Warning: I have discovered my love of creating dice result charts, please correct me if I did my math bad. WALL OF TEXT INCOMING. (edit note: already noticed I forgot to clarify nat 1 and crit failure possibilities and how this compares to other dcs/rolls npcs can do versus pcs. Also do note: high, moderate and low bonuses are most common ones, extreme ones are rarely used due to how imbalanced there are, so I'm mostly arguing on principle here.
I will also add that while I think extreme saves should be able to fail at level 22 and 23, adding +3 to dcs might make crit fails too common for regular bonuses, so maybe there should be alternate solution to level 22 and 23 extreme saves like them just being lower than they currently are?)
Definitely agree that there should be caster bonus to spell attack rolls because spell attacks are rarely that much more powerful than single strikes(the ones that are also have save), but not sure about DC. Like dc increase would help make casters more useful vs boss enemies, but it would make them comparatively much more dangerous vs bosses that aren't already ridiculous.(and with those bosses they can focus on buffs and area control spells)
Well to find out, I decided to compare to maxed out caster vs trivial, low, moderate, severe and extreme solo enemies (so level 20 wizard vs level 20-24 enemies), this is how math works out:
Fails on roll vs DC 45 | Extreme | High | Moderate | Low | Terrible |
20 | 5 | 8 | 11 | 14 | 17 |
21 | 3 | 6 | 9 | 12 | 16 |
22 | 1 (can only crit fail, no regular failure) | 5 | 8 | 11 | 14 |
23 | only 1 (can't crit fail) | 4 | 7 | 10 | 13 |
24 | only 1 (can't crit fail) | 2 | 6 | 8 | 12 |
My main observation besides obvious "casters hate their lives vs boss enemies if they are dc focused" is that save growth chart behaves like NPC have profiency level above legendary. But yeah, if caster had +3 rune to DCs, suddenly even level 20 enemy with high dc fails on 11 on dice. That isn't however bad thing necessarily. There is also another benefit, since Class DCs/spell DCs seem to be only PC ability that can only be succeed with level 22-23 extreme bonuses. (level 24 extreme bonuses in general require nat 1 on their roll versus pc. Extreme strike of 46 vs max heavy ac of 47, extreme skill of 48 versus max save dc of 48. Like yeah, level 23 extreme skill +46 has same effect of succeeding on roll of 2, but at least crit fail chance exists.) So if +3 item bonuses to exist, 23 extreme would fail on 3 and 24 on nat 1(so also crit failing) which "feels" more in line with everything else to me.
I think difficult thing here is that I think its partly intentional math wise that targeting boss creatures with Spell DCs is supposed to be hard and that majority of casters have no reason to start with 14 in key stat, so all of them are likely to have either +5, +6 or +7 in casting stat depending how experimental they want to be. So its kinda design choice of "do you want to base DCs around stat of +5 or +6 and give extra boost to +7, or do you want to base them around maximum possible due to how powerful spells are?". I personally honestly like idea that chances should be based around +6 at max since giving apex item to -1 or +0 stat to boost it to+4 should be a valid option as well.
Besides that, I can also understand desire for "I maximized this stat, so I want to do good in it even versus powerful foes" especially when you compare difference between PC saves vs enemy DCs. Like basically PC saves work decently versus their own level npcs regardless of proficiency(they range at level 20 from expert tolegendary) as long stat itself is good. So with resistance runes minimum possible expert profiency save is +26(-1 in stat) and maximum possible legendary save is +38(+7 in stat)
so to compare difference between PC saves/DCs to NPC saves/DCs, level 20 extreme/high/mod dc are 47, 42 and 39. So max save character succeeds on those on 9, 4, and nat 1. For extreme level 24 DCs(52, 48, 45), they succeed on 14, 10 and 7.
So basically pc investing on their optimizing their save is eventually rewarded so that even versus final boss they have decent chance to succeed (compared to lower level extreme foes) while casters are rewarded only when facing equal or lower level foes.
...Okay I wasn't planning to touch on spell attack rolls(I thought caster legendary +35 spell attack vs master martial +36 and legendary martial +38 was pretty self evident especially due to how late casters gain legendary proficiency), but I figured out I might as well check if math favors spell attacks to saves at level 20. My own theory is that math favors spells that don't require saves or attacks for high level bosses, but let's see.
+35 vs AC hits on result of | Extreme | High | Moderate | Low |
20 | 13 | 10 | 9 | 7 |
21 | 14 | 11 | 10 | 8 |
22 | 16 | 13 | 12 | 10 |
23 | 17 | 14 | 13 | 11 |
24 | 19 | 16 | 15 | 13 |
Hmm I think its fair to say that casters are really harmed by level 22-24 npcs having sort of "legendary+ proficiency" jump. Like with bosses you are supposed to rely on debuffs and flanking for flat footed. Like thing to remember about boss ac is that it should always incentive PCs to flank if boss isn't immune to flat footed, but ranged characters trade easy flat footed for distance. Still though, extreme final boss with high in stat succeeds on DC on 3 and caster hits them on 16 on spell attack roll, so weirdly enough, they do have better chance of doing something to them on spell attack roll than save. Still though I think none of this makes lack of +3 item bonus to attack rolls fairer.