without peace/war and cower is a retreat and formal peace treaty (8-15 turns). Which brings us to the key concepts of ‘core reaction’ and the temporary modifiers.

Core reaction is the AI’s fairly stable reaction to your proposals is made up of:
Core1: Base relations – the starting relations on first turn of new game between you and AI as described in the manual. This is far more stable than the current relations as displayed on the Races screen. It only changes by -5 for each war (whoever initiated).
...Relaxed +12
...Neutral 0
...Unease -12
...Wary -24
...Restless -36
Core2: Personality modifier – based on AI personality:
...Peaceful +20
...Honorable +10
...Erratic -40 to +40 (rolled at start of each turn)
...Aggressive -20
...Ruthless -30
...Xenophobic -50
Core3: An adjustment with a maximum of +30:
...tech tribute +5 each
...oath break -5 each time you break a peace treaty, an alliance, a NAP or a trade agreement

Temporary modifiers are a more volatile component of AI reaction to your diplomacy which represent the patience of the AI being tested by repeated proposals on the same turn or soon after a previous proposal (accepted or not). Different temporary modifiers are used for different options:
1. temp(treaty) – for NAPs and alliances
2. temp(trade) – for starting or increasing trade
3. temp(exchange) – for technology exchanges
4. temp(peace) – for peace treaty proposals
5. temp(diplomat gone) – for leaders patience

Temporary modifiers start at 0 but every parlay inflicts a -10 on all these temporary modifiers and -30 in the field actually brokered then they increase back to 0 at +10 per turn. This is why standard advice on trying to get an AI to agree peace is to wait 3 turns after the last refusal to allow the temporary modifier to increase back to 0 to give another decent chance of acceptance.

All this is fine but if core starts off mainly being affected by personality (mostly negative) and the temps start at 0 the chance of the formula described above reaching ‘accept’ is remote. Even a Relaxed/Peaceful would have only a 32% chance of accepting that trade agreement you were offering and you would have no chance at all with the grumpy ones (Xeno/Ruth/Agg or Erratics on a bad year). Reality is somewhat different with the AI accepting most early proposals. I almost forgot that early trade rejects were even possible until one was reported from the Unease/Ruthless Mrrshan in Imperium 14. So what’s wrong or changed?

Core Reaction practical
After scanning the savefile and running a long series of tests I have drawn the following conclusions:
1. Temporary modifiers start the game at 0 then increase by a random 10-15 (approx) the first 4 turns then a more varied increase the following year, reeling in the higher numbers, then by a flat 10 for 5 years if still under 100. This makes the AI less susceptible to diplomatic initiatives in the first few years. The final result is the temporary modifiers stable position is a random number between 100 and 109 and after decrease they automatically return, gradually, to this value range. There is no ‘diplomat gone’ temporary modifier since he flees if any of the temps plus ‘bad’ personality modifier drops to -100 or below - no bonus for Peace/Hon who leave at -100 but Agg/Ruth/Xeno leave at -80/-70/-50. Now an extra 100-109 points in the formulae above gives great chances of early success and little chance for failure.

2. The core reaction includes the current relations not the base relations. This is a more subtle difference which hardly impacts early diplomacy but is very important later on. I can only assume this is a bug because this makes AI reaction, and its own diplomacy if it is affected similarly, far more volatile. The AI will be far more difficult to deal with when angry and far more agreeable when pleased. This might sound good but it makes positive (and negative) diplomacy far too effective since it is counted multiple times – vote, ‘enemy of my enemy’ increase relations and temporary modifiers while tech gifts increase core adjustment, relations and temp(peace)!

3. Minor issues:
NAP accepted from +11 (Neutral is -11 to +11)
Alliance accepted from +51 (Affable is +48 to +59)
No minimum relations needed for DOW, DOW(pact) or Break Alliance but current relations impact on chance of success
DOW(pact) behaves like DOW with same accept value so easier to achieve this than Break alliance (a bug)
Break alliance is for alliances only, not for NAP’s mentioned in guide
On Threaten, appease reaction enforces withdraw and peace treaty just like cower
Peace treaty from war may be 8-15 turns as per guide but enforced peace from appease/cower to your threat is at least 6-20 turns
Threaten must have additional factors or lower limits for success considering rate noticed (can threaten more powerful Xenophobe!)
Breaking a trade agreement is not an oath breaker (the AI will still not be pleased though – see below)
Tech tribute gives a variable gain to core adjustment (+3 to +10 noticed) subject to overall +30 limit
Oath breaker gives +10 to core adjustment (a bug) not subject to +30 limit but you will not be popular for quite a while (see below)
No counters offered to (early) trade or NAPs and accept appears to be 75 & 50 respectively
At least some responses seem non-linear with too many counters to accepts or rejects. For instance with peace treaty offers better relations or personality modifier improves accepts but increased temp just seems to affect number of counters with few if any direct accepts.

Temporary modifier changes
NAP/Alliance – no loss if below minimum relations, else accept or reject adds about -30 to -60 to temp(treaty) and adds -10 to other temps
DOW or Break Alliance request – no change if rejected else adds about -30 to -60 to temp(treaty) and adds -10 to other temps
…Note that where outright reject suffers no penalty, rejecting a counter-offer suffers the full penalty.
Peace treaty – accept or reject adds about -50 to -80 to temp(treaty) and adds -30 to other temps – oddly temp(peace) is not worst impacted
Trade - forget-it adds about -10 to -40 to temp(trade) and adds -10 to other temps, accepting/rejecting trade adds about -45 to -95 to temp(trade) and adds -20 to other temps
Exchange - outright reject adds -30 to temp(exchange) but max of 50 and adds -30 to other temps, accept/forget-it/escape adds about -60 to -110 to temp(exchange) and adds -40 to other temps
Threaten – appease/cower/ignore puts temp(treaty) to -220 and other temps reduced by 100 each, AI dow puts temp(peace) to -230 and other temps to -300, in either case diplomat will be gone awhile
Break Trade agreement – adds -100 to all 4 temps (but no oath breaker)
Break NAP – puts all 4 temps to -300 and marks you an oath breaker, diplomat gone a long while
Break Alliance – puts all 4 temps to -300 and marks you a double oath breaker, diplomat gone a long while
Break Peace treaty (by action not dialogue) – temp(peace) to -120 and other temps to -190, diplomat gone but no oath breaker!
DOW by you (action not words) – same as break Peace treaty but damage caused impacts temps also. If you have a treaty in place then AI will suffer an oath breaker penalty against you!
Dialogue warning! The guide says there is no penalty if you exit dialogue before finalising action but, apart from penalties noted above for trade & technology exchange, note that entering ‘Threaten/Break Treaty or Trade’ dialogue adds -100 to all temporary modifiers even if you immediately exit without actually breaking any treaties or issuing a threat!
Positive/negative diplomacy affecting relations also impact temporary modifiers eg. inflicting damage on an AI which reduces relations will reduce treaty, trade and exchange temporary modifiers by the same amount and increase peace temporary modifier by one quarter of the loss (more likely to agree peace if he’s hurting).

Trade offer accept chance at defaults
………...…Peace..Honor..Errat..Aggre..Ruthl..Xeno
Relaxed…100%..100%..100%..100%..100%...92%
Neutral….100%..100%...99%..100%..100%...80%
Unease….100%..100%...97%...98%....88%...68%
Wary…..…100%..100%...92%...86%...76%...56%
Restless…100%..100%...87%...74%...64%...44%

Note some conclusions are from direct information in the savefile, some are from repeated tests subject to the Moo pRNG. Let me know if you see or suspect something different from that mentioned.

[RefSteel]

Incredibly comprehensive job on this! I just read it for the third time to make sure I understand the implications. A few questions:

Do the temp modifiers' annual increase from zero toward 100-109 begin on first diplomatic contact, or at the start of the game?

If I'm reading the way temp modifiers work correctly, it seems you can ask for breach of alliance (or DoW) any number of times until you get a counter-offer or acceptance (or give it up as a lost cause) without chances dropping due to failures. Is that right, or is there something I'm missing? (As an aside, I'd noticed that DoW requests against allies have better odds than break alliance requests, and I'd considered a personal "house rule" forbidding me to ask for a DoW against a race's ally - in other words, requiring me to successfully request breach of alliance before requesting DoW - but the prolificness of early AI cheese alliances has thus far convinced me to retain the direct option.)

Does Oathbreaker affect relations with all races, or just the one with which you originally received the penalty? Also, I presume that as AI dialogue suggests, it doesn't matter who broke the NAP or Alliance - the player is an "Oathbreaker" whenever the treaty is broken by either side. Is that confirmed, or still just assumed based on "Why should we trust one who has broken so many treaties?"

[sargon0]

Temp modifier increases begin at game start - they will all reach their steady state (100-109) by 2310, unless disturbed by some action.

Any request/offer hits the temp modifiers (and so increases chance diplomat will leave) so any subsequent attempt will have less chance. I have to exit the dialogue and save to show this but if you repeat you can see double and higher penalties accumulating for multiple attempts. On your aside, I would agree a house rule to avoid DOW(ally) although point taken about the early AI cheese alliances. The guide considers the chances so low it says you should 'break alliance' first then DOW.

Oath break only affects relations with the individual race. Both races count of 'oath breaks' against each other is incremented but only one side gets their 'core adjustment' changed (in wrong direction). The interesting bit is who gets blamed. If you directly break NAP/alliance then you get blamed. If you 'cause' a war and the AI dow's then the AI takes blame although does it matter. If you get blamed (and it worked correctly) the AI would be less responsive to your diplo. If the AI gets blamed does it change your attitude which is decided by you not by the computer.

Another interesting bit is effect of being Honorable. The guide says that all oath break (and diplo point) penalties for diplo against an Honorable leader are doubled. However I have seen evidence that the penalties are doubled for diplo by an Honorable leader. This makes more sense. An Honorable leader would not expect much from, say, a Xenophobe, but anyone would expect an Honorable leader to live up to his word or be offended. The difficult part is that this means you need to know your generated personality to know how offended the other races will be with your actions and it is not listed!

[Dvd Avins]

Thank you. I didn't even know if anyone besides me still played this game, and I only break it out once every couple of years. But I was just reviewing the guide as I was playing a game and realized the formulas they had just don't add up. The way they had it, the maximum possible modifier for a non-human player was 57 (talking to a pacifist, with whom one had given 6 technologies for free, with whom was originally favorably disposed) and many rolls of d100 needed a 200 for success. You'd think Emrich and Hughes would have noticed that and corrected it, but I don't think this book is Emrich's best work.