While working on the Red Dragon series I have discovered that the hit points tend to run somewhat lower than in Pathfinder, the base system. This is in large part because there are no longer Constitution bonuses (which add up fast).
This might work if damage generally runs much lower, but I am not convinced this will be so. I think I would like to see a larger difference between “the toughest” and “the squishiest”.
Time to adjust the hit point formula again… especially since the last time I did this, I still had ability scores, and Constitution was part of the formula.
Hit Point Formula Change
If I change the formula from
(level + Martial Training Bonus + Great Fortitude modifier) * tier
(level + Base Attack Bonus + Great Fortitude modifier) * tier
where ‘Base Attack bonus’ is equal to (Level Bonus + Martial Training Bonus), and thus ranges from [floor(level/2)..(floor(level/2)+ceiling(level/2))] the highest hit point totals are increased by about (actually, a little less than) a third. This brings them close to the mean hit points rolled under Pathfinder rules, for the spot checks I made.
I think I’d like the range to go higher than that. Changing the formula instead to
(level + Martial Training Bonus) * (tier + Great Fortitude tier)
will provide high hit point values that are more or less double the original formula. This is a little higher than the mean hit points rolled under Pathfinder rules, but still quite a bit lower than the maximum possible — much of which, under Pathfinder rules, tends to come from Constitution score.
This might make Great Fortitude more popular than it likely is already, but I don’t know that it moves it into ‘must-have’ territory. I expect that many character and creature builds for ‘tough things’ are likely to have Great Fortitude, but that seems appropriate: “Great Fortitude is how you build a tough creature”.
Magic Point Formula
Assuming I keep a similar formula for magic points might be more problematic. Doubling the number of magic points available can have a profound effect on how many spells can be cast between rests. One way to mitigate ‘nova behavior’ is to strongly limit how many spells can be cast per unit time but increase recovery rates so the unit time can be short. That is, rather than blowing the entire day’s budget of magic in one fight, limit how much can be used ‘per fight’ but allow a recovery high enough that that much, or close to it, is available each fight.
Right now a low-level dedicated caster (Caster Training Bonus maximized) can afford about (tier * 1.5) spells per day cast at maximum caster level. Each spell costs a number of magic points equal to the caster level used for the spell, regardless of spell level, with a minimum cost equal to the level needed to cast the spell (five points for magic missile, nine points for fireball — (D&D level + 2)*2-1… I want to make that tidier, but it’ll keep).
This change to the formula, using Iron Will in place of Great Fortitude and Caster Training Bonus in place of Martial Training Bonus, would allow a maximum number of spells per unit time equal to the tier times three. In Pathfinder rules a PC can usually expect to get 5-7 spells of each level below the top (wizards get up to four spells per spell level, sorcerers get up to six, and it’s safe to assume they will have one bonus slot per spell level from high ability score), so ultimately this is likely a reduction in the number of spells per day… but they are likely to come from the top end. On the other hand, removing the caster level cap on spells and the spell level from the save DC may encourage the use of lower-level spells because they can be cast at lower caster level — expanding the capacity per day.
Ah well, see how it goes.
Hit Point/Magic Point Table
In the table below, Level and Tier have their normal meaning. The remaining columns are Training Bonuses, which have a maximum value of one-half the Level, rounded up. Empty cells indicate that the entry is not valid.
For hit points, the Martial Training Bonus is used. For magic points, the Caster Training Bonus is used.
Great Fortitude and Iron Will modifiers can be added to the tier multiplier for hit point calculations and magic point calculations respectively. This means that all values shown in the table above can be increased to up to double the number shown. That is, if an eighteenth-level creature (Champion tier, tier 5) has Champion Great Fortitude and the maximum Martial Training Bonus for eighteenth level (+9), that creature will have (18+9)*(5+5) = 270 hit points.
To compare, a CR 14 red dragon (conveniently, Pathfinder Adult Red Dragons are CR 14) has 17d12 Hit Dice and Constitution 23, for total Hit Dice of 17d12+102. This has a minimum value of 119 and a maximum value of 306, with a mean of 212.5. Thus, the toughest 18th-level creatures might have more hit points than the mean hit points of a lots-of-hit-points creature of the same level in Pathfinder, but less than the maximum that creature could have. The squishiest 18th-level creature has less than half the hit points the dragon does (90).
It looks like this puts me closer to the range I want to be in. Great Fortitude and Iron Will will likely be popular talent choices for tough characters and casters, but I’m passingly okay with that — it’s good to have a clear choice available for certain character types.
For now, I’ll use these values going forward with the Dragons in Echelon series, and will update the already-posted dragons to use these new values soon.
Some months ago I did a spreadsheet comparing D&D hit points to “same-level” echelon hit points and developed the impressions that, while not terribly far off, echelon fighter-types were a little squishier than their D&D counterparts. So I’m glad to see this change.
I was slightly confused by the comparison with PF dragons; presumably a Dragon Warrior at a given tier will have the same HP to a maxed-out human fighter, with no randomness, since they will use the same formula. If somebody wants dragons to be the toughest monsters somebody will face, maybe they need to build them at a higher tier.
Mostly it came up because right now I’m gauging most things off Pathfinder. I’m trying to model Pathfinder dragons, so it makes sense to me to compare my output with Pathfinder dragons.
As it happens, these come in more powerful in some ways (spell casting especially) but relatively quite squishy. Glass cannons are of dubious value because it becomes a game of rocket tag — first person to make contact wins.
You’re right that human fighters should be capable to be as tough as the toughest dragon of the same level (which they are, as far as hit points are concerned), and that if a dragon is supposed to be markedly tougher it should just be a higher-level or higher-tier creature (or that fighters are supposed to not eat their Wheaties and be big and strong as dragons are capable of — no same-tier Great Fortitude for you!).
I like the “just use higher-tier dragons” answer myself.
The mathematician in me doesn’t like the idea of “combine a selection of various numbers in different ways until you get approximately the intended answer”, but I guess it’s only a game, I should really just relax!
Since level/tier are supposed to actually be a useful measure of “power” in some fashion, it should be intuitively obvious that a level N red dragon warrior is about “as powerful” as a level N human fighter, and ditto for the level N dwarvish wizard. So by extension a group of four level N characters would be best challenged by a level (N+k) dragon. I guesstimate k at 4.
I’m an empiricist at heart, so I’ll just frob the formula until it looks about right to me. This should be pretty close, and it’s fairly simple.
I have been trying to keep the dragons as legal (within the limits of my knowledge) characters. I am mildly concerned about talent balance — the dragons might be hitting a sweet spot and getting all the good stuff, and anything else is weaker, but I doubt it… which raises some question in my mind about how badass the high-level characters are going to be.
That’s a good thing, I think.
I agree, k = 4 is the intuitive answer. That’s how CR math works in D&D and Pathfinder as I recall. It may be fairly close here, but will require experimentation and evaluation.