The 3 phases of pace conference

[kaguravitro]

After thing about it for a long time, the peace conference, witch is very need should have 3 components.

Recognition: every tag should have a recognition value, fix in the start and only disagreed by focus or events. A like value prevent to realise or puppet in a peace conference (but still available for player, focus or event), a mid value depending on the alignment of the operator in the could be for example a satellite in fascist but a puppet in Communist or not available for non aligned (is a genetic example). A high value offer the totality of the options, so prevent many wanted border gore.

War conference (like Yalta): events fixed or dynamics where some values in the peace conferences are presset, like a peace conference style, like Italy and Germany over the peace of territory of se france. If a new non puppet tag get into the war and try to violate a prearrange conference needs a lot of acute and could get a war goal against it.

2 phases conference: a first draft set tag per tag witch countries could affect they territories, for example with enough score you get direct access to puppet a complete territory but could be cheap split the score in 2 countries and fight for the land by score as usual. The types of war goal get very important so the retake core should run first automatic so they should be guaranteed. In the second phases tags witch are still contended sober one by one only with the tags how set interest in first phase. Minor with small amount of provinces should not be splited excerpt by retake core, as mid recognition should poop from all the cores, could be a bit of border gore in some cases but this process should lock up merch of they causes of the horrible gore in late game and even help with some achievements.

I have post some of this ideas in their specific party of the forum, but I believe this needs much discussion. It's very unhappy experience get a gore ending with Russia in a 3 ways war even in historic.

 

[TheMeInTeam]

War conference (like Yalta): events fixed or dynamics where some values in the peace conferences are presset

This can and reliably will result in broken outcomes that make no sense. I know this, because it has resulted in broken outcomes that make no sense.

KR's war between Egypt's faction and Turkey, for example, can straight up steal land from 3rd party participants or force people out of the war who don't want to be out. This is what you get with scripted peace outcomes; they ignore what actually happens during the war. Which is already a big enough problem without adding this extra pitfall.

2 phases conference: a first draft set tag per tag witch countries could affect they territories, for example with enough score

The algorithm for score is itself one of the biggest problems with peace conferences. To the point where I don't think it's possible to have reasonable peace conferences without fixing it.

Making the AI have preference for contiguous territories and not use the more ridiculous TAGs as puppets would probably limit gore sufficiently in most cases.

 

[kaguravitro]

This can and reliably will result in broken outcomes that make no sense. I know this, because it has resulted in broken outcomes that make no sense.

KR's war between Egypt's faction and Turkey, for example, can straight up steal land from 3rd party participants or force people out of the war who don't want to be out. This is what you get with scripted peace outcomes; they ignore what actually happens during the war. Which is already a big enough problem without adding this extra pitfall.



The algorithm for score is itself one of the biggest problems with peace conferences. To the point where I don't think it's possible to have reasonable peace conferences without fixing it.

Making the AI have preference for contiguous territories and not use the more ridiculous TAGs as puppets would probably limit gore sufficiently in most cases.

Prescrict deals is not working, needs a change or new tool to make it works cose noire it does thinks as you describe. That's the point, it could also triggers to make it fall appart and not be used in the peace conference.

Some kind of score is need. Get non prescripted peace deals,I mean dynamic as is pretended actually, it is quiet complex that restricting the choices and the viable tags you for sure get a better outcome. I don't like the idea of just taking territories, puppet usually it is better.

In that sense of taking contiguous land, you will first set to take land form the contiguous tag, this could be easy this way. Maybe also the resources/ic of tags calculated at start the conference could be used by ai to prevent take low value territory and puppet it, usable in African rescramble peace deals

 

[TheMeInTeam]

Prescrict deals is not working, needs a change or new tool to make it works cose noire it does thinks as you describe. That's the point, it could also triggers to make it fall appart and not be used in the peace conference.

IMO, if you want this then it would probably be better to give AI preference for territory in peace conference, rather than bypassing the normal rules.

Right now there is no mechanic for getting war goal/declaring for contested territory taken by someone else in peace deal. I'm not opposed to that though. I'm really not a fan of situations where one nations loses land without having a chance to say "you and what army?" in response to to the new territory claims.

Some kind of score is need.

For sure. It's not trivial to calculate it. The current system creates some busted incentives and (when gamed) can completely break peace deals, so it doesn't need a lot to be an improvement. Making it actually GOOD, however, won't be trivial.

 

[kaguravitro]

IMO, if you want this then it would probably be better to give AI preference for territory in peace conference, rather than bypassing the normal rules.

Right now there is no mechanic for getting war goal/declaring for contested territory taken by someone else in peace deal. I'm not opposed to that though. I'm really not a fan of situations where one nations loses land without having a chance to say "you and what army?" in response to to the new territory claims.


For sure. It's not trivial to calculate it. The current system creates some busted incentives and (when gamed) can completely break peace deals, so it doesn't need a lot to be an improvement. Making it actually GOOD, however, won't be trivial.

It's not trivial and I even don't get how you calculate the score.
Bombing score is horrible

 

[TheMeInTeam]

It's not trivial and I even don't get how you calculate the score.

The wiki is accurate from what I can tell: https://hoi4.paradoxwikis.com/Warfare#War_participation

Notably absent is anything about actually damaging the enemy. You get score taking, but not giving, casualties...which is stupid.

Casually reading the wiki does not fully convey just how broken the "occupations" metric is, however. A favorite example of mine is that you can do this:

• Pick Brazil.
• Declare on Germany.
• Declare on Argentina, so it joins Axis.
• Repeatedly attack and push units off of Argentina's capital, without capturing it. Don't join any factions or give military access, so nobody else captures land there.

If you do this non-stop from around when WW2 starts until Germany eventually folds to allies + Soviets, you will easily be top 2 in score (quite possibly 1st) with many thousands of points from "occupations". You can then use the trick of puppeting a small piece of land and feeding it to your puppet in subsequent rounds to get past the 10x cost for non-faction modifier, and take the majority of land in the WW2 peace conference. Despite never sending a single soldier outside of SA and mostly just beating on Argentina, you can own all of Italy (as puppet) and a majority of Germany this way.

But what if you don't do crap like this? The answer is that you get hosed out of peace conferences frequently, because the AI does crap like this inadvertently. It constantly attacks (successfully or otherwise) onto victory points, often sustains it for 48h to get full value, and will push into/back from provinces repeatedly to multi-count them. The Argentina farm example above is an extreme application of this principle, but it's nevertheless the same principle...score calculations as they are designed right now are nonsense. USSR frequently gets huge takes from WW2 even if the Germans are still knocking on the door of Moscow (or even having captured it!) when Germany caps to allies...mostly due to a combination of bad-incentive casualties score and double-counting those losses with "occupations from losing land score".

But frankly, repeatedly face banging one province in SA, or in Sicily, or London etc is not fun to do. It does not do more to help win the war than you did to reach that point. It is obvious that this is not a legit way to "contribute" to who actually wins or loses fights in HOI 4. I know it won't ever be possible to perfectly align score incentives with what is done to win wars in the game, but IMO this should be made as close as possible, with obvious issues like score farming (inadvertent or otherwise) being addressed.

 

[Lamartine]

I thought that the three phases of Peace Conferences were Denial, Bargaining and Anger. Only after you pass through each phase can you grieve the loss of your hard earned victory.

 

[TheMeInTeam]

I thought that the three phases of Peace Conferences were Denial, Bargaining and Anger. Only after you pass through each phase can you grieve the loss of your hard earned victory.

Unless you do the above, in which case you can reliably "out-contribute" USA, UK, and USSR vs the Axis as Ethiopia by landing in Sicily lol.

 

[davidc929]

The wiki is accurate from what I can tell: https://hoi4.paradoxwikis.com/Warfare#War_participation

Notably absent is anything about actually damaging the enemy. You get score taking, but not giving, casualties...which is stupid.

Casually reading the wiki does not fully convey just how broken the "occupations" metric is, however. A favorite example of mine is that you can do this:

• Pick Brazil.
• Declare on Germany.
• Declare on Argentina, so it joins Axis.
• Repeatedly attack and push units off of Argentina's capital, without capturing it. Don't join any factions or give military access, so nobody else captures land there.

If you do this non-stop from around when WW2 starts until Germany eventually folds to allies + Soviets, you will easily be top 2 in score (quite possibly 1st) with many thousands of points from "occupations". You can then use the trick of puppeting a small piece of land and feeding it to your puppet in subsequent rounds to get past the 10x cost for non-faction modifier, and take the majority of land in the WW2 peace conference. Despite never sending a single soldier outside of SA and mostly just beating on Argentina, you can own all of Italy (as puppet) and a majority of Germany this way.

But what if you don't do crap like this? The answer is that you get hosed out of peace conferences frequently, because the AI does crap like this inadvertently. It constantly attacks (successfully or otherwise) onto victory points, often sustains it for 48h to get full value, and will push into/back from provinces repeatedly to multi-count them. The Argentina farm example above is an extreme application of this principle, but it's nevertheless the same principle...score calculations as they are designed right now are nonsense. USSR frequently gets huge takes from WW2 even if the Germans are still knocking on the door of Moscow (or even having captured it!) when Germany caps to allies...mostly due to a combination of bad-incentive casualties score and double-counting those losses with "occupations from losing land score".

But frankly, repeatedly face banging one province in SA, or in Sicily, or London etc is not fun to do. It does not do more to help win the war than you did to reach that point. It is obvious that this is not a legit way to "contribute" to who actually wins or loses fights in HOI 4. I know it won't ever be possible to perfectly align score incentives with what is done to win wars in the game, but IMO this should be made as close as possible, with obvious issues like score farming (inadvertent or otherwise) being addressed.

I think they took Mussolini's statement about needing a few thousand casualties to be able to sit at the peace conference a bit too seriously.

I think casualties sustained should play a part. Clearly Stalin used that as part of his justification for territorial gains. But he also used the damage inflicted on the Germans as justification. So both should play an element.

Gore doesn't bother me so much. It can give good justifications for a third world war if you want to play on. But I'm sure at the time many people thought the new borders that were fostered on countries were terrible and "unnatural". The only reason we think they are right is because we are used to them.

 

[TheMeInTeam]

I think casualties sustained should play a part. Clearly Stalin used that as part of his justification for territorial gains. But he also used the damage inflicted on the Germans as justification. So both should play an element.

I know what leaders said. But what actually *mattered*? If casualties taken mattered as much as the game implies, China should have been able to demand a lot of land. That's not what happened, though.

I would argue that the actual reason Stalin got what he did was "I have a large number of boots and tanks on the ground in Germany". But PR-wise that's probably not a great thing to say, even when everyone knows it...

I could see maybe casualties taken mattering a little, but it should matter much less than inflicted in game terms, because again we (should) want to align "war participation/contribution" to actions that actually do, in fact, contribute to winning a war in HOI 4. Losing manpower/equipment does not, beyond what gains/damage those losses allowed you to achieve.

 

[davidc929]

I know what leaders said. But what actually *mattered*? If casualties taken mattered as much as the game implies, China should have been able to demand a lot of land. That's not what happened, though.

I would argue that the actual reason Stalin got what he did was "I have a large number of boots and tanks on the ground in Germany". But PR-wise that's probably not a great thing to say, even when everyone knows it...

I could see maybe casualties taken mattering a little, but it should matter much less than inflicted in game terms, because again we (should) want to align "war participation/contribution" to actions that actually do, in fact, contribute to winning a war in HOI 4. Losing manpower/equipment does not, beyond what gains/damage those losses allowed you to achieve.

I agree damage done should be taken into account more. With maybe a 3 to 1 ratio compared to casualties taken for example. I wonder if they maybe viewed the occupation score as the factor that would account for what a country does.

 

[TheMeInTeam]

I agree damage done should be taken into account more. With maybe a 3 to 1 ratio compared to casualties taken for example. I wonder if they maybe viewed the occupation score as the factor that would account for what a country does.

Possibly, but occupation score double-counts combats that way. Specifically, it double counts casualties taken right now, and if casualties inflicted were a factor it would double-count that too.

IMO, it is also not particularly intuitive that "occupations" means "points given from contesting victory points with combats". Normally, one would predict this is scored based on territory captured specifically as a separate category. Since capturing land actually does matter to the outcome of wars, this should probably be kept but reworked so it means what it says on the tin. Though that might also require fixing who gets control of territory upon entering it...that sometimes still doesn't work properly.

For example, as Baltic Union I captured the entire Baltic coast from USSR and was pushing east. USSR then declared on Finland, who joined Axis. A moment later, the game stole that territory between Finland and Estonia from me and gave it to...Germany. We were all part of the Axis, but there's no way that's a competent application of territory control. Though maybe scoring would still be based on units entering the province, rather than who controls it...

 

[davidc929]

Possibly, but occupation score double-counts combats that way. Specifically, it double counts casualties taken right now, and if casualties inflicted were a factor it would double-count that too.

IMO, it is also not particularly intuitive that "occupations" means "points given from contesting victory points with combats". Normally, one would predict this is scored based on territory captured specifically as a separate category. Since capturing land actually does matter to the outcome of wars, this should probably be kept but reworked so it means what it says on the tin. Though that might also require fixing who gets control of territory upon entering it...that sometimes still doesn't work properly.

For example, as Baltic Union I captured the entire Baltic coast from USSR and was pushing east. USSR then declared on Finland, who joined Axis. A moment later, the game stole that territory between Finland and Estonia from me and gave it to...Germany. We were all part of the Axis, but there's no way that's a competent application of territory control. Though maybe scoring would still be based on units entering the province, rather than who controls it...

That might be where two different mechanisms come into conflict. I suspect handing the territory to Germany is necessary for the supply system to work but then would disrupt the occupation score as you say. Whatever system they put in will have ways to game it. But it does need work to get closer to a reasonable peace system.

 

[TheMeInTeam]

That might be where two different mechanisms come into conflict. I suspect handing the territory to Germany is necessary for the supply system to work

I don't see how. It was complete bonkers. Baltic Union + Finland still owned literally all surrounding states. Germany had no hand in capturing that spot. There's just no sensible reason that one state would ignore the usual "we occupy it from x direction" rules and just flip to another faction member randomly mid-war.

In terms of supply, since we were all one faction it would be the same regardless.

This gets much, much worse if it happens across factions (such as wrong capitulations, though fixing score algorithm would help there).