@Albatross

I read a missuderstanding about brackets between the lines of your posts, forgive me if I'm wrong.

Imagine there are 2 brackets :
0% taxation until 1 000 000
50% taxation above 1 000 001

If your income is 1 000 000, your gain will be 1 000 000 since all is taxed at 0%
If your income is 1 000 002, your gain will be 1 000 001 since 1 000 000 is taxed at 0% and only the 2 gold over 1 000 000 will be taxed at 50%

So there is no playing the meta. If you increase your income, you always increase your gain. There is no reason to stay below a bracket. If you can have more income, you will earn more.

Edit: There is the drawback of how randomness is handled in the current test version, where earning more can lead to more loss. I feel ike this is an unintended case of maths, that will likely disappear when Goldo reads through these pages

I will continue to play with the spreadsheet and show some case figures for gameplay with the current system.
And based on this propose new metrics that keep the idea of the system without punishing as much as it is today.

@Albatross: The beauty of percentage-based taxes, especially a flat rate, is that you don't really need to know anything about exact income values - the tax will adjust itself.

And Goldo's been struggling with getting similar data to balance around for years. What makes you think there will be enough 'typical' income information to balance around?

As to abuse, it's the fixed rates that suffer the worst. As A&O said, taxing a proportion of your income that goes into the next bracket doesn't make much of a difference when you're straddling the line. But with fixed sums, it's all or nothing and you'll either end up with a thousand-bracket monstrosity or situations where you do lose money for earning more. And significant amounts at that.



@Pekpuuu: Interesting idea, but I think it's somewhat too arbitrary to build your brothel around. I'm of the opinion that customers are what should be driving your diversification, not orders from on high. The 'which girl to choose' part of matchmaking is a tiny step in that direction, but it's not nearly as strong as it could be.

For some data collection, I'm gonna start a new playthrough on Normal difficulty.

Each chapter I'll take the time to fill all rooms in the brothel and rank at least most of my girls up to the highest available rank.

I'll document my income values during tax collection, as well as my last daily incomes before and after chapter change.

As always with a rogue, because charisma/love based brothels are my way to play the game. And I won't go into Fox if avoidable, to not cheese income by stealing/doubling. I still will take golden girls with Fox of course.

Might take some time, but maybe some hard values cound do some good, even if it will be from a a single sample size. Though don't expect me to have the values within some days, BK playtime is limited and random.

Until soon^tm

Quote from: Alpha_and_omega on Mar 11, 2025, 09:35 AM@Albatross

I read a missuderstanding about brackets between the lines of your posts, forgive me if I'm wrong.

Imagine there are 2 brackets :
0% taxation until 1 000 000
50% taxation above 1 000 001

If your income is 1 000 000, your gain will be 1 000 000 since all is taxed at 0%
If your income is 1 000 002, your gain will be 1 000 001 since 1 000 000 is taxed at 0% and only the 2 gold over 1 000 000 will be taxed at 50%

So there is no playing the meta. If you increase your income, you always increase your gain. There is no reason to stay below a bracket. If you can have more income, you will earn more.

Edit: There is the drawback of how randomness is handled in the current test version, where earning more can lead to more loss. I feel ike this is an unintended case of maths, that will likely disappear when Goldo reads through these pages

I will continue to play with the spreadsheet and show some case figures for gameplay with the current system.
And based on this propose new metrics that keep the idea of the system without punishing as much as it is today.

I'm sorry, I'm not quite sure I 100% understood all you are saying, but your idea sounds good. Trying to explain what I meant more clearly I had something like this in mind:

0-1000 : No tax
1001-3000: 150 tax
3000-5000: 400 tax
5000-7000: 800 tax

And so on. The numbers are made up, but it would be small brackets with nothing that drastic. But if you replace it with your idea where each bracket is taxed upon separately in % rates and then charged cumulatively it would work even better. If applying a difference % in each part of the income is possible I think it is definitely a great way to go. Even better if the rates vary with game difficulty.

My main idea wasn't so much based on the use of flat taxes, but rather with the use of direct income brackets without factors like time pressure since while I understand they are meant to stop people from stalling, they can end up snowballing against the players who are already "behind" in profit from where they should be in that point of time.

Quote from: Jman on Mar 11, 2025, 10:58 AM@Albatross: The beauty of percentage-based taxes, especially a flat rate, is that you don't really need to know anything about exact income values - the tax will adjust itself.

That is not really the case since the current tax is percentage-based and people are still having trouble with it. While I now agree with the argument that percentage is overall better than my initial flat rate idea, any decision whether it be flat or % will only have meaning if we have at least some notion of how it is affecting the player's experience. 

I imagine if someone makes twice the amount they need to buy everything they want, even if they get absolutely ripped off with a 50% tax rate they will still be fine. But to other players what might sound like a reasonable rate at first, might lock them out of buying/accessing many parts of the game or getting enough money to progress as it did for me. You need information no matter which route you go.

Quote from: Jman on Mar 11, 2025, 10:58 AM@Albatross
And Goldo's been struggling with getting similar data to balance around for years. What makes you think there will be enough 'typical' income information to balance around?

In this case, the data doesn't need to be a spreadsheet or anything too professional, The feedback and suggestion from the players itself is the data. Which is something that will always be around.

I'm just throwing out what I think it could be a reasonable solution (Though keep in mind I'm a layman at this so don't hold it against me if it's a bad one)  based on my specific experience with the tax system.

But even if he doesn't use anything from my suggestion and finds a much better way to do it, my suggestion and description of the problems I found in my experience with the system is still part of the data in that sense. Other players are doing the same and that will give Goldo an rough idea of what the 'typical' experience is, at least within the limited data he has access too (Gotta work with what we got).

Quote from: Albatross on Mar 12, 2025, 02:24 AMI'm sorry, I'm not quite sure I 100% understood all you are saying, but your idea sounds good. Trying to explain what I meant more clearly I had something like this in mind:

0-1000 : No tax
1001-3000: 150 tax
3000-5000: 400 tax
5000-7000: 800 tax

And so on. The numbers are made up, but it would be small brackets with nothing that drastic. But if you replace it with your idea where each bracket is taxed upon separately in % rates and then charged cumulatively it would work even better. If applying a difference % in each part of the income is possible I think it is definitely a great way to go. Even better if the rates vary with game difficulty.

My main idea wasn't so much based on the use of flat taxes, but rather with the use of direct income brackets without factors like time pressure since while I understand they are meant to stop people from stalling, they can end up snowballing against the players who are already "behind" in profit from where they should be in that point of time.

For your given numbers, the brackets would be like this :
0-1000 : 0% -> up to 1000g will be taxed at 0% for max 0 tax
1001-3000 : 7,5% -> up to 2000g will be taxed at 7,5% for max 150 tax
3001-5000 : 20% -> up to 2000g will be taxed at 20% for max 400 tax
5001-7000 : 40% -> up to 4000g will be taxed at 20% for max 800 tax

And these taxations already scale with difficulty.


What we should not forget, taxation is a very new feature that is only in the 0.3 test version. And for now Goldo has not changed the bracket intervals, difficulty/time penalties etc. We are effectively playing with the alpha version of this feature. And Goldo knows this, as he has opened this thread to discuss it.

So all of your expeciences playing with (or suffering from) taxation are incredibly valuable in this thread!

I will try to collect real ingame data, so that the discussion can be enhanced/supported numbers and not only feeling. And that the brokenness of some existing features (Random +/- 25% on total tax to pay !!!) can be proven by these numbers

@Albatross: Your example brackets already demonstrate the issue: when you make 1000 gold, you get all of it. When you make 1010, you get 860! Make more to earn less...

Percentages ensure this problem won't arise.

Also, 800 tax from 2000 income at 5-7K is not a small tax, either, at 40% for that bracket. And such small brackets will mean you do end up with a multitude of them and thus making an overcomplicated mess that'd be better approximated with percentage-based brackets anyway.



For the second quote, currently the problem isn't due to percentages not adjusting to income. They do. It's about how high these percentages can become. They still adjust to income without needing to know the exact values, it's just that Goldo went overboard with how high they go. This is not a problem, it's a feature. A tax system that doesn't let you set how high the tax is going to be is useless.


And I absolutely reject the idea that you should be guaranteed a reasonable amount of income no matter what your pre-tax earnings were. If it's impossible to do well without extreme optimisation while getting taxed a similar proportion or less (and currently, a poor player does get taxed considerably less overall), it's the sources of income, money sinks and overall balance that need to be looked at, not only or primarily the tax rate. An even slightly balanced game doesn't let your income differ by orders of magnitude.

A management game should be about bettering your management, not auto-coasting to victory. Perhaps a 'very easy' difficulty setting, but not the core experience.


Basically, I'm wary of any tax rate over 50% because even at 50%, you'll be taking at least twice as long to make the same amount of money. It's a significant blow, but more or less bearable. Going further means the time and effort needed to make whatever you need is progressing faster and faster (a 1% from 90% to 91% adds ~10% more time/effort), and any setbacks you suffer can easily put you in the red.


Now, you're right that progressive taxation does need information about where to put the brackets. A flat tax rate without brackets ('non-progressive taxation') doesn't, and that's actually what I personally like most. Easy, less anachronistic, very transparent and doesn't 'punish' the player for doing well.



Finally, any useful data does need to be numbers. Even something like Duke Greene's screenshots over at F95. Just saying 'taxes are too high' is the kind of vague and almost useless feedback that can drive a developer nuts, at best making them tweak things blindly and at worst they'll just ignore it.


My personal takeaways are that

• the time pressure modifier needs to go or at least be restricted to higher difficulties;
  
• the RNG needs to be capped at least;
  
• tax rate progression needs to slow down and not go to 95%, at least not for people who do not play challenge games (and personally, I'd do away with progressive taxation altogether and introduce a flat rate per chapter/difficulty);
  
• something needs to be done about amassing gold before tax day and getting hit by security - bigger safes, a vault at the Banker, partial payments, something;
  
• the game needs more gold sinks that at least give some eye candy in return, and preferably something more tangible.
  

"Your example brackets already demonstrate the issue: when you make 1000 gold, you get all of it. When you make 1010, you get 860! Make more to earn less... "

Oh wait, that's not how it goes.

The first 1000 you make will always be at the first %.. 0 in this case. So only that 10 will be taxed at the 7.5%. So 1010 would be taxed at .75 gold

It's like having buckets.  A 0% bucket that's 1000 gold sized, a 7.5% bucket that's 2000g sized, ext. You fill up the first bucket then move on to the next, then ONLY tax each bucket at the amount listed.

So under Alpha's plan, if you earned 5000 gold:

You fill the 0% bucket first with 1000 gold.  0 tax so far.

You still have 4050 gold so next you fill the 7.5% bucket.  2000g goes in.  That's 150 tax so far in total.

You still have 2050g left so next up the 20% bucket.  Fill another 2000g in there. +400 for 550 tax .

So 5000 gold would be taxed for 550 in all.

If you made another 50 gold for 5050 earned in all, That's a new bracket. ALL of the rest of the gold stays where it is. ALL you do is take the 50 new money and dump THAT in the 40% bucket. That's 20 more tax.

So 5000 gold was 550 tax. 
And 5050 gold is 570 tax.

So no amount of money suddenly ends up with you making less than before. But as you make more the NEW money made gets a bigger chomp.

What part of 'Percentages ensure this problem won't arise' made you think I don't understand this?

As has been said the most important thing is to actually cap it at 95% or whatever number is decided. Simplest way is to remove the rng or apply it individually to each tax bracket, capping at 95% rather than after.

Next way to improve this would be to have a whole month to pay the taxes. Currently you're told on 15th what you will have to pay on the first giving you only 15 days to come up with the money. This means you will only make half of your monthly income unless you start saving before you know the amount.

There are ways around this though. First is getting taxgirl as a trainer. This is time gated to hell though and if you rush to the last chapter you'll be looking at 95% tax rate for 9+ months as your love slowly increases every month. This could be fixed by giving us a way to boost her love faster, maybe gifts or something.

Second way around it is untaxed income, as far as I know taxes pretty much only include tips from work/whoring and streetwhoring. This means you could sell slaves, do quests or sell items etc to make more money. The problem is that these give very little. Doing 5k quests you won't make a dent in 22 million taxes with it. Even selling girls gives very little compared to their daily income.

The only real way that scales? TJB special. You make 10% tax free return on your investment every week. If done 3 times a month you get 33.1% of your initial investment per month. I do it 3 times a month to give a week of leeway to make sure I don't have to pay taxes while my money is locked in the challenge. The only requirement is making 100k a week to complete the challenge which at the point where taxes become relevant is trivial.

Thank you, I read everything and it was very interesting!

I still stand by my original proposal, though. I am not convinced the feature needs a complete overhaul, when we haven't yet made a pass at basic rebalancing.

A few comments:
- "95% tax is too high": this is an absolute maximum that would only kick in in a normal game when you reach Elon Musk levels of money (or should, under my proposal). Sure, it means you may only keep hundreds of thousands of denars out of millions of outcome, but the game doesn't have enough money sinks for you to spend all that extra gold anyway so it's kind of a moot point. And if you think that number is outrageous in the context of Zan's deliquescent monarchy... please consider that if most of you don't currently live under a King, it's likely because they pushed their luck too far on taxation.
- "Progressive taxation is hard to understand/anachronistic": Yes, but the player doesn't need to understand it since they are presented with a lump payment each month, and their is enough RNG baked in that it's pointless to concern themselves too much with the particulars. I use progression under the hood because it is a tried and true technique to make taxation non-linear while ensuring there aren't threshold effects, and it is easier for me to wrap my head around when balancing various brackets (and open to player customization).
- Player customization: Do note that you can already adjust most tax values directly in BKsettings.rpy (you can even set a single tax bracket if you want), so feel free to play around with other values.
- Difficulty levels: I need to run some numbers but it seems we have a problem with difficulty modifiers not changing the outcome enough between easy and hard difficulties. Maybe I should consider having a hard cap that is different according to difficulty (and keep the 95% cap for the hardest difficulty levels).
- I like the idea of giving the player something in return for taxes to soothe their pain at least a little, but it can't be over-ambitious or I will get bogged down with it before finishing 0.3. I might have an idea that will gamify taxes just a little.

Quote from: Goldo on Mar 18, 2025, 02:53 PMthe game doesn't have enough money sinks

That's the main issue, really.

Quote from: Goldo on Mar 18, 2025, 02:53 PMmost of you don't currently live under a King, it's likely because they pushed their luck too far on taxation.

Yeah, and why is that guy still on the throne and there isn't a merchant league of free cities stealing 95% of the trade from Zan and waging a cold war on the high seas at the same time? You can only squeeze the golden goose so much.

Kings and feudalism in general kinda sucked at taxation and never even dreamed of progressive taxation, because these were his buddies getting hit with that. And once we're past feudalism and hitting the nobles in the pocketbook, any King's power starts going away real fast.


Renaissance Elon, never mind the real thing, is never going to pay 95% or even 50%, one way or the other.

Quote from: Goldo on Mar 18, 2025, 02:53 PMI use progression under the hood because it is a tried and true technique to make taxation non-linear while ensuring there aren't threshold effects, and it is easier for me to wrap my head around when balancing various brackets (and open to player customization).

A flat percentage rate per chapter is all that without the above drawbacks.

Quote from: Goldo on Mar 18, 2025, 02:53 PMhard cap that is different according to difficulty
giving the player something in return

Those I can get behind, if you're dead-set on progressive taxation.

Quote from: Jman on Mar 18, 2025, 03:24 PMRenaissance Elon, never mind the real thing, is never going to pay 95% or even 50%, one way or the other.

Yep, but you're just a lowly brothel owner who got lucky. The actual renaissance Elon Musks are the ones squeezing you (but maybe by the end of the game you'll be able to change that, when the story gets that far...). Lore-wise, it's fair to say not much of the tax money eventually makes it to the King's coffers (which is also not entirely historically inaccurate).

And I mean very wealthy or not, if the powers that be wanted your shit... They'd get it.

QuoteA flat percentage rate per chapter is all that without the above drawbacks.

I may be misunderstanding you, but a flat percentage rate is the very definition of linear?

Flat rate per chapter. New licence, more benefits, bigger cut to the Guild.


Edit:

Quote from: Goldo on Mar 18, 2025, 03:36 PMif the powers that be wanted your shit...

I have no problem with the taxmen taking your shit. This is how it used to work in Rome, your possessions were assessed and either the taxmen or more likely the people who bought the rights to tax you came and took a percentage (well, they likely did the 'assessing', too, at that point). A small percentage of a few percent.

If the Guild took a cut of any excess you had, that would work just as well if not better to curb overflowing income and would encourage the player to be proactive with their money.

Ok, I get it now! My gripe would be that a flat percentage would introduce more friction when switching chapters (people already say the gap in facilities from the old brothel to the new can be hard to overcome), as well as some odd min-max incentives (such as waiting to complete chapters until the day after taxes are calculated...).

I did run some numbers, and it appears that the big problem with the current system can be summed up to variation. Taxes can vary much more wildly within the same difficulty setting than the mean does between, say, normal difficulty and highest difficulty. It isn't right and my proposed solution plus some tweaks here and there should address that, and make taxes much more consistent for a given difficulty level.

If you wanted to emulate flat tax rates per chapter, though, you could easily do it; change the settings in BKsettings.rpy to the following:
## GUILD TAX ##

    tax_brackets = [(10**12, 0)] # Neutralize the progressive tax rate

    tax_chapter_penalty = {1 : 0, 2 : 0.05, 3 : 0.075, 4 : 0.1,5 : 0.125, 6 : 0.15, 7 : 0.2} # Change the chapter penalties here to the desired value

Edit: Changing the cap down from 95% would make a moderate difference for average difficulty on the higher end of the tax range only, and none at all for lower difficulties, so it isn't a very influential parameter compared to the other tweaks I suggested.

Well, you could give a grace period for the tax ramp-up. Like the Guild takes a month to process the licence, and you can't buy higher-ranked slaves or rank up during that time.

You could also introduce a delay between moving and start of operations, so you always have a full month before taxes are due. Although... If you're only taxed for the time you've spent in the new chapter, it shouldn't be too bad anyway.

A lot of the change-of-chapter financial pressure has been toned down already, and personally I'm a great fan of the 'start over' feeling.



Moving the cap down might look small potatoes until you realise the brackets are already pretty damn high (most of the way to max) at 50-100K. In a game where you're eventually making millions.

And, in any case, the biggest problem here hasn't been that what you get after taxes is too little to actually live off of - although it seems at least Albatross had that problem, too. It's that they're a very 'in-the-face' method to take the player down and kick him until he stays down.



Anyway, having taxes cap at, say, 50% for Whorelord (or maybe Hustler) and commensurate reductions elsewhere, together with capping randomness under 100%, would already do wonders for differentiating difficulty levels.