Right and wrong. That behavior is the result of empire size dramatically affecting how people play not the bonuses/negations. The reductions are a must have because empire size exists and its too punishing to ignore. People wouldn't be pursuing -100% empire size builds if it wasn't so ridiculously important to manage it. Seriously any other title would see how people are flocking to empire size reductions and probably change empire size to be less stupid. People are taking total crap traditions strictly because the empire size reduction is included. Its an unhealthy system for the game. I'm in the minority here but realistically the old system was just straight up better. Anyways point being, removing the negations and doing nothing about empire size itself is a massive mistake.
As if god whispered in your ear to give you this idea. A non intrusive system that wouldn't actively hurt playing the game normally. It would be way better. Sadly i doubt anyone would go for this.
Yes, but since reduction stacking has a huge impact, I think it fundamentally has to be addressed as part of reworking the system. If we're going to try to balance building tall verses wide, there's going to have to pretty large swings in what tall verses wide gets. But I think we should also be ready for the fact that no system is perfect, it isn't like Stellaris tries to be carefully balanced. I am not saying some level of balance isn't a good goal, but tight balance is just never going to happen. Tbh, the whole system bothers my sense of aesthetics because realistically bigger is better. The overhead increase for management of bigger sizes scales really well in the real world. Though, in science fiction, we often see small empires that are just that good -- the Time Lords would be a classic example, even if they'd also be a fallen Empire (they were never large even before they became stagnant), but there are plenty of others. And a lot of players don't enjoy a lot of expansion or persuing it aggressively.
Stellaris does probably handle Empire Size in the most heavy-handed way of pretty much any game I've ever played. I know many have said it feels bad and it does. It makes you want to hit it with as many hammers as you can. Heck, the fact even small empires can easily go beyond 100 is something I still find strange.
And frankly, it's weird that the big things it affects are just tech and unity. If there was going to be a penalty, shouldn't there be some sort of administrative overhead at least from managing a large civilization?
I think some factors to consider are:
1. We can probably replace a lot of the current malus system with a system that starts positive, goes to 0 at a certain size (500, 1000, or whataver), and certain techs increase the bonus if it is positive or reduce it if it is negative, but the 0 point always stays at the same size.
2. Current system punishes players who have smaller planets or habitats, or small empires that get a lot of systems. There's no reason for this. It adds nothing to the game. They already have to micro more and defend more places. Get rid of the empire size effects of systems and number of colonies, or make the colony cost depend on how good the planet is (e.g. size and if it is a habitat, ring segment, ecu, regular planet, etc) -- but even then, what is really giving power isn't the planet existing, it's the districts and pop, so it seems to me size should mostly be just about those. Have the district effect modified by what kind of planet it is though, since that just makes sense. To make advanced planets still good, their bonuses and the cost and effort of getting them can be adjusted -- influence cost for an ecu could probably be removed, for instance.
3. Really need to get rid of how support districts make larger colonies exponentially better than smaller ones. It's just a bad idea. Sure it empowers civs with a small number of large colonies, but it also does so for large civs, and weirdly encourages players to not colonize good planets based on size -- which is just weird if we think about how a real empire would behave. Probably should just have a strict ratio of support to production district, say each one gives a +40% bonus to 4 districts or something, and more don't do anything.
4. Having increased admin costs, stability costs, and bigger organized crime would make sense though. Even machines can have problems with too much data flow for their capacity to manage, leading to erroneous behavior in various applications (I'm a network engineer in real life). I think a combination of energy, mineral, and trade overhead, as well as having jobs that provide stability more necessary and the ability to scale the jobs the buildings provide with policy or automatically with need, would make a lot of sense. So at a certain point, large empires need a building or two on each colony for management, basically, and proportionately have a much higher amount of resources going to keep things running smoothly. I say jobs should scale to an extent because I could see this becoming a little ridiculous if you need to keep putting in more buildings, so I'd think at most two different buildings would make sense, and they just do more jobs or maintenance goes up and output goes up for the jobs if you get bigger. It also means you don't have to micromanage old colonies so much. I do think there absolutely have to be effects that need to be managed locally.
And I don't think this would feel bad the same way that the current system does, which feels extremely arbitrary. But that's just me spending too much time thinking something up off the top of my head. It probably has problems, but I think it's good to think of other ways things could work.