On "leader trait cycling" and how to fix it

[Zentay]

The recent dev diary described upcoming changes to leaders:

Leaders now cost Unity to hire rather than Energy. They also have a small amount of Unity Upkeep. We understand that this increases the relative costs of choosing to hire several scientists at the start of the game for exploration purposes, or when “cycling” leader traits, as you are now choosing between Traditions and Leaders..

In older versions of Stellaris leader hiring costs scaled with empire size. This proved unpopular and was disabled. This in turn resulted in "leader trait cycling" which is hiring and firing leaders until you are offered the one with the trait that you want. I agree that this doesn't feel right. Making leaders more costly to hire is only a partial solution however. Another change that is required is a balance pass on leader traits because so many of them are weak or even useless:

1. Exploration related traits become useless once there is no more exploring to do. The game should not offer us scientists with useless traits. Maybe scientists doing exploration should have a high chance to gain an exploration related trait on level up, and hirable scientists should never have exploration related traits. Or give us a policy to permit or prohibit exploration related traits on scientists for hire.

2. The lifespan-enhancing and experience-enhancing traits are very weak as first trait. My feeling is that it's not worth it to hire a leader with these two traits and hope that they later gain a useful trait. These traits should not appear on recruitable leaders but be something that can be gained on level up. The cost reduction trait is also questionable but who knows how the changes will affect it.

3. Governors in particular have too many bad traits, like tile blocker clearing cost reduction or crime reduction. These can be okay for some time in some situation but this situational usefulness conflicts with the leader level system which rewards keeping the same governors assigned. My feeling is that in the end it only makes sense to use governors with traits that are always useful, like the research or admin related traits. If no governors with such traits are offered, I have to do "leader trait cycling". If you increased the cost of doing so it would cause frustration. In the end governor traits are also not an interesting choice because they don't meaningfully change how the game plays. To make it worse, when you really need the tile blocker clearing cost reduction from a governor, you can just hire them for 1 day, queue clearing of the blockers and then fire the governor.

Most of my use of "leader trait cycling" is with scientists because it's a way to increase research speed. It doesn't make a large difference but I feel bad missing out on extra research due not having a scientist with the right expertise assigned to research. Then I feel bad about all the micromanagement that this creates. So it's time this is changed. The current effect of the various "expertise in field" traits is to increase research speed and affect the draw chance of relevant techs. This could be changed to the following: +1 tech offered from the relevant field of expertise (the game generates tech options as normally, and then generates an extra choice of tech from the field of expertise). Then scientists are merely a way to influence research direction rather than a way to optimize research speed directly.

Rulers traits aren't too bad but there are balance problems as well.

 

[Quinzal]

Maybe they could just let us pick and choose traits we want on our Scientists, or maybe restrict some from appearing in the leader pool? This could be tied to administrative tech, or maybe one of the ascension paths; Psionic maybe, since their whole schtick is mind magic.

I dunno. I've only ever cycled trying to get Expertise: Psionics, so it's never really been a huge thing for me.

 

[Losttruppen]

We have only 4 exploration traits, would it be terrible to add a colour and 5% bonus to 3 of them: say Carefree gets +5% society, Meticulous +5% physics, Archaologist +5% engineering, and then have roamer grant ~3% to everything?

I actually think the "leader cycling" is pretty thematic as far as how a hiring process happens and how much it can cost to find the right employee, but mechanically it is dull and I certainly won't be using my precious Unity to do so.

I would love to see some more random events that provide good leaders like we get promising admirals and a few scientists from anomalies.

 

[Pancakelord]

I'd just go with Leader Modification.

Let us pay some resource (unity? energy? food? research?) to "retrain" a leader, or, instead, let us pick their traits on level up (with an optional "auto-level" button - mostly for generals).

Could even use the species modification screen as a base. its already somewhat similar to what this would need.

Of course, if you're doing this, I'd say go all out and make leader upgrades more structured, maybe with some higher-tier versions of leader traits gated behind certain conditions (e.g. "Corvette Commander - Admiral must lead a fleet and have the corvettes in that fleet kill X enemy vessels = +corvette stats).

• You could also opt to take negative/positive traits e.g. corrupt admirals add piracy to their current system, but reduce upkeep (offbook supply lines).
• Whilst other negatives currently in the game could become timed traits (e.g. "nervous breakdown" for 1800 days this admiral has reduced stats, with a chance to gain this during combat).
• Certain traditions, ethics or civics might unlock special traits too. E.g. Pacifists might get the "Peacekeeper" trait option for Generals, any planet-side armies they lead add 0.5 stability, each, on worlds during defensive wars.

Then it would not much matter what your starting leader traits are (starting age is already pretty negligible). You could just have them all start with a free skill point, gaining more as they rank up, or do special things (like lead the fleet slaying a drake), letting you customise to your needs.

 

[DukeLeto42]

The simple solution is to make a leader pool a proper pool - rather than immediate replacement of offered leaders, they slowly replenish in the pool when lost. Hired (or as it could be changed to, simply "employed") leaders cost unity upkeep, but fired leaders return to the pool so you can't cycle them.

 

[blahmaster6k]

The simple solution is to make a leader pool a proper pool - rather than immediate replacement of offered leaders, they slowly replenish in the pool when lost. Hired (or as it could be changed to, simply "employed") leaders cost unity upkeep, but fired leaders return to the pool so you can't cycle them.

That would just make the game even more RNG dependent on whether or not you get good leaders or not. If you can manage to get a battleship focus ruler and a retired fleet officer governor, you can get 30% cheaper battleships than someone who doesn't. That means you'll have a 42% bigger fleet just because you got good leader RNG. At least the way the game works now you can eventually get good traits and won't be crippled by bad ones, except maybe in the very early game when you need the energy for other things.

 

[Grobb]

This would probably be complicated to implement but... what about offering everyone up front, with randomized costs?

It produces the exact same result of digging around, spending resources until you get the perfect roll, but condensed into one or two clicks. And it's not like you can then "leader cycle" for a better price, because there's no refund. You're just checking to see if you'd rather get your Leader of choice at the offered rate, or settle for some other Leader at reduced cost.

 

[currylambchop]

whats wrong with leader trait cycling anyway?

 

[DukeLeto42]

whats wrong with leader trait cycling anyway?

There's two separate complaints at play:

1. "I can't get what I want when I want it, making for suboptimal play"
2. "There is a clear optimal play, but it's tedious and once you can afford it makes the pool meaningless since you're cycling anyways"

The first one is a complaint about the general game design - namely that you need to make the best play with what is available rather than having a persistent meta. The second is a complaint about how the mechanics don't fully mesh and that using them to their fullest extent is therefore tedious.

 

[Nevars]

How to fix it is easy, just produce more unity, problem solved.

 

[TSBasilisk]

Maybe something similar to the Schools in Surviving Mars? Essentially, you build an Academy of some kind and specify what traits you desire; every year the Academy will generate one leader with one of the listed traits who can be recruited as a separate selection from the mixed pool. The downside is you have to tie up a building slot educating them.

 

[Bezborg]

WE SHOULD DO AWAY WITH LEADERS ENTIRELY!

Hear me out... we change it to "institutions", immortal and non-personal entities attached to the specific context.

Meaning:
- 3 branches of research are academies, engineering, physics, society... these represent the entirety of your civilization's talent, academia and science
- fleets have a "command". an immortal "leader" entity with effects, and it learns as it fights. Can even have an admiral portrait
- armies have their own command entity (would need a change in how armies are made, to something similar to fleets, meaning a "fleet" is an entity, with ships being parts of it
- sectors have their own administration structure, also an entity, personal governors are not necessary except maybe nominally, so they qualify as leaders.

How these things upgrade is through us actively investing (direct transfer, edicts, tech), and over time (old colonies would have better administration than new ones).

So in essence, remove the personal leader entity entirely, and have these permanent structures do what they do, improve over time,suffer problems through events, etc... and personal leaders can be auto generated only for elections, if the govt type is open to any sort of leader.

I think this would be a great addition

 

[Ikael]

I think that the main problem is that the entire leader system based around random traits is not exactly conducive for balance, hence the leader re-trait recycling meta. Lowering leader unity costs at the start of the game is a necessary step (and I still think that making you think hard about wherever spend your unity in leaders VS traditions is a good idea at its core) but at some point, we ought to rethink this whole interchangeable leaders thing.

 

[GloatingSwine]

WE SHOULD DO AWAY WITH LEADERS ENTIRELY!

Do away with manually managing leaders in passive positions.

Any leader who is a passive stat stick should be autogenerated to fill their position, how often they cycle and change traits should depend on governing authority. (Science should get a new way to influence the next pick in the pool, some way to choose a field of study to be more likely to appear at a small cost).

Then generate a new type of leader called Explorers who work like Envoys, your empire just has a set number of them based on ethics and they do all of the active sciency jobs. Investigating anomalies, researching special projects, excavating dig sites, and assisting research.

 

[Archael90]

Maybe they could just let us pick and choose traits we want on our Scientists, or maybe restrict some from appearing in the leader pool? This could be tied to administrative tech, or maybe one of the ascension paths; Psionic maybe, since their whole schtick is mind magic.

I dunno. I've only ever cycled trying to get Expertise: Psionics, so it's never really been a huge thing for me.

I have suggested a formula for leader choosen traits.
It goes like this:
You have an option to select one of generated leaders or to start recruitment of specific leaders. The more variables You will choose for a leader the higher it cost.
Then You have formula:
(Base cost) x ((1/(%*/100)) x a**
Where:
* - is the % chance of that specific variable appear.
** - is just repeat of previous bracket for each available variable.
Example:
You wish to higher a leader from Your main species that is 75% of Your total pops, and want it to have maniacal trait which shows up in 20% of cases. It should have psionic trait which means 100% chance of occuring and brainslug which is 2% chance:
200 x (1/(75/100)) x (1/(20/100)) x (1/(100/100)) x (1/(2/100)) = 66667 unity cost.
Removing main species variable will make this leader costs 50000 unity, and without brainslug its only 1000

 

[Dinkelman]

One easy fix is to add a cooldown on the leader pool filling up.

But, I think leaders should have several traits even at the hiring stage. It could be both positive, negative and neutral traits. It could be like personality traits. It would give them some individuality and it would give the player more to consider. It would be less of a non-choice.

 

[Mastikator]

Add a button to fire a potential leader, it costs the same as hiring them, but it just skips to firing them. Leader trait cycling isn't bad, it's a thing you can do if you have too much unity and not enough good leader traits. But the act of cycling is tedious, so make it not tedious.

 

[Critical Ethics]

This would probably be complicated to implement but... what about offering everyone up front, with randomized costs?

It produces the exact same result of digging around, spending resources until you get the perfect roll, but condensed into one or two clicks. And it's not like you can then "leader cycle" for a better price, because there's no refund. You're just checking to see if you'd rather get your Leader of choice at the offered rate, or settle for some other Leader at reduced cost.

Honestly this. Bring up a list of starting traits and select one (pulling up a list of all valid gene and robot traits works fine for gene and robomodding, should also work fine for leader traits). Pay a small amount up front to start the search for talent then a random number of months later get a list of potentials of various species (weighted toward species with gene and robo traits that boost leaders) with variable costs attached.

Add an option to also choose the species at the cost of increasing the search time and cost based on the number of that species available.

You can even leave the old leader screen intact as cheap walk-ins you can hire for far less.

 

[Franton]

fired leaders return to the pool so you can't cycle them.

Not just fired leaders, but also those left unassigned. We should be allowed to swap leader assignments though, to prevent a swapped out leader to immediately get removed to the pool before getting the chance to reassign them.

Another option, based on some real world countries where leaders get a formidable pension after service, even if they worked only for a short period: assign every leader ever hired a pension that will be paid for a lifetime.

 

[Millbot]

So giving this further thought, rather than potentially clog up the beta thread with something that might not be considered on topic, but at the risk of the devs not seeing this.

Someone suggested having leader hiring costs, possibly upkeep as well, scale with sprawl. This seems like a solid solution to me since that means leader costs can lower at the start, but it never becomes possible for them to be super trivial to hire if the player isn't already guaranteed to win or going nuts with a super unity production build. Also makes running into marauders, leviathans, hostile aliens and even scientist eating events far less punishing.

Now, that doesn't solve the issue entirely because the leader system still isn't great. I'd like to see leader traits be neutral traits with pros and cons, but that is likely going to be a semi or major rework that the devs probably don't have time for right now. I'd suggest the compromise of removing the leader pool bonus stuff from the game and creating three starter traits for each leader type (though IMO generals might as well not exist). These three leader traits would either be neutral or if you don't want to or lack the time, you could also make the pool for level one traits just eager, resilient and adaptable. Those three traits seem like the closest to neutral traits we have right now and could be decent for pick your preference, while essentially killing the desire to cycle. I'd prefer to kill the whole microing of scientists, but killing he cycling problem first can at least make the unity cost setup less of an issue.

I'd also suggest that arrested development get nuked, it has be just obnoxious. Part of this is that it's way too common, but also something that kneecaps a leaders potential. The negative traits really should be results of events. So just a matter of making a few more for scientists and making ones for governors, that prompt the player to make some choices and this can result in getting negative traits and sometimes the player knows exactly what traits they are getting but it's a trait off between getting a boon at the expense of your leader. Admirals and generals are more or less covered by combat, but could also add such events for them.

Main thing is, at level one, there should be zero reason to fish for leader traits, while also ensuring that getting more leaders early game can't severely slow down tradition acquisition.