Sometimes I Surprise Myself

I haven’t posted anything in a few days. I’ve been doing an analysis of the monsters in the RSRD to see how some of the design changes from Echelon work out.

I can’t exactly model the RSRD monsters. This was never a design goal of Echelon. However, I will take it as a good sign if I can get reasonably close, most of the time. If I can do this at the same time I normalize other design elements (such as making level, Challenge Rating, and Hit Dice mean more or less the same thing) then I am a happy man.

I’m not happy yet… but I can see satisfaction looming.

Analysis

My analysis is fairly straightforward.

I built a spreadsheet of almost all the monsters from the Revised System Resource Document. I skipped the templates (no way to quantify them without applying them, though a few had examples in other entries) and skipped the lycanthropes entirely (template, and I suspect I will implement lycanthropes in a different manner when I get to them). Each line contains for a single monster the Challenge Rating, Hit Dice total, Base Attack Bonus, ability scores, hit point totals, and armor classes (including breakdown by type and by bonuses). I corrected a few errors from the copy of the RSRD I was using (less than a dozen, I think).

I then started adding columns for the derived values, using the Echelon rules I’d previous posted. The results were somewhat surprising.

Armor Class

Hadsil and I were talking the other day about how the Natural Armor Bonus for many or most creatures seems to be more or less arbitrary and used as a correction mechanism to get a creature’s armor class into a level-appropriate range. The Balor, for example, has a +19 natural armor bonus, and a Colossal Monstrous Scorpion has a +25 natural armor bonus.

I find with the Echelon rules the calculated armor class (using each creature’s Base Attack Bonus rather than Challenge Rating or Hit Dice) for almost all creatures is within five points of the armor class presented in the RSRD, not counting natural armor. Of the 522 creatures in my spreadsheet, 29 have an Echelon armor class more than five points higher than the RSRD armor class, and 20 have an Echelon armor class more than five points lower than the RSRD armor class.

Well. I hadn’t looked at it using percentages, but it appears I’ve got over 70% within 2 points of the RSRD, and over 90% within 5 points of the RSRD.

I though it looked good, but that it was generally this close rather surprised me.

Let’s look at some examples of where this doesn’t quite fit.

Echelon Higher than RSRD

It looks like most of these have Hit Dice much higher than their Challenge Rating. The following aren’t too far apart:

• Harpy Archer (14 vs. 11)
• Frost Giant Jarl (22 vs. 17)
• Hound Archon Hero (17 vs. 16)
• Celestial Charger (15 vs. 13)
• Ghaele, alternative form (10 vs. 13)
• Noble Salamander (15 vs. 10)
• Black Pudding (10 vs. 7)
• Harpy (7 vs. 4)
• Formian Queen (20 vs. 17)

Many of the above are advanced monsters (templates and/or character classes) and may therefore be suspect. Some of the others are just odd (the Ghaele alternate form is incorporeal, for example, and it wouldn’t be entirely unreasonable to give the black pudding a natural armor penalty because it is so malleable).

At this point I am inclined to think that this is something to keep an eye on while adapting the monsters for Echelon, rather than a real problem.

Echelon Lower than RSRD

Unlike where Echelon gave a higher armor class than the RSRD, these usually have a much closer relationship between Hit Dice and Challenge Rating. Most of them, though, really should have some kind of natural armor (I see creatures here made of metal or stone or earth, or that have chitin or some other form of hard outer covering).

As above, I think these are worth looking at closely to see if they can be improved, but not otherwise a problem yet.

Hit Points

I haven’t examined hit point totals nearly as closely.  I calculated hit points for all living creatures (not constructs or undead), again using Base Attack Bonus rather than Challenge Rating or Hit Dice.

Defining criteria for ‘far out of line’ is a little more difficult. I started with simple percentage difference, but this gets thrown off at low levels because the hit points are deliberately much higher (a starting elf warrior can expect to have four times the hit points he otherwise would (Con 10 + 1*(6+mCon), compared to 4+mCon), for example). I find that all the living creatures have at least as many hit points as they would average in RSRD RAW (will-o’-wisp is the closest, with 43 instead of 40). I didn’t include the unloving creatures (constructs and undead) in this analysis because they don’t have Constitution scores.

In Echelon they will, at which point they will have hit points calculated just like all the other creatures. If I am trying to model the RSRD constructs and undead they are likely to have the Toughness talent, or they may have the bonus hit points as part of the Construct or Undead packages.

Ability Scores

As part of my analysis I tried normalizing the monsters. I wanted to see what the creatures would be like as ‘basic creatures’ – partly so I could see what adjustments and modifiers would be needed to model them in Echelon, partly out of curiosity.

I reduced all ability scores by one-quarter of the Challenge Rating (stripping off the tier capstone ability score bonuses) and backed out the size modifiers.

I’ll have to post the findings at another time, but one thing I noticed right away, especially for some of the problematic creatures, is that they don’t seem to follow the guidelines for size changes. In many cases (usually the really big creatures) they shouldn’t be able to move under their own power because at Medium size they end up with negative Strength scores.

However, for the most part creatures tended to have normalized ability scores in the 3-18 range. Where they tend to be somewhat higher the creatures are usually fairly high-CR. The top normalized ability score totals are held by:

In many cases the totals are possible, if unlikely. The pit fiend is a CR 20 creature and has normalized ability scores no higher than 24. This is achievable by a 19^th-level human in Echelon (rolled using 27-25-23) that spends his top-tier slots on improving his ability scores (24, 24, 20, 21, 21, 21: 14+10, 14+10, 12+8, 13+10, 11+10, 11+10). The titan, on the other hand, isn’t quite possible without another bonus somewhere (normalized Constitution of 30 – starting with 18 and with Legendary Great Constitution you can get to 28, but a +2 racial modifier could take care of it; the tier capstone bonuses were removed in normalizing and thus can’t be counted here).

However, all in all they are in valid ranges for PCs, though their aggregate totals can run a little high. The most troublesome ones tend to have high scores across the board. I suspect than in many cases this is neither necessary nor appropriate. As I’ve said in other monster design conversations (and character design, for that matter) I find consistently high ability scores to be boring. I’d rather see monsters and characters a little more specialized.

Conclusion

There are some creatures that are somewhat problematic to model, but they look as if they often don’t really follow normal monster design guidelines. They might have very mismatched capabilities (mostly bizarrely high or low armor class or ability scores for their level), be too broadly capable (really high ability scores across the board), or be otherwise unbalanced with themselves.

I didn’t go into it much above, but the cases where the Echelon armor class closely matched that of the RSRD armor classes often had natural armor bonuses and Hit Dice higher than Challenge Rating. These cases usually had hit points rather higher in Echelon than in the RSRD. These things combined suggest that I have a lot of room to reduce the Hit Dice to bring them more in line with the Challenge Rating (which is part of what part of what I meant by ‘satisfaction looming’).

I think Echelon is headed on the right track.