Pathfinder 1E Pathfinder: Encounter Design Simplified

[joela]

If folks are interested, I've posted a free preview of Trailblazer in honor of GM's Day.

Those of you who have been following my posts closely, in this forum and others, have likely seen most of the featured content-- but there is at least one new section there (Customizing Monsters) that is brand new.

Link is in the sig.

Well this is interesting....

 

[joela]

First of all, the chart is adjusted to require "Challenging" encounters as the baseline, rather than "Average." (Because that's how Pathfinder will do it moving forward.)

Oh! Was that in the design forums and beta?

 

[Wulf Ratbane]

Oh! Was that in the design forums and beta?

Probably the alpha, actually-- that's when I originally did the work.

I'll have a little more room to stretch outside of a preview, so I can provide an XP chart for various rates of advancement, just like PF.

I believe the "average" encounter was too easy anyway, and so along with all of the other changes in Trailblazer, more difficult encounters will be the norm, rather than the exception. One monster per PC, at least. Gangpiling a single creature of equivalent CR to your level is neither interesting nor challenging.

 

[meomwt]

Thank you for the GM Day Preview: I am now champing at the bit for the full pdf "pre-Pathfinder." By then, I reckon I might have enough in the kitty to splash out for once

I do like the adding of abilities without adding Hit Dice mechanism. Very nice. Lots of math, true, but nice. I may have to try and put it into practice.

Gangpiling a single creature of equivalent CR to your level is neither interesting nor challenging.

Quoted for truth. As a mean DM, I rarely, if ever, offer equivalent level CR encounters to my players. They haven't lynched me yet

 

[Scott DeWar]

in the table top game i am in there are 8 lv 6 characters. that makes the simplified monster genrator a must for th gm. I think he likes it quite a bit. He also uses 'greater then average' encounters i am sure and we are enjoying it as a group.

don't tell wulf this, but no one has died yet in the party, but I don't think there is any thing wrong with the system.

 

[Roman]

Hello Wulf,

I downloaded the Traiblazer preview PDF and it is interesting. The most interesting part for me is the breakdown of levels into various features (BAB, iterative attacks, HD, skill points feats, class features, spellcasting and so on) and the costs associated with each feature per level. Having said that, however, it seems like this is highly arbitrary and I don't really see the logic behind the breakdown (apart from obviously that higher say BAB should have higher costs - but why BAB is weighted at 0.10 to 0.20 of a level, rather than say 0.20 to 0.40 (just an example) is a mystery. BAB costs given, however, at least scale proportionately relative to each other, but that is not the case with HD. d4 is worth 0.20 of a level, whereas a d12 is worth 0.33 of a level, even though d12 provides 2.6 times as many hit points on average than a d4 (so if the costs scaled linearly and the value of a d4 was truly 0.20 of a level [though it is not clear to me why it should be 0.20 rather than say 0.25 or something else entirely], the value of a d12 should be 0.52 of a level) and at maximum 3 times the number of hit points. OK, you could argue there are diminishing returns, but that is actually probably less the case with hit points than say with attack bonuses which you scale linearly.

I guess I am an inherent sceptic when it comes to what seem to be arbitrary numbers unless I see how they were derived. Would it be possible to show how you arrived at at least some of the numbers in the table? I am sure my confidence in the numbers would increase if a good rationale for them were to be provided.

 

[Wulf Ratbane]

I guess I am an inherent sceptic when it comes to what seem to be arbitrary numbers unless I see how they were derived.

As am I, Roman.

In fact you'll note in my text, I make specific reference to the seemingly arbitrary source of some of Craig Cochrane's numbers-- from which my work is derived. (Craig is known as Upper_Krust on these boards.)

Yet there's no doubt in my mind that whatever the origin of UK's numbers, his methodology is remarkably predictive of CR.

You may question whether that means that UK hit the bulls-eye, or whether he's simply proven out that "close enough" is good enough with respect to CR.

I tend to think it's closer to the latter, and I encourage you to look at the big picture.

Would it be possible to show how you arrived at at least some of the numbers in the table? I am sure my confidence in the numbers would increase if a good rationale for them were to be provided.

BAB costs given, however, at least scale proportionately relative to each other, but that is not the case with HD. d4 is worth 0.20 of a level, whereas a d12 is worth 0.33 of a level, even though d12 provides 2.6 times as many hit points on average than a d4 (so if the costs scaled linearly and the value of a d4 was truly 0.20 of a level [though it is not clear to me why it should be 0.20 rather than say 0.25 or something else entirely], the value of a d12 should be 0.52 of a level).

The value shown for Hit Dice also includes a flat-rate constant for ability score increases, and a flat rate for feats: Regardless of the type of HD you use, you get 1 feat per 3 HD and +1 ability score per 4 HD. This is explicitly called out in the text, and it's why it doesn't scale the way you expect. This is your error.

Back out those fixed constants and here are the HD values:

d4 = .083
d6 = .1165
d8 = .15
d10 = .183
d12 = .2165

There is much rounding off in the text, but those figures are directly derived from the average hit points per HD.

Thus they'll match the ratios you expect to see as HD increases.

I could go into greater detail but it's unnecessary to answer your question-- because it won't address your concerns regarding the arbitrariness of the numbers. You can still drill down to the cost per hit point and wonder why it is set where it's set, and let it annoy you no end.

I can't make you comfortable with the arbitrariness. At some point you're going to have to get there on your own.

Don't lose sight of the big picture. Greater precision does not necessarily guarantee greater accuracy-- and the system that UK has deconstructed is reasonably accurate and reasonably predictive.

 

[Roman]

As am I, Roman.

In fact you'll note in my text, I make specific reference to the seemingly arbitrary source of some of Craig Cochrane's numbers-- from which my work is derived. (Craig is known as Upper_Krust on these boards.)

Yet there's no doubt in my mind that whatever the origin of UK's numbers, his methodology is remarkably predictive of CR.

You may question whether that means that UK hit the bulls-eye, or whether he's simply proven out that "close enough" is good enough with respect to CR.

I tend to think it's closer to the latter, and I encourage you to look at the big picture.





The value shown for Hit Dice also includes a flat-rate constant for ability score increases, and a flat rate for feats: Regardless of the type of HD you use, you get 1 feat per 3 HD and +1 ability score per 4 HD. This is explicitly called out in the text, and it's why it doesn't scale the way you expect. This is your error.

Back out those fixed constants and here are the HD values:

d4 = .083
d6 = .1165
d8 = .15
d10 = .183
d12 = .2165

There is much rounding off in the text, but those figures are directly derived from the average hit points per HD.

Thus they'll match the ratios you expect to see as HD increases.

I could go into greater detail but it's unnecessary to answer your question-- because it won't address your concerns regarding the arbitrariness of the numbers. You can still drill down to the cost per hit point and wonder why it is set where it's set, and let it annoy you no end.

I can't make you comfortable with the arbitrariness. At some point you're going to have to get there on your own.

Don't lose sight of the big picture. Greater precision does not necessarily guarantee greater accuracy-- and the system that UK has deconstructed is reasonably accurate and reasonably predictive.

Thanks Wulf! Yes, for some reason I missed the fact that the 1/3 feat progression and the 1/4 ability score progression were bundled with the hit dice - I guess it was because everything else was separated, so I carelessly assumed these would be too. In any case, that does explain the linearity of progression issue.

As to the arbitrariness issue - fair enough. I guess predictive power is empirical evidence in favor of the system. Still, I would be more comfortable if the underlying mathematics of the numbers were clear. I think it would actually be possible to do determine the relative values of the combat-oriented statistics such as BAB and Hit Dice (or at least their hit point component), which would still leave things like feats with arbitrarily assigned numbers, but would nonetheless help a great deal. Unfortunately, it would involve lengthy calculations, but I am too busy with other things (not gaming-related), so the payoff for me of trying to do those calculations is likely not worth the effort.

 

[Wulf Ratbane]

As to the arbitrariness issue - fair enough. I guess predictive power is empirical evidence in favor of the system.

Certainly so.

Still, I would be more comfortable if the underlying mathematics of the numbers were clear. I think it would actually be possible to do determine the relative values of the combat-oriented statistics such as BAB and Hit Dice (or at least their hit point component), which would still leave things like feats with arbitrarily assigned numbers, but would nonetheless help a great deal.

You can walk back through UK's process yourself if you like.

Start by deconstructing the fighter, since he has nothing going for him but that numerical d20 spine, and his feats. You can use this to "zero in" on the value of a feat (0.2 per feat)-- just keep in mind that not all feats are equal. Close enough is as close as you're going to get (but as close as you need to be).

Moving forward, terms like "feat-equivalent" have to be applied to class abilities and even spells, and here things start to feel a lot more arbitrary-- but still close enough to be predictive.

You don't need CR to predict the outcome of every thrust and parry. (It couldn't do it if you tried-- the game is far too varied and textured.) But it should provide at least a reasonable estimate of overall encounter difficulty and certainly avoid TPKs-- and for this it serves.

 

[Kid Charlemagne]

Can this be extrapolated to the effects of items and the Big Six items? Reason I ask is for my next game I'm considering giving out a feat per level (as per the BOXM 1 & 2) and using action points (per UA), and I'm wondering if this would allow me to reduce the amount of magic drastically without overly impacting the rest of the system...