Idea group balancing

[TheMeInTeam]

EU4 should be balanced for single player.

In cases where MP balancing significantly differs from SP balancing, changed values should be an option in the game setup.

If you need to balance differently between SP and MP, you are making an AI that doesn't try to win, however a particular game defines winning.

I disagree with that design choice even at the theoretical level. If the AI is stepping into the shoes that are occupied by human players otherwise, it should be at least attempting the same objectives, albeit likely less effectively due to developer/programming resource limitations. Making an AI that game throws and then "balancing" around that seems to me a like a recipe for undue extra development work and asking for mechanical imbalance just to offset the design imbalance (opponents that don't try).

I've seen people try to claim that an AI that tries was a failed experiment in Civ 5, but that AI certainly did *not* try to win. It's actually pretty rare to see it in modern strategy titles.

What's most perplexing is that straight up weird balance choices often get blamed on "MP-focused balancing" even though there's no credible rationale for that and some counter-evidence...especially apparent in moves like "buff trade companies", "buff survivability of large nations", or "nerf the weakest positions on the board without coherent basis or any internal consistency", all examples straight from EU 4's checkered patch history. I'd argue that had there *actually* been a MP emphasis on patch changes with some coherent framework, these changes would not have happened as they did and EU would be better off right now.

 

[arosstheriver]

If you want a fun rebalance of ideas, and a lot more customization in your countries, via idea groups, then look in to the mod Idea Variation. It presents a massive amount of power creep...but power creep that everyone has access to...unless you're non-European, due to tech deficiencies.

 

[kramsikrams]

If you need to balance differently between SP and MP, you are making an AI that doesn't try to win, however a particular game defines winning.

In my opinion the only idea group that sucks inherently is expansion (why did they remove the CB?). I would never pick economic, naval or maritime. The latter 2 might be MP choices for England, Brunei and someone who owns the entire new world. In SP I would say that idea groups are far from being the largest AI weakness as the AI only have one opponent that is smarter than it iself so picking "mp idea groups" might not be that dumb.

 

[KRBLACK]

If you need to balance differently between SP and MP, you are making an AI that doesn't try to win, however a particular game defines winning.

I disagree with that design choice even at the theoretical level. If the AI is stepping into the shoes that are occupied by human players otherwise, it should be at least attempting the same objectives, albeit likely less effectively due to developer/programming resource limitations. Making an AI that game throws and then "balancing" around that seems to me a like a recipe for undue extra development work and asking for mechanical imbalance just to offset the design imbalance (opponents that don't try).

I've seen people try to claim that an AI that tries was a failed experiment in Civ 5, but that AI certainly did *not* try to win. It's actually pretty rare to see it in modern strategy titles.

What's most perplexing is that straight up weird balance choices often get blamed on "MP-focused balancing" even though there's no credible rationale for that and some counter-evidence...especially apparent in moves like "buff trade companies", "buff survivability of large nations", or "nerf the weakest positions on the board without coherent basis or any internal consistency", all examples straight from EU 4's checkered patch history. I'd argue that had there *actually* been a MP emphasis on patch changes with some coherent framework, these changes would not have happened as they did and EU would be better off right now.

I think you misunderstood. The game should be balanced for single player first and IF there are massive imbalances in multiplayer due to the way the game is balanced the multiplayer build could be tweaked a bit.

Perhaps changing 5% discipline in idea groups and policies to 3 or 4% (just naming a random figure).

I do agree with you that the AI should be more competent, however I also know how much work that takes.
The same AI runs for every country if I am not mistaken (France and Ming have the same AI code is what I'm trying to say), so the AI you are trying to make has to:
-Be applicable and adjustable to every country on the map.
-Advanced enough to know when to contest a player or great power. This can range from trying to take the curia, HRE, EoC or simply taking land from a great power.
-Know when to concede defeat and take its losses.
-Be lightweight enough to not make the game run terribly slow.

I have very little knowledge about programming and AI, but the way I see things more code= more ''intelligence'' (assuming the code works and is not unnecessary).

 

[a_sophist]

I think you misunderstood. The game should be balanced for single player first and IF there are massive imbalances in multiplayer due to the way the game is balanced the multiplayer build could be tweaked a bit.

I took the argument to mean that if there is an imbalance in multiplayer it's because the AI has been built to role play or play in some other sub-optimal manner, which is undesirable in a strategy game, hence the comparison to Civ, where leaders have "personalities" that often force the AI to play poorly for the sake of flavor. That being said, for me a sense of realism and the ability to craft a narrative are what I'm primarily interested in with EU (I'll take chess for a strategy game that stresses mental faculties) so I don't mind sub-optimal flavor as long as it serves a coherent purpose for some goal that makes sense for the way a country is being played. I'm not saying that blobbers shouldn't be able to blob, of course, rather that I appreciate mechanics that serve a variety of approaches to the game. Perhaps, though, this is the best argument for balancing around MP, since that format is much more "win" oriented and is necessarily more military quality dependent than single player.

 

[grommile]

That is damning confirmation that the idea groups are imbalanced in SP.

Let us consider two idea groups A and B, which are competing for a player's attention.

Both of them are worth taking in MP between players of (roughly) similar skill, depending on your situation.

One of them is (almost) never worth taking in SP, because you can outplay the AI without it.

What this tells us is not that the idea groups are unbalanced in SP, but rather that the AI is bad at the game.

The correct fix is to "balance for MP" (whatever that actually means in practice) and improve the AI. The challenge is to improve the AI in ways that are still (a) fun to play against and (b) capable of running at acceptable speeds when instantiated 200+ times on your customers' existing computers.

EDIT: oh, and (c) does not cause the immersionist camp in your player base to ragequit en masse.

 

[KRBLACK]

Let us consider two idea groups A and B, which are competing for a player's attention.

Both of them are worth taking in MP between players of (roughly) similar skill, depending on your situation.

One of them is (almost) never worth taking in SP, because you can outplay the AI without it.

What this tells us is not that the idea groups are unbalanced in SP, but rather that the AI is bad at the game.

The correct fix is to "balance for MP" (whatever that actually means in practice) and improve the AI. The challenge is to improve the AI in ways that are still (a) fun to play against and (b) capable of running at acceptable speeds when instantiated 200+ times on your customers' existing computers.

EDIT: oh, and (c) does not cause the immersionist camp in your player base to ragequit en masse.

Fair point.

But lets be honest Naval, Expansion and Espionage are lacking.
Expansion is only worth taking when colonising, and in this case Exploration is always a better pick. (I know you can double down on colonisation but still)
Naval only make you be able to win naval battles. Wars are rarely decided by naval battles and even when they are the +50% naval forcelimit from Maritime will suit you during war and peace.
Espionage just requires so much micromanaging unless you are taking this group for the passive bonuses the 100% spy network grants you. In case you want siege ability just take Offencive, if you want less AE then take Influence.

 

[TheMeInTeam]

I think you misunderstood. The game should be balanced for single player first and IF there are massive imbalances in multiplayer due to the way the game is balanced the multiplayer build could be tweaked a bit.

Perhaps changing 5% discipline in idea groups and policies to 3 or 4% (just naming a random figure).

I do agree with you that the AI should be more competent, however I also know how much work that takes.
The same AI runs for every country if I am not mistaken (France and Ming have the same AI code is what I'm trying to say), so the AI you are trying to make has to:
-Be applicable and adjustable to every country on the map.
-Advanced enough to know when to contest a player or great power. This can range from trying to take the curia, HRE, EoC or simply taking land from a great power.
-Know when to concede defeat and take its losses.
-Be lightweight enough to not make the game run terribly slow.

I have very little knowledge about programming and AI, but the way I see things more code= more ''intelligence'' (assuming the code works and is not unnecessary).

I disagree at the fundamental level; you want balance first around options such that good players don't consistently pick the same stuff when playing against other good players.

This is for options, not start positions. Those are imbalanced by necessity.

 

[Novacat]

The problem with Naval, Maritime, and Espionage ideas is that navies and espionage cannot win wars. Wars are won exclusively by land combat and land sieges, so you need everything either going into land combat, or things that can support land combat, or things that can enhance your ability to control greater tracts of land.

Thus, the current meta.

This will not change unless the meta was changed so that you can win wars with espionage or navies alone. That is probably not going to happen.

Perhaps changing 5% discipline in idea groups and policies to 3 or 4% (just naming a random figure).

This would probably make countries that have discipline in their NIs/Governments even more powerful, and would not change the meta one iota.

 

[brifbates]

SP has different priorities compared to mosh-pit MP games. Idea groups that don't take this into account get left behind.

Core cost modifiers aren't the only way to make things engaging. Give Expansion its CB back along with +5% admin efficiency and it goes from being trash to being a very interesting strategic alternative to picking Admin every game.

Expansion doesn't need its CB back, it needs to go back to being a diplo point group like it always should have been. The only idea in the group not diplo point related is regiment recruitment speed which is military, it has no business in the admin list.

Admin is going to be must have as long as it remains the only real source of RCC and doubles down on merc spam bonuses to boot. There's no getting around that.

 

[KRBLACK]

The problem with Naval, Maritime, and Espionage ideas is that navies and espionage cannot win wars. Wars are won exclusively by land combat and land sieges, so you need everything either going into land combat, or things that can support land combat, or things that can enhance your ability to control greater tracts of land.

Thus, the current meta.

This will not change unless the meta was changed so that you can win wars with espionage or navies alone. That is probably not going to happen.



This would probably make countries that have discipline in their NIs/Governments even more powerful, and would not change the meta one iota.

I understand this is the current meta, which is why Naval, Expansion and Espionage should be balanced. Being able to win wars solely with espionage or naval power would be ridiculous, but at least make Espionage a good support option.

I would like to see Espionage give you something like +1 monarch point when one of your allies' monarchs has better stats than you own monarch prioritising your lowest stat.
Example: Own ruler is a 1/2/2 and ally's monarch is a 5/5/5 you would gain +1 ADM/month. This would of course not stack multiple times.

Another option would be to give Espionage lower prestige and legitimacy cost for disinheriting heirs or abdicating monarchs. Perhaps a simple -20% strengthen government cost modifer could be added.

If PDX is willing to change the espionage system altogether then perhaps siphoning income could be added. I'd say you would receive 10% of a nation's income at 100% spy network whilst undiscovered. This would give the system a fair (I think) boost. The finisher of 50% rebel support efficiency could be changed to +50% income from siphoning. This would grant you 15% of a nations income while you can maintain enough spy network strength. I understand this will be very frustrating when 20 AI nations have over 80 spy network, which is why I think it could either function like trade power and dividing the spoils, or only the nation with the highest network strength receiving the gold to make spying more competitive.

I could also see Espionage being changed into ''offencive diplomacy'' allowing you to lower the AI's reasons to join wars against you. Of course this is very risky and would require serious balancing, but I can't count the number of times I have played OPM's and wanted to lower a nation's reasons to join a war against me by just 2 or 3.

A simple option that I have suggested before in multiple threads is to give the age ability of allowing fabrication a claims bordering claims to Espionage as a finisher. This would be fitting and usefull since you can avoid having to declare no-CB-wars to conquer far away lands.



Naval will probably be overkill unless naval warfare is drastically changed, until then Maritime is better in every way in my opinion. That said, Naval ideas could just be totally removed and spread between other idea groups so that every military idea group has one or two naval boosts.

And I know changing discipline in MP would not necessarily work, but it was just a random example of a quick and simple change to balance multiplayer differently than singleplayer.

 

[Avernite]

The big use of espionage used to be the ability to infiltrate your enemy and see exactly where their armies are.

In SP, this is basically never relevant because you need to start as a single 3-dev province minor to still care where the AI's armies are by that point. An option would be to simply lift all tech-based restriction on spy actions if you have the espionage group... seeing where the Ottoman armies are for that first war could be rather useful. Seeing it for your first war against Kanem Bornu is a bit less obviously useful.

 

[PhoenixG]

I used to love espionage when they put the advanced action in to idea set. But no, pdx wants to be "fair" and put into tech. :<

 

[KRBLACK]

I used to love espionage when they put the advanced action in to idea set. But no, pdx wants to be "fair" and put into tech. :<

I'm sorry I don't quite understand what you are trying to say.

 

[hajutze]

I'm sorry I don't quite understand what you are trying to say.

Some of the Dip tech actions were part of the espionage group after its rework. Emphasis on "were".

 

[misiceman]

Naval and maritime have nothing to do with "meta" when it comes to single player, and no amount of "rebalancing" will fix this. The naval AI is just not capable of using fleets properly. until this specific aspect gets "fixed" balancing those groups will be useless and quite frankly a waste of time. Also seeing as they do work for some countries in MP, and can work really well, the groups themselves seem to be working fine when you can operate your navy competently.

Espionage's problem is its a poor mans diplo. They cant make the powers of espionage crippling to other players because what happens when 10+ nations use that power against one power repeatedly or specifically the player (hint it is severely un-fun - as it was before). They did change the group for a reason, but quite frankly it still isnt good when compared to diplo or influence. Also I just dont think the group can do what people want it to do. If its a hybrid group it will by necessity not be as optimal as a specialized group (look at aristo/pluto - or expansion) and in EU4 if you want more power, stacking like bonuses instead of spreading around bonuses is just so damn powerful.

Martimes finisher of repairing at sea is the real deal in the conversation i think. It CHANGES play style (quite frankly i would have it also not consume sailors while on coast as well). Doing something like that for espionage that alters play for the user of the group without effecting other players (AI or not) and I think it can turn into a cool group that has benefits that would make it engaging for the player to pick.

 

[PhoenixG]

I'm sorry I don't quite understand what you are trying to say.

The action slander merchant, sabotage recruitment, sow discontent, sabotage rep, agitate for liberty and infiltrate administration that you get from dip tech were in the espionage idea group when they rebalance the idea group for the first time. But few patches later they move those action from idea group to dip tech with the reasoning "everyone should use it"

 

[KRBLACK]

The action slander merchant, sabotage recruitment, sow discontent, sabotage rep, agitate for liberty and infiltrate administration that you get from dip tech were in the espionage idea group when they rebalance the idea group for the first time. But few patches later they move those action from idea group to dip tech with the reasoning "everyone should use it"

Thank you, I didn't know this.

Naval and maritime have nothing to do with "meta" when it comes to single player, and no amount of "rebalancing" will fix this. The naval AI is just not capable of using fleets properly. until this specific aspect gets "fixed" balancing those groups will be useless and quite frankly a waste of time. Also seeing as they do work for some countries in MP, and can work really well, the groups themselves seem to be working fine when you can operate your navy competently.

Espionage's problem is its a poor mans diplo. They cant make the powers of espionage crippling to other players because what happens when 10+ nations use that power against one power repeatedly or specifically the player (hint it is severely un-fun - as it was before). They did change the group for a reason, but quite frankly it still isnt good when compared to diplo or influence. Also I just dont think the group can do what people want it to do. If its a hybrid group it will by necessity not be as optimal as a specialized group (look at aristo/pluto - or expansion) and in EU4 if you want more power, stacking like bonuses instead of spreading around bonuses is just so damn powerful.

Martimes finisher of repairing at sea is the real deal in the conversation i think. It CHANGES play style (quite frankly i would have it also not consume sailors while on coast as well). Doing something like that for espionage that alters play for the user of the group without effecting other players (AI or not) and I think it can turn into a cool group that has benefits that would make it engaging for the player to pick.

Well balancing could also mean just removing the groups entirely or merging the two together. If Maritime would give some of the boosts of Naval then I'd say that is worth it. Meanwhile another military group could be added instead of Naval. I have seen plenty of good suggestions Logistic ideas for example.

I'm not saying removing the group is the way to go, but it might be. I have never picked up Naval ideas, and I have never seen anyone defend it on the forum. (Which is saying quite a lot seeing how triggered people can get when you insult their favorite idea groups )

And adding the ''Allow fabrication of claims bordering existing claims'' to Espionage would add this unique mechanic I think. It just depends on what the developers are planning. Do they think the idea groups are fine as is, are they planning to rework them in one of the coming patches or do they want to revamp naval warfare and espionage first before touching the idea groups?

 

[Tavior]

I most certainly am willing to embrace the idea that MP vs SP balancing idea group should be different.

The only real question left is how?

 

[misiceman]

Because they ARE useful in MP. Well naval is. Even limited in scope, what point does removing the group actually do? Just leave it as is until it can be worked on (or the stuff attached to it that is kinda bokred). If its a false choice... well it kinda isnt. Naval is a great example... it does EVERYTHING it does VERY well.... its just that in SP its not needed.... unless you want it, because flavour. Is it an optimal pick... well no, and in the end that's fine too. Whenever the naval AI gets some help naval COULD be better for some nations but probably just better for the nations its already geared toward.

IMO this game will NEVER be 100% balanced because its not supposed to be. This isnt competitive starcraft. That and many players choose non mass conquering/optimized builds because they are having fun (I do both, WCs and just playing for stupid goals/rp reasons). Sandbox games are like that. Naval and maritime fall into this perfectly (so does expansion but hey, that group does need some help though). The problem in the logic is espionage... which will now sorta fall in that category but it really isnt "espionage" in that case with -liberty desire and - corruption, but hybrid trees are also like that (aristo/pluto, espionage, influence, expansion). I kinda like the idea of the age ability to form chain claims on the end of the espianage group, that would be cool.

Now if something is totally out of whack (which IMO only humanist falls into as it should be VERY powerful in Dharma) it should be brought down a bit or modified, or other picks brought up, but they cant just bring up the rest of the groups to humanist level, there is enough power creep already.

Finnaly MP vs SP doesnt have to be balanced separately, different groups will have higher priorities in one vs the other because they play very differently. Especially competitive - but it is true with coop as well, and in the end accomplishes the same thing. Asymmetrical starts and real world diplo will have larger effects then most idea groups... unless Prussia or equivalent with all necessary ideas..... but then its the other players fault. Or if players can not coordinate properly, but again thats in play not "ideas". In the end MP isnt just Dev clash style MP, you can play even that many different ways and none of them are "right", so we need one rule set that applies properly for the game mechanics, not meta mechanics.

I need to stop wall of text.....