Stability feedback thread

With the new stability mechanics finished for the moment (of course still due for fixes and rebalancing), it makes sense to focus all feedback in this thread so it doesn't get lost.

This is both for bugs or confusing effects, and the big picture, i.e. things like "X doesn't collapse often enough" or "X always collapses in Y but it shouldn't" or "I want to do Z as X but I collapse".

Edit: I've already seen collapses or partial collapses resulting in weird new city owners, like Persia collapsing and ending up under Greek control, reviving them. Happened only during autoplay for me though so details are welcome there.

Okay, a quick guide for how the stability mechanic works so people can actually judge what's going on.

First of all, stability checks don't happen periodically anymore. They are triggered by one of the following events:
- losing a city
- razing a city
- researching a tech
- becoming a vassal
- vassalizing another civ (+)
- declaring war / being declared war on
- making peace (+)
- changing civics
- changing state religion
- moving your palace
- building a wonder (+)
- starting a golden age (+)
- receiving a great person (+)

Under certain circumstances, the check is canceled because the civ is currently immune to it. The main reasons for this are that the civ has just spawned, just received a crisis and is therefore protected or is in anarchy or a golden age.

During the check, the civ's stability score is calculated. This score only depends on the current state of the civ without any permanent influences. The score is the sum of the results in the five categories expansion, economy, domestic, foreign and military. More details on the score calculation follow later on.

Then, a threshold is determined with to which the score is compared. The threshold is higher at higher stability levels and lower at lower levels. This has the result that unless something changes within your civilization and therefore your score at the next check, you eventually arrive at a certain level and stay there. Here's what can happen:

score > threshold + 10: your stability level increases by one
threshold < score <= threshold + 10: nothing
threshold - 10 < score <= threshold: a crisis occurs
score <= threshold - 10: your stability level decreases by one, a crisis occurs

Checks triggered by events marked with a (+) can only result in a stability increase, never in a decrease or crisis.

A crisis affects your civilization negatively in some way. The type of effect depends on the category your score was worst in, and the severity of the crisis depends on your stability level. The level is reduced before the crisis starts in the last case above, so it's one level more severe than the level you entered the check with. Here's how the stability levels correspond to the crisis levels:

solid -> no crisis
stable -> minor crisis
shaky -> moderate crisis
unstable -> severe crisis
collapsing -> terminal crisis

A terminal crisis always results in complete collapse no matter the type. As you can see above, it's possible to be on collapsing stability level and still never experience a crisis, so it's not guaranteed that you collapse on this level (but very much possible until your situation improves). Also, solid makes you completely immune to crises until your stability drops again. Here's a short summary of the effects:

EXPANSION
Minor: a random city secedes
Moderate: all cities in foreign territory secede
Severe: all cities except your core secede

ECONOMY
Minor: lose 10% of your current research (first in gold, then in beakers)
Moderate: lose 50% of your current research, one turn of anarchy
Severe: lose 100% of your current research, three turns of anarchy, cottages degrade, great people points are reset

DOMESTIC
Minor: one turn of anarchy, 25% chance for all cities to have unrest for three turns
Moderate: three turns of anarchy, 50% chance for all cities to have unrest for five turns
Severe: five turns of anarchy, all unhappy cities secede

FOREIGN
Minor: two random open border agreements are canceled
Moderate: five random open border agreements are canceled, all defensive pacts are canceled, all peace vassals are freed
Severe: all open border agreements are canceled, all defensive pacts are canceled, all vassals are freed
(every canceled treaty also includes a relationship hit and makes the civ not want to talk to you)

MILITARY
Minor: all cities lose their defenses
Moderate: all cities lose their defenses, 25% chance per unit to desert
Severe: all cities lose their defenses, 50% chance per unit to desert, cities secede to civs you're at war with if they're in their target area

Here's a short summary of what can affect your stability score in the different categories:

EXPANSION
- ratio of core vs. non-core population (cities on historical territory are weighted much less than ahistorical, conquered, or foreign core cities)

ECONOMY
- economic growth (change of total commerce compared to 10 turns earlier)
- penalty for trading with richer civs under Mercantilism
- penalty for trading with Free Market civs under Central Planning

DOMESTIC
- relationship of happiness/unhappiness in your empire
- number of unhappy cities
- civic combinations
- contemporary/outdated civics
- religious unity (state religion in all cities, not too many non-tolerated non-state religions)

FOREIGN
- open borders with collapsing civs
- stable/unstable vassals
- defensive pacts with stronger civs
- being the worst enemy of a stronger civ
- having furious relations with someone
- being at war while in Autocracy
- being at war with heathens / brothers of faith while in Fanaticism

MILITARY
- winning/losing wars
- losing military strength compared to 10 turns earlier

How exactly do you define a richer civ for the purposes of economic stability?

How exactly do you define a stronger civ for the purposes of foreign stability?

Are you planning on pop-ups associated with crises?

Leoreth said:

EXPANSION
Minor: a random city secedes
Moderate: all cities in foreign territory secede
Severe: all cities except your core secede

Click to expand...

Losing a city to independence is not what I would call a minor crisis, especially when that city is randomly chosen from within your empire. It would appear that you could be stable and randomly lose your 2nd best city to this crisis, which is not minor whichever way you look at it. Or is it actually possible to lose your capital to the RNG here also? The city furthest from your capital rather than randomly selected would make more sense.

A minor crisis would be to have one or more of your cities go into revolt, not secede.

ECONOMY
Minor: lose 10% of your current research (first in gold, then in beakers)
Moderate: lose 50% of your current research, one turn of anarchy
Severe: lose 100% of your current research, three turns of anarchy, cottages degrade, great people points are reset

Click to expand...

Given that one of the triggers is researching a tech, the penalty of losing some % of your current research would seem not that harsh if you'd only just begun to research the next tech.

DOMESTIC
Minor: one turn of anarchy, 25% chance for all cities to have unrest for three turns
Moderate: three turns of anarchy, 50% chance for all cities to have unrest for five turns
Severe: five turns of anarchy, all unhappy cities secede

Click to expand...

Minor effect does not seem minor to me here either. 1 turn of anarchy with a 25% chance for 3 more turns of anarchy?
Moderate effect doesn't seem particularly moderate either, 3 turns of anarchy with a 50% chance that it is actually 8 turns of anarchy? Wow.

FOREIGN
Minor: two random open border agreements are canceled
Moderate: five random open border agreements are canceled, all defensive pacts are canceled, all peace vassals are freed
Severe: all open border agreements are canceled, all defensive pacts are canceled, all vassals are freed
(every canceled treaty also includes a relationship hit and makes the civ not want to talk to you)

Click to expand...

The random aspect doesn't appeal here. Could this be implemented from the OB agreements with the civs with the worst attitude first?

How long will it take for the cancelled treaty penalty of making the civ not want to talk to you to wear off?

Playing as the US, when Spain lost their colonies (way too early. About 1790), Mexicola (the city east of Tijuana) flipped to me and the message "Americans renounce their foreign rule" popped up. This caused me to be in a war with Spain, when I had nothing to do with that far away city. It's not even a part of my historical boundaries.
Peru needs to get Bolivia when they respawn; they are too weak without it.
Austrian stability is not strong enough, as it had a terminal collapse about 1850, creating a super-Prussia which conquered Poland and defeated Napolean's France while gobbling Italy.
The Ottomans need to be weakened a lot with this new stability system.
The Mughals need to vassalize to England easier. England is always defeated (1700 scenario).
I think, due to the new stability mechanisms, that international congresses should be disabled until they are looked at again.

I suppose nothing stops me know from razing my neighbours' border cities down to gain access to their resources? Furthermore, the current expansion mechanic greatly favours the super-city approach in anything but your core, as those will generally have less population than two normal cities.

After a game I have to agree with Blizzard that the minor domestic and expansion minor crises are too severe while the minor economic crisis is not severe enough. Several turns of revolt in a cultural region (expansion) or X number of unhappiest cities (domestic) seem to fit with minor crisis. For economic I kind of like the idea of improvement decay. Roads/watermills/windmills/workshops fall into disrepair. Hamlets, villages, and towns shrink as urban population dwindles. More severe versions could even include economic buildings disappearing or a depression/recession mechanic that causes a temporary global commerce penalty.

The foreign minor crisis is more of a pain than it ought to be right now because it only ends one side of an open border agreement, not both. I kept finding out that my neighbors had the right to enter my territory but not visc versa which forced me to notice that what appears to be an open border agreement isn't actually one, end the current open border, and wait several turns to resign it. It requires more player time to resolve then I would prefer.

There will eventually need to be some sort of notification for crises. It was very frustrating to realize that several turns ago one of my cities went barbarian.

blizzrd said:

Losing a city to independence is not what I would call a minor crisis, especially when that city is randomly chosen from within your empire. It would appear that you could be stable and randomly lose your 2nd best city to this crisis, which is not minor whichever way you look at it. Or is it actually possible to lose your capital to the RNG here also? The city furthest from your capital rather than randomly selected would make more sense.

A minor crisis would be to have one or more of your cities go into revolt, not secede.

Click to expand...

Core cities are exempt from secession.

Given that one of the triggers is researching a tech, the penalty of losing some % of your current research would seem not that harsh if you'd only just begun to research the next tech.

Click to expand...

I guess you're lucky then.

Minor effect does not seem minor to me here either. 1 turn of anarchy with a 25% chance for 3 more turns of anarchy?
Moderate effect doesn't seem particularly moderate either, 3 turns of anarchy with a 50% chance that it is actually 8 turns of anarchy? Wow.

Click to expand...

The percent chances are per city, i.e. some will go into unrest while others won't. The effect starts right away so you can deduct the anarchy turns where the cities would've been inactive anyway.

The random aspect doesn't appeal here. Could this be implemented from the OB agreements with the civs with the worst attitude first?

How long will it take for the cancelled treaty penalty of making the civ not want to talk to you to wear off?

Click to expand...

Depends on the AI personality, it's the same time they are unwilling to talk if you followed another AI's request to stop trading.

NerfCothons said:

The foreign minor crisis is more of a pain than it ought to be right now because it only ends one side of an open border agreement, not both. I kept finding out that my neighbors had the right to enter my territory but not visc versa which forced me to notice that what appears to be an open border agreement isn't actually one, end the current open border, and wait several turns to resign it. It requires more player time to resolve then I would prefer.

Click to expand...

Oh, that is unintentional, will be fixed.

There will eventually need to be some sort of notification for crises. It was very frustrating to realize that several turns ago one of my cities went barbarian.

Click to expand...

I have added interface messages; are you on revision 554?

Suggestion about crisis

EXPANSION
Minor: Lose half of cities in foreign territory.
This will automatically lead to loss of territory by foreign culture and makes cities much more vulnerable to foreign culture spread.
Moderate: (probabilistically) 50% of Military forces in foreign area becomes rebels or join to foreign military force.
Severe: all your cities in outside of historical area secede and (probabilistically) 50% of military forces in historical area becomes rebels or join to foreign military force.

ECONOMY
Minor: increase inflation rate temporary.
Moderate: increase inflation rate permanently.
Severe: lose all of your currently reserved gold, three turns of anarchy, cottages degrade, great people points are reset.
It will be better to find some ways to reset the inflation rate. I suggest let players available to reset the inflation rate to initial value by paying certain amount of golds after severe crisis occured.

I cannot figure out what 'foreign' stability and crisis means in real world. Please give me some examples or explanations...

Novicenoble said:

I cannot figure out what 'foreign' stability and crisis means in real world. Please give me some examples or explanations...

Click to expand...

North Korea is currently experiencing a foreign stability crisis.

blizzrd said:

North Korea is currently experiencing a foreign stability crisis.

Click to expand...

Easy example though... but I thought that diplomatic disadvantages are already represented as diplomatic attitudes of other civs.

Well I need to think about it more

Leoreth said:

I have added interface messages; are you on revision 554?

Click to expand...

You're right, I missed 554.

This is about 4 turns before the sack of Aden.
I am the Ottomans on a 3000BC M/N start.
As you can see, I am at war with the Arabs & the Moors.
I've pushed them into a very unstable position, and I've hoped to collapse them.



This is immediately after I take the city.
All of a sudden, nearly all of Arabia's Egyptian possessions flip to me,
with the exception of one city, who resurrects the Egyptians
and pits them at war against me, all while giving me WAY more than I asked for.

That's likely the consequence of a severe military collapse. There was an error in the code that gave away all enemy target cities instead of just one as intended.

How do courthouses and stability related buildings fit into the new system? Do they help any particular category, or do they have no stability effects any more?

Swarbs said:

How do courthouses and stability related buildings fit into the new system? Do they help any particular category, or do they have no stability effects any more?

Click to expand...

From my experience, they no longer reasonably affect stability,
however, at least the Courthouse maintenance effect is invaluable
and the others are decent if you want to run an Espionage economy.

Swarbs said:

How do courthouses and stability related buildings fit into the new system? Do they help any particular category, or do they have no stability effects any more?

Click to expand...

Courthouses in non-core cities make their population count less.

By the way, how is your impression of the speed improvements now that the old mechanics are disabled completely? It seems that autoplay goes somewhat faster now at least.

The game goes much quicker now.

I have to say, there is a surprising amount of variety in most of the rolled starts now.
Across 8+ different games, I have seen:

-An Uber-Arabia that occupied the near extent of the Umayyad Caliphate + Russia & Scandinavia with a vassalized Ottoman Empire competing against an Uber-France (all of Europe + Ireland + Quebec) and its motely crew and a third place dark horse in the form of Republic of China with Japan as a vassal competing for supremacy.
-A still kicking Achaemenid Empire that stretched from Vienna to Pakistan and was actually winning against the Seljuks.
-Numerous instances of a powerful and stable Roman Empire.
-One where India circumnavigated the globe in 1200s (so failed to trigger conquerors).
-The first time I have ever seen the Mongols enter historical interaction with China:
They collapsed them, and proceeded to take the whole, and in the 1300s, China respawned (as the Song though and not the Ming).
In the same game, the Mongols had about 2/3 the extent of the OTL Mongol Empire as well.
-Surviving and stable Byzantine Empire that showed no signs of slowing down.
-Various other superpowers that such as Tamils, surviving Greeks and Polish from time to time.

The AI is incredibly aggressive now, more so than before, which I like.
These are all on 3000BC starts, which are normally crazy, but it's never been this crazy before.

Yeah, I'm rolling 3000 BC into the Renaissance as well right now and can confirm this. The variety is nice and I really have no idea where the increase aggressiveness comes from (maybe it's that collapses are now more likely to flip cities to major civs instead of independents which tended to slow them down).

I still think that civs that are "supposed to" collapse survive a bit too easily right now. Especially Rome and Arabia. I think I have to adjust the modifiers for passing the fall year a bit still.