If you can completely redesign Stellaris military systems, which would be your first priority?

[Iosue Yu]

This is a casual discussion. Let's see if we combine each person's first priority, are we going to see a complete and interesting combat gameplay.

My first priority would be to introduce Dog Fights.

It means we should get a focus on "Strikecrafts". There needs to be various types of fighters you can launch. The Amoeboid Fighters will be one of the good ones. And then we will have regular winged fighters, variable fighters from Macross (or a rip-off from it), armoured troopers like from Votoms and Yamato, then a few "larger" robots like Mobile Suits (from Gundam) or Valiancers (from Buddy Complex), and finally some giant robots like Mazingar Z, and even gianter ones, like Gunbuster.

There should be a strikecraft designer like what we have in Ship Designers. Then we can choose a "Hull", from smaller aeroplanes that each squad has more units or a 3-MS Strike Group or just 1 Giant Robot as a single unit.

Dog Fights would be engagements without the Ships. You launch your strikecrafts while the ships continue to sail, until one side is overpowered.

They are used as "uncommitted fights". So if it's a Dog Fight, it's just a Dog Fight against strikecrafts. Then it can be escalated to be a Fleet Battle so both sides just commit everything they have.

Does it improve gameplay? We don't know until we have it. But it improves coolness.

 

[PK_AZ]

My first priority would be logistic aspect of warfare, and making fleets much more bound to empire infrastructure.

 

[InvisibleBison]

If I was redesigning Stellaris's combat, the first thing I'd do would be to remove strikecraft entirely. It's not really a gameplay motivated change; it's more that I just don't like the idea of space airplanes. To me, they seem to be part of a trend of making space warfare the same as naval warfare - specifically WWII era naval warfare - which seems to me to defeat the point of playing in space entirely. Also, strikecraft don't really make for a coherent setting: If it's possible to make strikecraft that can beat bigger ships, why would you spend all the money on building bigger ships instead of just more strikecraft and carriers?

 

[Iosue Yu]

If I was redesigning Stellaris's combat, the first thing I'd do would be to remove strikecraft entirely. It's not really a gameplay motivated change; it's more that I just don't like the idea of space airplanes. To me, they seem to be part of a trend of making space warfare the same as naval warfare - specifically WWII era naval warfare - which seems to me to defeat the point of playing in space entirely. Also, strikecraft don't really make for a coherent setting: If it's possible to make strikecraft that can beat bigger ships, why would you spend all the money on building bigger ships instead of just more strikecraft and carriers?

It's a balance issue. I think one biggest difference between a ship with FTL would be the energy source. You'd expect a ship to have its own perpetual engine and FTL capacities but strikecrafts, and smaller combat robots, the lack of them. And also, the pilots will be a big concern. You generally need a lot of good pilots who can handle big G in combat. That's not true for ships. You can staff a ship with physically weaker individuals and they can still perform well somehow. But for pilots, they really need to undergo intense conditioning, especially under G force.

The other thing is that the cockpit of any aircraft and robot would be too small to live inside comfortably. You'd still need a ship to travel and rest.

But if you want to talk about bigger ships, then what's the point of bigger ships if weapons can only be shot 1 against 1 without causing an area damage?

 

[Iosue Yu]

My first priority would be logistic aspect of warfare, and making fleets much more bound to empire infrastructure.

This is interesting. Care to elaborate?

 

[johnty5]

+Puts on helmet+

Entirely get rid of the ship designer. Maybe just leaving the “ship sections” component for a bit of customisation - but, overall, move towards a system more like EU where new and improved units unlock at various tech points.

 

[Iosue Yu]

+Puts on helmet+

Entirely get rid of the ship designer. Maybe just leaving the “ship sections” component for a bit of customisation - but, overall, move towards a system more like EU where new and improved units unlock at various tech points.

Limited freedom, but would make the AI make fleets that make sense. Also easy to balance. But I wonder how players would react to reduced freedom.

 

[johnty5]

Limited freedom, but would make the AI make fleets that make sense. Also easy to balance. But I wonder how players would react to reduced freedom.

I suspect players would be very unhappy. There’s a large group of players who love it. I just find ship design to be tedious and/or needlessly fiddly. I’ve never looked back on a game of Stellaris and thought “the ship design was fun in that playthrough”.

Agree that fewer ship design options would allow for better AI/balance and allow the complexity to be added elsewhere in the war system.

 

[InvisibleBison]

And also, the pilots will be a big concern. You generally need a lot of good pilots who can handle big G in combat. That's not true for ships. You can staff a ship with physically weaker individuals and they can still perform well somehow. But for pilots, they really need to undergo intense conditioning, especially under G force.

I have never encountered a work of science fiction where some ships are subject to G forces and others aren't. In every work I'm aware of, either they have inertial compensator technology that prevents all ships from experiencing G forces when accelerating (eg, Star Wars, Mass Effect), or they don't have inertial compensators and all ships experience G forces when accelerating (eg, the Expanse, the Night's Dawn trilogy). And incidentally, I'm pretty sure Stellaris falls into the former category.

This is also a good example of how people think of space combat as being the same as naval combat. Naval ships, after all, don't experience significant G forces. In the absence of an inertial compensator, though, spaceships absolutely would. After all, if someone shot a missile at your ship, you'd want to move away from it as fast as you can, which generally means at as high an acceleration as you can get without killing the crew.

 

[Elordis]

IMO, ship combat is good enough not to require both "first priority" and "complete redesign". I'd rather see war exhaustion rework and ground combat rework first.

 

[Iosue Yu]

I have never encountered a work of science fiction where some ships are subject to G forces and others aren't. In every work I'm aware of, either they have inertial compensator technology that prevents all ships from experiencing G forces when accelerating (eg, Star Wars, Mass Effect), or they don't have inertial compensators and all ships experience G forces when accelerating (eg, the Expanse, the Night's Dawn trilogy). And incidentally, I'm pretty sure Stellaris falls into the former category.

This is also a good example of how people think of space combat as being the same as naval combat. Naval ships, after all, don't experience significant G forces. In the absence of an inertial compensator, though, spaceships absolutely would. After all, if someone shot a missile at your ship, you'd want to move away from it as fast as you can, which generally means at as high an acceleration as you can get without killing the crew.

If someone shoots a missile at you your sensible move would be to fire countermeasures and detonate the missile before it is armed, or just before it can damage you.

Evasion is more a dog fight thing.

That's why Ships don't experience excessive sudden G forces since that's what you always want to avoid. But combat robots and fighters are all very G-intensive.

 

[InvisibleBison]

If someone shoots a missile at you your sensible move would be to fire countermeasures and detonate the missile before it is armed, or just before it can damage you.

Ideally, yes, but if those don't work dodging is better than sitting still and letting the missile blow you up.

Evasion is more a dog fight thing.

On planets, yes. Airplanes are much more able to dodge than oceanic ships. But in space, all spaceships are equally maneuverable. A big battleship can dodge just as easily as a tiny fighter (assuming competent ship design, of course). This is why it's bad when people treat space combat as being the same as naval combat: The physics involved are completely different and give rise to different tactical situations.

 

[Pancakelord]

Let's see if we combine each person's first priority, are we going to see a complete and interesting combat gameplay.

Whilst I don't have a priority, there are broadly 3 categories.

1. Ship-design (pre-war),
2. Tactical gameplay (moving ships/micro, automated ship combat)
3. Strategic gameplay (deciding what to attack to get the best victory/least losses)

For 1 Ship-design

• Technical limitations let long-range guns (notably L/XL) do damage instantly, before the other side can fire / before the animation finishes.
  • This leads to a plethora of workarounds - but ultimately each "block of damage" should be applied on a time-delay that matches up with range to target.
  • Artillery builds will be less impressive without needing to use any distortive nerfs/tweaks
• M-slots under perform, and are generally boring.
  • I attribute this to the point above, a lack of extra ship-class specific modifiers (+bonus dmg to corvettes etc) and feel like they'd be a good place to get small stat buffs and put some more interesting weapons like short-mid-range unguided rockets.
• Starbases are red-hot trash.
  • They need scalar modules that level up alongside the tier of starbase (e.g. +100% <>+600% Starbase armor). Scalar because if you give them flat bonuses they end up over or under performing per-alloy cost in certain cases. See this for some more ideas of what I mean.
  • They also need more guns in general - some have suggested making defence platforms share shields with starbases so they last longer, which I like the Idea of.
  • Beyond that, making Def Plats indestructible + disable them for invaders (so you can capture a starport, the starport will fire back at its original owner, but the defence platforms will be dormant until their OG owner comes back) would also lead to them getting more use without the annoyance of rebuilding them often.
• There needs to be more of a reason to use mixed fleets, be this stacking buffs or imposed ship-class limits (either based on ethos or GC restrictions).

For 2 - Tactical

• Combat computer AI instructions are flawed for a few ships. This might not be obvious to people that havent screwed with modding, but Ship AI is entirely driven by the ship_behaviour rules AI in battles, weighted by range of guns, position of Attacker fleet and Defender fleet. For example. this is the default carrier AI. And we all know how the default carriers love to YOLO rush in to battle, right?
• Spoiler: Vanilla ship behaviour

  ship_behavior = {
      name = "carrier"
      preferred_attack_range = 150
      formation_distance = 150
      return_to_formation_distance = 150
  
      combat_target_anchor = root
  
      collision_awareness = 0.75
      collision_radius = 1.0
      collision_strafe = 2.5            
      attack_move_pattern = stay_at_range
      passive_move_pattern = stay_at_range
  
      desc = "CARRIER_BEHAVIOR_DESC"
  }

• Well, here is what we could tell the carrier to do instead (see the #### to indicate my changes)
• Spoiler: Altered ship behaviour

  ship_behavior = {
      name = "carrier"
      preferred_attack_range = 150
      formation_distance = 150
      return_to_formation_distance = 300        #### changed from 150
  
      combat_target_anchor = parent            #### Changed from root
  
      collision_awareness = 0.75
      collision_radius = 1.0
      collision_strafe = 2.5            
      attack_move_pattern = flee                #### Changed from stay_at_range
      passive_move_pattern = stationary        #### Changed from stay_at_range
  
      desc = "CARRIER_BEHAVIOR_DESC"
  }

• What I've done here is
  • let the carrier hang back from the rest of it's fleet (which is actually called a formation in ship behaviours - making me wonder if fleets-of-fleets were ever planned) [150 >> 300]
  • told it to prioritise movement based on its own location rather than that of it's fleet [root >> parent]
  • Told it that when it gets within attack range of any enemy ship [150] actively run away
  • when it gets >150 units away, sit there.
• This makes it so that carriers very very hard to catch for any ships that are out-of-range when i've tested it (to the point that they'd probably need slower engines). So combat behaviours need some serious rethinking.

For 3 - Strategic

• The AI needs to be instructed to "mass troops" prior to declaring war.
  • Right now you can put all your troops on the border and then declare war, immediately blitzing the AI before they can respond (assuming they even have the economy/navy to respond).
  • Teaching AIs (not all, but some) to also move most of their troops to within 1 jump of enemy space, before declaring war would go some way to at least making their opening moves more threatening [and would mean they stop declaring wars on people they cant even get to].
• In much the same vein, the AI needs to understand the concept of stacking fleets. Right now it performs better with bigger (modded for size) fleets, than an equivalent number of ships scattered over small fleets, Just flag them in to an armada and move as one.
  • The 2-tier leader structure in HOI4 is a good example of how this may work, too: In Stellaris terms, you put an admiral (A "Grand Admiral") in charge of several other admirals fleets, which lets the game treat them as one fat fleet.
• Further, borrowing from HOI again, we need real strategic targets in wars that can offer outsized benefits.
  • Taking a Sector capital should reduce morale of all defence armies in that sector by 50%, and maybe provide some intel bonuses.
    • Taking all sector capitals + the main capital of a nation could let you pacify it in multi-empire wars, effectively turning it neutral until its allies can free its worlds.
  • Taking an Agri/Mineral/Generator-world could apply a stack/capped "Basic shipments disrupted" modifier to other worlds in a sector - or worlds within N jumps for 1-2 years/till-recapture them reducing Specialist job output.
  • Knocking out an enemy titan or juggernaut should award flat war score.

None of this is an egregious overhaul (the exception being not-instant-applying ranged damage, which would take some coding), but it would make ships fight more intelligently, close up some statistical gaps that are just too large with ship weapons, and generally give you a reason to prioritise planet X or Y, beyond it just "being on the way to their capital".

 

[PK_AZ]

This is interesting. Care to elaborate?

In current Stellaris, battlefleets are basically independent of any form of infrastructure. They can go wherever they want and stay here as long as they desire (as long as you have open borders). Compare it to Hearts of Iron 4 (where ships cannot execute missions if there is no port nearby) or even Europa Universalis 4 (where ships take attrition damage if there is no port nearby).

What I believe would create better war dynamics, is to have fleets unable to really operate beyond naval bases range, or even in range, for prolonged time. So if you want to invade enemy, you have to make sure you have enough naval bases at the border. If you are afraid of invasion, you want to create extensive network of bases to counterattack. When you penetrate enemy borders, you have to secure naval bases to support your push forward. If enemy is too daring, you can take his base some distance from "frontline" and therefore slow down his operations.

The point is to change perspective: instead of fleets being your mean of force projection, now naval bases are the ones that project your force (i.e. fleets). Instead of fleets just being in some system, they are now assigned to some naval base, and they can be on mission in another star system. Note there is no mechanical change in this paragraph, it is all about the way we see things. It would force some big changes to UX, obviously.

Now, there are at least two glaring problems with that sugestion:
1. Hyperlanes doesn't really work conceptually with such system, warp would be better
2. What counts as naval base is an open problem. Some megastructures obviously. Planets and star bases are obvious candidates, but both bring some problems.
2a. What counts as naval base IN OFFENSIVE OPERATION is even bigger problem.

 

[Iosue Yu]

Ideally, yes, but if those don't work dodging is better than sitting still and letting the missile blow you up.

On planets, yes. Airplanes are much more able to dodge than oceanic ships. But in space, all spaceships are equally maneuverable. A big battleship can dodge just as easily as a tiny fighter (assuming competent ship design, of course). This is why it's bad when people treat space combat as being the same as naval combat: The physics involved are completely different and give rise to different tactical situations.

No. Physics has it that F = ma, and also if you apply a force not directly on the centre of mass, the object spins.

Since your ships are not anchored in space, you need to apply pushes with balances so that it moves towards one direction with its 2 ends balanced. Do be very agile, your ship first needs to be quite light in mass. And then your nozzles have to be placed around its centre of mass, unless you are placing multiple nozzles scattered across your ship. Then your ship needs to be relatively spherical in shape without too much of a rectangular shape else it's a nightmare trying to turn any direction other than just the 4 ordinary directions. If not a nightmare, you still get reduced push since you need to balance your head and tail.

It's equally bad when people think movement in space is free.

Whilst I don't have a priority, there are broadly 3 categories.

1. Ship-design (pre-war),
2. Tactical gameplay (moving ships/micro, automated ship combat)
3. Strategic gameplay (deciding what to attack to get the best victory/least losses)

For 1 Ship-design

• Technical limitations let long-range guns (notably L/XL) do damage instantly, before the other side can fire / before the animation finishes.
  • This leads to a plethora of workarounds - but ultimately each "block of damage" should be applied on a time-delay that matches up with range to target.
  • Artillery builds will be less impressive without needing to use any distortive nerfs/tweaks
• M-slots under perform, and are generally boring.
  • I attribute this to the point above, a lack of extra ship-class specific modifiers (+bonus dmg to corvettes etc) and feel like they'd be a good place to get small stat buffs and put some more interesting weapons like short-mid-range unguided rockets.
• Starbases are red-hot trash.
  • They need scalar modules that level up alongside the tier of starbase (e.g. +100% <>+600% Starbase armor). Scalar because if you give them flat bonuses they end up over or under performing per-alloy cost in certain cases. See this for some more ideas of what I mean.
  • They also need more guns in general - some have suggested making defence platforms share shields with starbases so they last longer, which I like the Idea of.
  • Beyond that, making Def Plats indestructible + disable them for invaders (so you can capture a starport, the starport will fire back at its original owner, but the defence platforms will be dormant until their OG owner comes back) would also lead to them getting more use without the annoyance of rebuilding them often.
• There needs to be more of a reason to use mixed fleets, be this stacking buffs or imposed ship-class limits (either based on ethos or GC restrictions).

For 2 - Tactical

• Combat computer AI instructions are flawed for a few ships. This might not be obvious to people that havent screwed with modding, but Ship AI is entirely driven by the ship_behaviour rules AI in battles, weighted by range of guns, position of Attacker fleet and Defender fleet. For example. this is the default carrier AI. And we all know how the default carriers love to YOLO rush in to battle, right?
• Spoiler: Vanilla ship behaviour

  ship_behavior = {
      name = "carrier"
      preferred_attack_range = 150
      formation_distance = 150
      return_to_formation_distance = 150
  
      combat_target_anchor = root
  
      collision_awareness = 0.75
      collision_radius = 1.0
      collision_strafe = 2.5          
      attack_move_pattern = stay_at_range
      passive_move_pattern = stay_at_range
  
      desc = "CARRIER_BEHAVIOR_DESC"
  }

• Well, here is what we could tell the carrier to do instead (see the #### to indicate my changes)
• Spoiler: Altered ship behaviour

  ship_behavior = {
      name = "carrier"
      preferred_attack_range = 150
      formation_distance = 150
      return_to_formation_distance = 300        #### changed from 150
  
      combat_target_anchor = parent            #### Changed from root
  
      collision_awareness = 0.75
      collision_radius = 1.0
      collision_strafe = 2.5          
      attack_move_pattern = flee                #### Changed from stay_at_range
      passive_move_pattern = stationary        #### Changed from stay_at_range
  
      desc = "CARRIER_BEHAVIOR_DESC"
  }

• What I've done here is
  • let the carrier hang back from the rest of it's fleet (which is actually called a formation in ship behaviours - making me wonder if fleets-of-fleets were ever planned) [150 >> 300]
  • told it to prioritise movement based on its own location rather than that of it's fleet [root >> parent]
  • Told it that when it gets within attack range of any enemy ship [150] actively run away
  • when it gets >150 units away, sit there.
• This makes it so that carriers very very hard to catch for any ships that are out-of-range when i've tested it (to the point that they'd probably need slower engines). So combat behaviours need some serious rethinking.

For 3 - Strategic

• The AI needs to be instructed to "mass troops" prior to declaring war.
  • Right now you can put all your troops on the border and then declare war, immediately blitzing the AI before they can respond (assuming they even have the economy/navy to respond).
  • Teaching AIs (not all, but some) to also move most of their troops to within 1 jump of enemy space, before declaring war would go some way to at least making their opening moves more threatening [and would mean they stop declaring wars on people they cant even get to].
• In much the same vein, the AI needs to understand the concept of stacking fleets. Right now it performs better with bigger (modded for size) fleets, than an equivalent number of ships scattered over small fleets, Just flag them in to an armada and move as one.
  • The 2-tier leader structure in HOI4 is a good example of how this may work, too: In Stellaris terms, you put an admiral (A "Grand Admiral") in charge of several other admirals fleets, which lets the game treat them as one fat fleet.
• Further, borrowing from HOI again, we need real strategic targets in wars that can offer outsized benefits.
  • Taking a Sector capital should reduce morale of all defence armies in that sector by 50%, and maybe provide some intel bonuses.
    • Taking all sector capitals + the main capital of a nation could let you pacify it in multi-empire wars, effectively turning it neutral until its allies can free its worlds.
  • Taking an Agri/Mineral/Generator-world could apply a stack/capped "Basic shipments disrupted" modifier to other worlds in a sector - or worlds within N jumps for 1-2 years/till-recapture them reducing Specialist job output.
  • Knocking out an enemy titan or juggernaut should award flat war score.

None of this is an egregious overhaul (the exception being not-instant-applying ranged damage, which would take some coding), but it would make ships fight more intelligently, close up some statistical gaps that are just too large with ship weapons, and generally give you a reason to prioritise planet X or Y, beyond it just "being on the way to their capital".

This is absolutely fantastic. As for the weapon's part, I feel like Stellaris lacks the "World Building" that SF authors do before writing their stories. We have no idea why some weapons work this way or that way. Maybe first we should get some basic lores on the weapons so we get some designs. Otherwise, it's like the game is trying to please the god of "Generic". So you just have to be as blurry as possible to support the highest number of possibilities.

This is not a problem per se. It's just when we have too much freedom, it often lacks focus. Weapon systems are better designed with a very clear sense of how technology progresses and what and how those magical weapons work.

In current Stellaris, battlefleets are basically independent of any form of infrastructure. They can go wherever they want and stay here as long as they desire (as long as you have open borders). Compare it to Hearts of Iron 4 (where ships cannot execute missions if there is no port nearby) or even Europa Universalis 4 (where ships take attrition damage if there is no port nearby).

What I believe would create better war dynamics, is to have fleets unable to really operate beyond naval bases range, or even in range, for prolonged time. So if you want to invade enemy, you have to make sure you have enough naval bases at the border. If you are afraid of invasion, you want to create extensive network of bases to counterattack. When you penetrate enemy borders, you have to secure naval bases to support your push forward. If enemy is too daring, you can take his base some distance from "frontline" and therefore slow down his operations.

The point is to change perspective: instead of fleets being your mean of force projection, now naval bases are the ones that project your force (i.e. fleets). Instead of fleets just being in some system, they are now assigned to some naval base, and they can be on mission in another star system. Note there is no mechanical change in this paragraph, it is all about the way we see things. It would force some big changes to UX, obviously.

Now, there are at least two glaring problems with that sugestion:
1. Hyperlanes doesn't really work conceptually with such system, warp would be better
2. What counts as naval base is an open problem. Some megastructures obviously. Planets and star bases are obvious candidates, but both bring some problems.
2a. What counts as naval base IN OFFENSIVE OPERATION is even bigger problem.

Yes. Considering flying from one side to another side of one country takes at least a year, probably having a stress on supplying at least Food would be the scene that is lacked right now.

We often say supplying is the secondary battlefield. I concur.

 

[ASGeek2012]

I would like to see the whole "repair/reinforce your fleets from enemy starbases" revisited.

Not necessarily throw away the mechanic completely, but as someone in another thread pointed out, it seems to stretch credulity a bit that you could use potentially incompatible tech to repair your ships. I want to see something in place that will still allow the mechanic to work, but not before you take some other action. Some ideas would be:

1. You first have to have a minimal level of intel so you actually understand how their stuff works (AND the enemy cannot have the "Enigmatic Engineering" perk)
2. You have to first research some tech to allow you to adapt alien systems to your own
3. You have to spend resources and/or influence to adapt a particular starbase to be used as a forward command center, which means the starbase won't instantly be available to you
4. You have to pick an ascension perk to specifically unlock the ability.
5. You have to land troops on the station to keep the alien technicians in line, and even then there's a non-zero chance they will commit sabotage against your fleets.

I really like #3 from above because it encourages strategic thinking. You have to weigh the risks and ask yourself is that position behind the lines secure enough to warrant the time and resources to adapt it. It also gives the enemy a reason to take back that position specifically because it's aiding your advance. Pehaps that combined with #5 makes the idea of using an enemy base something that requires a lot of forethought.

 

[grommile]

To me, they seem to be part of a trend of making space warfare the same as naval warfare

Starfighters are such a stock trope of the exploding-starships genre that once they're in your game, you can never remove them, and the most iconic starfighter sequence of all time is not the Battle of Midway in space; it's Operation Chastise in space.

 

[Xenrek]

Starfighters are such a stock trope of the exploding-starships genre that once they're in your game, you can never remove them, and the most iconic starfighter sequence of all time is not the Battle of Midway in space; it's Operation Chastise in space.

im *assuming* by operation Chastise, your referencing the Death Star Trench Run, yes? if not, enlighten me, because i cant for the life of me think of anything closer in any of the mainstream or cult-classic scifi i can remember

 

[grommile]

im *assuming* by operation Chastise, your referencing the Death Star Trench Run, yes?

Yes.

If there's a more iconic starfighter sequence, I'd love to see it!

 

[Nebbie Zebbie]

First priority? Orbital bombardment being faster (and directly ruining buildings), but defenders having more things to build. Planetary shields should just straightup be a health bar that stops all damage until depleted, and armies should prioritize taking it out. Bombardment should pretty much start right before armies land, and continue while they fight to help them out a ton. Armies would also be made more interesting than just "click to spend minerals and wait a long time" per-planet, instead being things that gain experience training and thus you want them early on the ready, costing a bit of CG and food upkeep.
I say this because whenever I'm doing war, I just let the AI bombard and have my spammed armies go around on aggressive, because bombardment is such a chore and sucks the fun out of actually conquering.

A very close second would be fixing surrender willingness. Right now, you can completely eliminate all enemy fleets and take all starbases, and the AI just won't care that you did, giving you a meager bonus for your fleet being more powerful. On top of that, the "unoccupied systems" counter counts systems where the starbase is occupied, but the planets aren't, which I think is incorrect because it's a sort of double taxation.
Fixing this would make wars where there is a clear winner early not drag on like crazy, helping to limit the time spent at war and establishing rapid victory if you're an unstoppable military juggernaut.

I definitely wouldn't stop there, though. Next up would be having a "crew" resource like the Star Trek: New Horizons mod, so that it's not just alloys and the stupidity of "naval capacity". Then, I'd make peace not force borders open (instead only open to the side that forced a surrender) and add EU4's "exiled" mechanic to Stellaris fleets caught by borders closing on top of them. Then, I'd take a look at fixing the overcomplexity of the ship designer and balancing things (with a special look at making strike craft more a utility thing you almost always want in a small amount rather than just a ridiculous option for starbase defense).