I briefly came back

Democracy 4
1

I’ve been saying on this forum that ministers are a game breaking problem for about a year now. Here’s the problem:

Every minister has a maximum effectiveness which they will ever reach. It is not clearly displayed, but the best display you get is after an election when being invited to shuffle the cabinet. You can see the effectiveness and experience then. When they reach 100% experience, whatever effectiveness they have is the best they will ever reach. The starting ministers usually cap out somewhere in the 90+% range, but any replacements can cap out as low as 65%. The replacement ministers will NEVER be as good as the starting ones. This means that losing one of the starting ministers causes IRREPARABLE damage to the player. Yet, due to the randomized starting loyalties of the ministers, these losses are often unavoidable. Further to this, the loyalties of potential replacements are random, often forcing the player to draw from low on the list and in doing so settle for a low maximum effectiveness.

Why is this a problem? In any game players generally want a challenge, but they want a fair challenge. Players don’t mind losing if their own choices or actions led to the loss. Players do mind losing when they were punished as if they had failed even though they did nothing wrong. Players should be faced with challenges, players should face ambiguous situations where the correct choice is not always clear, but there should always have been a correct choice available. If there are unavoidable challenges then those should be declared up front, or be consistent from one round to another. Having unavoidable pit falls generated randomly which are not consistent from one round to another and are presented as a failure by the player will only serve to annoy the player.

This is particularly true if said pit falls don’t even make sense within the context of the game. For example, if I have been playing to please liberals, capitalists, ethnic minorities and self employed by antagonizing socialists, conservatives, trade unionists and religious people, and I have been doing this for nearly two terms, then the draw of replacement ministers should not all have loyalty to one of the groups I’ve been antagonizing, forcing me to draw from the bottom of the list. These people would simply not be in my party, they would be running with the opposition!

I gave this game a couple more tries because I saw the patch drop, and I’m still tapping out of play sessions for the exact same reason I was a year ago. This minister system is broken and has not changed since this game entered early access, save for one bug fix where they were only loyal to 4 groups. I’ll keep an eye on the patch notes, if minsters get mentioned I’ll check this game out again.

6 Likes
2

Bumping!

3 Likes
3

Thanks for bringing this to my attention. I’m looking at this now.
The game calculates effectiveness as follows:

float basecompetence = GetExperience() * GetSuitability(PCurrentJob);
CurrentEffectiveness = (0.8f * basecompetence) + 0.2f;
POSCLAMP(CurrentEffectiveness);
EffectScalar = 0.8f + (basecompetence * 0.4f);

To explain this:

  • The effect of a minister will sway a policy up or down by 20% in either direction based on effectiveness.
  • Effectiveness itself ranges from 20% to 100%
  • Effectiveness is calculated by multiplying the minister experience by their suitability for the current job.

Further explanation:
Base experience for a minister when they are created is randomly chosen to be between 2 values in simconfig.txt, which are set to 0 and 40%. Each turn that a minister is serving in the cabinet, their experience grows, depending slightly on some autobalancing criteria, but its ± 30% of the default experience gain which is also set in that config file, and is 2.5%
(In other words, in general;, minister experience rises by 2.5% each turn).
Experience is capped at 100%, regardless what level it started at.

So the next factor is job suitability. How is this calculated? Suitability is based on the current job.

When a minister is created they are given a random (0-100%) suitability for every potential job. The game then picks the top 3 suitabilities to show to the player in the select minister screen. These values never, ever change.

So there is no actual code that makes starting ministers any better than any other minister you might hire. They are all treated exactly the same way. So how could this not be coming across? I took a look at the code for selecting ministers at the game start:

At the start of a new game, the AI picks the highest ranking suitability minister for every job. So it goes through the policy group in order, (Starting at Foreign Policy, clockwise round…), and from the pool of ministers, it picks the most suitable one available for each position.

So… The real thing that is being experienced here is actually just that the AI is very good at picking based on suitability, and it has knowledge of all 7 values of suitability for each job, whereas the UI in the game only shows the top 3. In other words, if you pick a minister for a role where they are NOT in the top 3…you are playing Russian Roulette, and could have a minister with 0% suitability for the role.

I agree this is not ideal.
Because there is limited room in the UI, showing the total list of suitabilities might be a bit cumbersome, but I shall experiment and see if I can come up with a good compromise. One idea may be to show the suitability for the job currently being chosen for, when you are on that screen? Like this?

image
image1190×459 114 KB

Thoughts?

3 Likes
4

Though I often get 90+ effectiveness ministers, I do agree there are some frustrating points. I use a heuristic method of checking minister effectiveness regularly and doing shuffles every time I spot a minister with below-the-experience-line effectiveness when his/her experience hasn’t even reached 60%. This lets me detect incompetent ones roughly twice the faster but I still face some problems such as failing to find an adequate replacement or getting stuck at a condition when a sizable effectiveness fall can cause a bad situation such as Environmental Protest.

The time I really feel frustrated is when I’m having no choice but to appoint near-zero experience candidates or ones who aren’t interested in the vacant job. Such ministers immediately mess their ministry and give me headaches. I remember a post complaining about this and I mostly agree to it. As I’ve already wrote in that post, “Democracy is about transfer of power and the system should be prepared for it happening at least once per 4 or 5 years.” So what I’m really concerned is how to have smoother replacement of ministers. It’d be really nice if I can keep one or two candidates at other non-minister jobs and have them gain experience at least to 60% or inertia gets introduced to the effectiveness of each ministries.

Regarding Cliff’s idea of adding a suitability-o-meter, I wouldn’t hate it but want to see more indirect approach such as just adding one or two stars next to desired jobs if they have decent suitability values.

Related Post: Can you spot the minister resignations?

1 Like
5

Just out of curiosity, I often get an impression that it’s kinda easy to meet exactly identical ministers in terms of desired jobs & sympathies when it should be unlikely to happen if it’s chosen randomly. Is there any adjustment factors or correlation between desired jobs & sympathies?

1 Like
6

I think I followed everything you said here, but none of it indicates that I’ve been making any wrong choices.

When I have to pick a replacement minister I look for one who is specifically mentioned as being suitable for the job, and generally from as high on the list as possible. The only reason I would deviate from this pattern is if they are indicated as having loyalty to a voter group which hates me. Again, they should be in the opposition party’s shadow cabinet, not running for mine.

However, when selecting ministers with this method it is not uncommon to see them reach 100% experience with 65% effectiveness, while any starting ministers I’ve managed to keep are in the 93-98% range.

Any idea why this is happening?

2 Likes
7

To be honest no, that should not be a persistent phenomena. There is nothing special about the starting ministers. I guess the only thing is that effectively the game has done a perfectly-informed reshuffle for you at the very start, with the absolute best people picked for each job.

So for example in the middle of a game, when your tax minister quits, say you have a list of potential replacements. It might be that the most suitable of these is still less suitable than the current welfare minister. Ideally, you would no a mini-reshuffle of them into that position, and pick the best candidate for the welfare minister from those remaining, but I suspect people rarely do this.

I think to replicate the suitability of the starting ministers, you need to do a full reshuffle for every time a minister quits or is fired.

I’m going to implement this suitability meter anyway, because its not hard, and certainly better than not having it. Currently that information is hidden from the player. We can assume that when appointing a new minister, your staff will do a comprehensive bit of analysis of the CVs of potential ministers and give you decent information right?

5 Likes
8

Ok I have put this in:

image
image1200×627 161 KB

4 Likes
9

Alright, I that I didn’t think to check the existing ministers to see if they’d be better suited than the candidates, however it could be an easier switch than the game currently makes it. Assigning an existing cabinet minister to a new post should be less controversial than outright firing one or reshuffling the entire cabinet.

A game I would suggest looking at for ideas would actually be Crusader Kings 3. It’s a very different game than this one obviously, but a similarity is that the player has a cabinet to manage. That game does an excellent job of helping the player find the best cabinet, even with the competing definitions of what “best” could mean. In it’s simulation switching two councilors or assigning an existing one to an empty post isn’t controversial at all. Firing councilors outright is a big deal however, it often results in them…uh…lobbying for a change of governance…in feudal ways.

Here, I found a guy who uploaded a stream to YouTube:

If you jump to 2 hours 18 minutes in he has his council open (this guy is really neglecting his council) and opens the menu to replace his spymaster. It shows the identity and stat line of the current spymaster at the top, and a list of potential replacements.
The top of this list has 0 intrigue (purple) skill, but is at the top because he is powerful enough that he expects to be on the council and is pissed off about being excluded. The rest after that are listed in descending order of intrigue skill.
The 3rd option on this list is his current chancellor. It would be a single button click to make these two switch jobs, and neither would be angry about it. He wouldn’t really benefit from doing so, but the option is there.
What he should really be doing is reassigning his chancellor to be his steward. He even has the personal character window for his current chancellor open and can see that 4 diplomacy skill is awful, but 17 stewardship skill is significantly better than the 11 his current steward has, and I’m getting much more annoyed than I probably should that he isn’t doing this.

I hope this example is helpful.

3 Likes