That "average penalty" will change depending on the morale of the neighbouring provinces.jubjub bird wrote:
I've been thinking about this a bit more and I think the "Neighbors" calculation is actually a big part of the problem.As far as I know, the target morale calculation has the following components:
- A Distance to Capital penalty, that seems to be approximately linear at a rate of -10 morale points for each day of distance from capital, maxing out at -30 once a province is 3 days or more from the capital
- An Expansion penalty, that is related to the number of provinces owned (?) and maxes out at -35
- A boost for Propaganda Centers (starting at +10 for a fully-built L1, scaling up to +23 at a completed L2, and then scaling up to +40 for a completed L3)
- An Enemy Neighbors penalty, calculated as -5 morale points per bordering enemy province
- A Neighbors penalty, which seems to be a variable number based on the current morale of neighboring owned provinces
Starting with the first two, they make sense conceptually and I don't think they're that bad when it comes to gameplay. Given the starting point of 102 morale points, the Distance to Capital penalty and the Expansion penalty, when maxed out, would take target morale down to 37--a reasonable number, low enough to slow down construction and production but not so low as to cause a potential revolt every day.
I think the Enemy Neighbors and Neighbors contributions to morale are the ones messing things up. For one, it's possible for Neighbors penalty to exceed the would-be Enemy Neighbors penalty, as in: it's possible that the morale drop due to Neighbors in an owned province surrounded by other owned provinces is greater than the morale drop if that owned province had been surrounded by enemy provinces instead. I don't think this makes much sense.
Here's an example:
The selected province below is surrounded by 6 friendly provinces and one enemy province.
Look at the morale calculation here. It has an actual morale of 0% and a target morale of negative 6%. The largest contributor isn't Distance to Capital or Expansion, it's the penalty that comes from having unhappy (but still friendly!) neighbors. At 38% for 6 neighbors, this average penalty (6.3%) is larger than the penalty that comes from having enemy neighbors (a flat 5%).
I've seen some target morales as low as -15%, which is an incredible hole to try to get out of: at -15% morale, even a fully completed L3 propaganda center isn't enough to guarantee a province won't revolt.
I think there are a few obvious issues here:
- A more central province will basically always have a lower morale than a coastal province, simply due to the fact that it has more neighbors, since the penalty is on a per-province basis
- This Neighbors penalty creates a negative feedback loop that spirals all the provinces downward. Each province has lower morale due to its unhappy neighbors, those neighbors have lower morale as a result of this province, etc etc
- When it comes to managing morale, at this point in the game it's not even worth it for me to attempt Propaganda Centers. I'm actually better off if provinces revolt because then the penalty is maxed at 5% per enemy neighbor rather than 6+% for unhappy friendly neighbors, AND when I recapture the province the morale gets reset to 25%--higher than it would have been if it hadn't revolted!
I think there are a few principles that would make sense in a solution:
- Don't penalize a province just because it has more neighbors
- Don't allow a province's target morale to get so low that an L3 Prop Center can't prevent revolts--if we want Prop Centers to be a thing, it should always be possible to build your way out of revolt territory (even if that isn't usually the best use of resources)
- An owned province shouldn't be hit with a larger penalty from owned neighbors than it does from enemy neighbors
- We should try to reduce the negative feedback loop that causes things to spiral out of control
Converting the Neighbors penalty from a per-province hit to a single value based on an average, with a max value that makes sense, might help with these. A max of -30 if the neighboring provinces average 0% morale, with enemy neighbors being calculated as 0% morale as well, linearly up to a drop of -0 if the neighboring provinces average 80% or more. This would mean a province surrounded by newly captured provinces, at 25% morale, would get a Neighbors adjustment of -21. With maxed out Distance and Expansion penalties, that'd be a target morale of 16%. A province entirely surrounded by enemy neighbors, with an effective average of 0% morale, would get the max -30--when combined with maxed Distance and Expansion penalties, that'd be a target morale of 2%. We're still getting target morales that put provinces at risk for revolt without getting such extreme results and negative numbers that can spiral things out of control.
I'm still open to the idea of dropping capital capture boosts from 10% to 5% to offset this slower decline in morale, which would smoothe morale a bit over the course of a full game.
Thoughts?
For example, 5 provinces with 20% morale will have a worse "Neighbours" morale effect than 5 provinces with 50% morale.
