Rules Hack: Fixing XP Costs

The main hub for homebrew, mechanics, splats and rewrites

Moderator: Epiphany

Post Reply
Posts: 57
Joined: Tue Apr 27, 2021 10:01 pm
Location: Canada
Has thanked: 271 times
Been thanked: 26 times

I came up with this back when I first read Exalted 2E, but didn't go very far with it.  When I saw that 3E had the same problem, I sat down to actually do the math, which turned out pretty basic.  

What? (or, TL;DR) 
• Attributes cost 12 XP/dot, Abilities cost 6 XP/dot.  
• If you have Cast, Aspect, or Favored Abilities (as Solars/DB), they cost 5 XP/dot.  
• If you have Caste or Favored Attributes (as Lunars), they cost 9 XP/dot.  
• If you're a Casteless Lunar, your non-Favored Attributes cost 11 XP/dot.  
• If anyone tries to convince you to play with scaling XP costs, they are a Fair Folk trying to eat your sanity.  

Why? (or, The Chargen-Minmaxing Mandate)  
• Exalted has a problem that I call the "Chargen-Minmaxing Mandate," which as far as I can tell has dogged every single White Wolf product in the studio's history.  The average Exalted fan is probably very familiar with this, but I will summarize for the benefit of anyone who is not.  
• The core of the issue is that there's a disconnect between what certain statistics cost at character creation versus what they cost in-play.  
• Specifically, Attributes and Abilities have a fixed cost at character creation (when bought with starting dots and bonus points), but a scaling cost in play (when bought with XP), such that higher ratings cost more XP than lower ratings.  
        • From the Core book: At character creation, Attributes cost 4 BP per dot (3 BP for a Tertiary attribute), and Abilities cost 2 BP per dot (or 1 BP if Caste/Favored).  Raising an Attribute or Ability in play, meanwhile, costs [current Attribute rating x4] XP or [Ability rating x2] XP.  

        This encourages - indeed, mandates - min-maxing at character creation.  No, the book doesn't come out and say "hey, make sure to give your character a wildly unrealistic stat spread," but the rules it contains mean that a player who min-maxes their character will have a real and significant advantage over one who does not.  

Math here:  

Attribute XP costs: 
• Minimum rating of 1 
• RAW price: [rating]x4 XP - thus, the second through fifth dots cost 4, 8, 12, 16 XP respectively 
• 4 to 5 dots (1 dot): 4x4 = 16 XP total; 16 XP/dot 
• 3 to 5 dots (2 dots): (3+4) x4 = 7x4 = 28 XP total; 14 XP/dot
• 2 to 5 dots (3 dots): (2+3+4) x4 = 9x4 = 36 XP total; 12 XP/dot  
• 1 to 5 dots (4 dots): (1+2+3+4) x4 = 10x4 = 40XP total; 10 XP/dot

Ability XP costs: 
• Minimum rating of 0 
• RAW first dot: 3 XP 
• RAW price: [rating] x2 XP - thus, the second through fifth dots cost 2, 4, 6, 8 XP respectively 
• 4 to 5 (1 dot): 8 XP; 8 XP/dot
• 3 to 5 (2 dots): 14 XP; 7 XP/dot
• 2 to 5 (3 dots): 18 XP; 6 XP/dot
• 1 to 5 (4 dots): 20 XP; 5 XP/dot
• 0 to 5 (5 dots): 23 XP; 4.6 XP/dot

        Example: We will consider the Attribute spread of three hypothetical characters A, B, and C, and the XP required for them to raise all Attributes to the maximum rating of 5 dots.  Character A is as min-maxed as possible, Character B is as un-min-maxed as possible, and Character C is merely as un-min-maxed as it is possible for a starting Solar to be under RAW.  For simplicity, we will assume that no Bonus Points are spent to improve the starting Attribute spread, allowing us to mostly ignore the distinction between categories.  
• Solar characters start with one starting dot in each attribute, plus 8/6/4 dots to spend on their primary, secondary, and tertiary attributes respectively, for a total of 9 + 8 + 6 + 4 = 27 dots.  
• A maxed-out spread of 5 dots in every Attribute is a total of 45 dots.  Thus, each character will have to buy 45 - 27 = 18 dots.  Put another way, this exercise will show us the different costs of buying 18 dots, and provide us some guidelines for determining fixed costs.  

Starting Attribute spreads
• Character A: 5/5/5/5/3/1/1/1/1  
        • Note that their starting points are concentrated in as few Attributes as possible, and as a result all their Attribute Scores are either as high or as low as possible (with one 3 hanging out in the middle).  They will spend as much XP as possible on the cheaper low-rating increases, with 4 Attributes already capped.  
• Character B: 3/3/3/3/3/3/3/3/3  
        • The worst possible scenario - all Attributes sitting at 3, blocking off the cheaper low-ratings and leaving only the expensive 3- and 4-dot increases.  
        • Note, however, that it is not actually possible to create a character with this attribute spread, due to the division of dots between categories.  For that, we must turn to:  
• Character C: 4/4/3/3/3/3/3/2/2 
        • Each category's dots are split as evenly as possible, leaving two primary Attributes at 4 dots and two tertiary Attributes at 2 dots.  This is as un-optimized as a starting Solar Exalted can be, at least in terms of attributes.  
        • C is better off than B, but not by much - they still have no attributes capped, and thus will have to buy all 9 top-level increases.  

XP costs to 45 dots:  
• Referring to the table above 
• Character A: 4x0 + 1x28 + 4x40 = 28 + 160 = 188 XP total
        • 188 / 18 = 10.444... XP/dot 
        • With 4 Attributes already capped, they cost 0 XP to raise.  
• Character B: 9x28 = 252 XP total;
        • 252 / 18 = 14 XP/dot 
• Character C: 2x16 + 5x28 + 2x36 = 32 + 140 + 72 = 244 XP total 
        • 244 / 18 = 13.555... XP/dot 

• Character A has a massive advantage (56 XP) over Character C, who is only slightly better off (8 XP) than the hypothetical Character B.  
• 56 XP suffices to buy between 5 and 7 Charms, Spells, or Evocations - a half to a third of a starting Solar's allotment of 15.  It is also 6 XP over the threshold for Essence 2.  Indeed, the rules for experienced Solars have them start with 5 more Charms (plus 3 more Merits and 3 more Bonus Points).  And this is for Attributes alone - imagine if Character A had the additional foresight to ensure that all the abilities they ever intended to were either Caste, Favored, or capped by the end of character creation.  

If you skipped that, don't worry.  I'll tell you the end result anyway: 
• Under the RAW of scaling XP costs, the difference between an optimized versus un-optimized character is more significant than the difference between starting at Essence 1 versus Essence 2.  

        "Discouraging" min-maxing does nothing.  In the best-case scenario, it just moves the min-maxing back a step, such that someone who starts with a character concept that lends itself to min-maxing (such as a swift and agile but otherwise physically weak or fragile assassin, who can believably have DEX 5 and STA or STR 1) will have the same kind of advantage over someone who starts with a character concept that doesn't (such as a career soldier, who should realistically have all 3 physical attributes at 3 or 4).  In the worst-case scenario (i.e. forcing your players to make equally unoptimized characters), you effectively prohibit such character concepts, with the added risk of alienating your players - for example, if an autistic player (like me) wanted to play someone like themselves (like the first character I made), they might very well make a character with nearly all Abilities at either 5 or 0, to represent a savant with incredible prowess in certain areas but staggering incompetence in others (me.  I'm talking about me.  That's what life is actually like for me), and I think we can all agree that barring them from that would be wrong.  

        I think that we can also all agree that limiting character concepts - even indirectly and non-explicitly, as scaling XP costs do - is bad for the game.  Exalted is fun because of all the weird & inventive stuff you can do and be.  Limiting that makes it less fun.  

        So, what to do?  

The Fix: Static XP costs 

        The core problem, as I see it, is scaling XP costs.  The solution, therefore, is to replace those with static costs, so that every dot in a given Attribute or Ability costs the same as every other dot in that Attribute or Ability.  Character A's massive head start in my earlier example vanishes - everyone starts with the same number of dots, everyone pays the same XP cost, and nobody gets to start at Essence 2 for free.  

        So, what should those costs be?  

More math!  

Our findings from the previous example: 
• Attribute dots cost, on average, between 10 and 16 XP depending on where you start.  
• Ability dots cost, on average, between 4.6 (or 5, for a better comparison) and 8 XP.  
• In aggregate, Attribute dots cost between 10.444... and 13.555... (or 14) XP, depending on how optimized the character is.  

From RAW: 
• Attribute dots cost 4 BP, and [rating]x4 XP.  
• Ability dots cost 2 BP, and [rating]x2 XP.  
• Caste, Aspect, and Favored Abilities cost 1 BP, and ([rating]x2 - 1) XP.  
• Caste and Favored Attributes cost 3 BP, and [rating]x3 XP 

From this, we can derive certain patterns: 
• Attribute dots cost (in general) twice as much as Ability dots.  
• Certain traits (Caste/Aspect/Favored Abilities, non-Favored Attributes for Casteless Lunars) cost 1 point less than normal.  

        If we can pick a fixed reference point to start with, we should be able to derive the other costs with relative ease.  The point that stands out to me is the average cost from 2 to 5 dots - 12 XP per Attribute dot, 6 XP per Ability dot.  

Arguments in favor of the "2 to 5" standard: 
• The Attribute cost of 12 XP/dot sits nicely in the middle of the aggregate average range - the average of (188/18) and (244/18) works out to exactly 12.  
• The "effective rating" (that is, the rating at which the RAW cost and the static cost match) works out to 3, in the middle of the standard 1-5 scale.  
• As a result of the above, it fixes the BP-XP "exchange rate" at 3 XP to 1 BP - the same as for other (non-Charm) traits like Merits and Specialties.  This makes sense, because Specialties (and several Merits) function basically the same way as Attributes and Abilities - incrementally adding dice directly to the relevant pools.  
• It's low enough that most people will see their XP costs come down (18 Attribute dots cost 12 x 18 = 216 XP), and only the most egregious min-maxers will end up paying more.  

        You are, of course, free to pick a different reference point.  You'll just have to live with being mathematically wrong.  


        The result: 12 XP per Attribute dot, 6 XP per Ability dot.  Caste, Aspect, or Favored Abilities cost 5 XP per dot, in keeping with the standard one-point discount.  The core book is safely de-scaled and safe from min-maxing.  

        But that's just for Solars, Dragon-Blooded, and mortals.  What about Lunar Attributes?  

Moon math ahead: 

Lunar Attributes:
• Caste/Favored:
        • RAW cost: [Rating x3] XP
        • Individual dot cost (2nd through 5th): 3, 6, 9, 12

        • 4 to 5: 12 XP: 12 XP/dot
        • 3 to 5: 9 + 12 = 21 XP: 10.5 XP/dot
        • 2 to 5: (2 + 3 + 4) x3 = 9x3 = 27 XP: 9 XP/dot  
        • 1 to 5: (1 + 2 + 3 + 4) x3 = 10x3 = 30XP: 7.5 XP/dot
• Non C/F: These cost the same as for Solars, but we'll reproduce them anyway 
        • [Rating x4] XP - 4, 8, 12, 16 per dot 
        • These cost the same as for Solars, but we'll review them anyway 

        • 4 to 5: 16 XP/dot  
        • 3 to 5: 14 XP/dot
        • 2 to 5: 12 XP/dot  
        • 1 to 5: 10 XP/dot
• Non-Favored (Casteless only):
        • RAW cost: ([Rating x4] - 1) XP
        • Per dot: 3, 7, 11, 15

        • 4 to 5: 15 XP: 15 XP/dot  
        • 3 to 5: 11 + 15 = 26 XP: 13 XP/dot
        • 2 to 5: 7 + 11 + 15 = 33 XP: 11 XP/dot  
        • 1 to 5: 3 + 7 + 11 + 15 = 36 XP: 9 XP/dot


We can see that the "2 to 5" standard still applies nicely: 
• Non-Caste/Favored Attributes cost 12 XP/dot, the same as for everyone else 
• Caste or Favored Attributes cost 9 XP/dot - 3/4 of the normal cost 
• Casteless get a one-point discount on non-C/F Attributes, bringing their cost down to 11 XP/dot.  

        We have now reduced all those messy scaling XP costs to nice, clean static costs, which concludes the Rules Hack itself, at least until the next splat comes out and mucks everything up again in new and exciting ways.  I will finish with a brief overview of other XP-related issues not covered by this hack, and in particular what this hack does not do: 
• This rules hack does not make all chargen strategies equally viable.  The book points out that certain things are more cost-effective to buy with BP than others, and I am inclined to take that as the game intentionally favoring some strategies over others, for the purposes of consistency and thematics.  Using this hack, your top priorities for Bonus Points should be Caste/Aspect and Favored Abilities, followed by Willpower (and tertiary Attributes, for non-Lunars), followed by general Attributes & Abilities, Merits, and Specialties, followed by Evocations, Spells, & Charms - in other words, exactly what the books say your priorities should be.  
• If anything, this hack has made the books' advice more viable - you can now have priorities other than MAX OUT ALL RELEVANT ATTRIBUTES/ABILITIES IMMEDIATELY.  
• Yes, this means every Lunar player with even the slightest min-maxing tendencies will probably start with Willpower 10.  Yes, that is also system-mandated character-creation min-maxing.  No, I don't care - the worst-case imbalance is a piddling 20 XP, and it's a stat that contributes to exactly 0 rolls.  Weighing the balance between the long-term benefits of Willpower 10 versus starting play with (for example) a Warstrider and a hidden fortress-Manse is left as an exercise for the reader.  If you feel that starting with a Warstrider is not sufficiently viable, you are free to make your own rules hack.  I'm not your mom.  
• As a corollary to the above: some min-maxing just doesn't matter.  On their own, scaling XP costs are just an annoying bug.  The fact that the scaling XP costs were attached to high-priority traits that directly affected how well you performed in-game - that was what made it a game-breaking problem.  If scaling XP costs were tied to Willpower, everyone would probably just go around with Willpower 5, and spend their BP / XP on more important traits.  
• Similarly, I think that the book's stated priorities make sense.  Your character had to attract their Exaltation somehow, so it makes sense for them to have a solid foundation of mundane-tier prowess before you start adding more superpowers.  


If you read all the way down here, I hope you found all that useful.  Please let me know what you think of it.  
send feedback pls

User avatar
Posts: 148
Joined: Mon Apr 26, 2021 6:28 pm
Has thanked: 100 times
Been thanked: 81 times

There's an argument that you should have lower costs, because some people will never raise some attributes up to 5
That is, if you have an attribute at 2, and you raise it to 5, you spend 8+12+16 = 36xp raising 3 points, for an average of 12, as you said.
Raising from 3 to 5 averages 14, raising from 1 to 5 averages 10

However, raising an attribute at 1/2/3 to 4 costs an average of 8/10/12

Putting the static cost of raising at attribute at 12xp makes the marginal costs equal at 3, but the total costs is equal at more like 4 or 5

Personally, I'm in favor of doing the reverse - keeping xp as scaling, and making character creation based on xp. The primary reason I like doing this is the same as why all-static is a good idea - it prevents weird incentives where things are cheaper to buy at different times. I like that scaling xp costs encourage spreading out in a gentler way - starting with 5s is a notable cost, but is still an option, while static costs can make it a bit too easy to be at 5s and 0s. I worry that at 12xp a pop, raising strength from 1 to 2 will basically never be mechanically a good idea, while raising dex will always be a good at. Scaling xp means that a str 1, dex 4 character can raise strength for 4xp, and dex for 16xp.
I do admit it leads to more complicated character creation, and I've never made a pure-scaling character without using a spreadsheet. But I like spreadsheets, so that's fine.
Post Reply