Contextual Power Rating

PhoneLobster · Post by **PhoneLobster** » Sat Apr 18, 2009 4:59 am

OK. So back before my computer died the sudden death in the night I had been arguing with Frank about unified and separate PC/NPC generation systems.

Frank brought up (an entirely spurious point) that Unified character generation systems are unbalanced in the context of the individual encounter level.

Effectively, with the sailor vs farmer scenario he suggested that by splitting your PC/NPC system you could magically make your NPC generation system automatically adjust for the context of every single encounter and you could just run that in parallel with a PC system that didn't adjust with the contextual changes and for some reason didn't need to.

There was a lot that was wrong about that and that was went into in all sorts of detail. So I don't really want to discuss the flaws of Failor vs Sarmer and the unified or aparthied character generation systems issue in too much detail.

HOWEVER.

It raised an interesting question. I brushed on it here and there in the ensuing argument, but now I want to talk about it in detail.

What if you did create a character design system (unified or not) that was centered NOT around assumptions about "average encounters" and "average opposition by level" or whatever, but instead by the specific context of individual encounters?

That after all was what was required to even begin to meet the demands of the Sailor vs Farmer scenario and if possible it might well have a wide variety of benefits.

So to get a catchy acronym I'm calling it...

Contextual Power Rating (CPR)
The game balance point in a CPR system is the individual and specific encounter. Character abilities are rated in their effectiveness based on the context of opposing character qualities and the environment of the encounter.

To determine the equivalent of challenge rating (which will actually be called "CPR" because it's cool) before an encounter you add up the CPR values of character abilities that apply within the context and compare them.

So all your abilities, like for instance a Dwarf's bonus vs giants, only count as adding to your value in an estimated match up if the opponents actually include giants.

Balancing with CPR
OK, so how do you USE such a balance point?

Well CPR would have to be as strictly disciplined in it's accuracy and detail as you can manage. The ultimate comparison between characters it creates should allow the GM to determine on an individual encounter by encounter basis how challenging encounters are and how effective or at risk individual characters are.

With that information the GM should be able to balance encounters (individually) to whatever preferred difficulty level. For any specific group of characters in any specific context.

He can further use that information to determine which Player characters aren't pulling their full weight in the context of the specific game or campaign and either adapt the context of the campaign and NPCs, or provide means of adapting the PC (like additional items and skills).

Thus allowing players to bring a mighty "Rock Wizard" OR a "Scissors Witch" on the great campaign of "Lair of the Paper Tigers and nothing else ever" and it still all being somewhat within the capabilities of the shifting balance point rather than falling into the gaps in an assumed average encounter RPS balance point.

Major Benefits
OK so here are some of the cool things this sort of system would land you.

1) "Average Encounters" is no longer an assumption required to make the game work. 30% of encounters do NOT any longer need to occur within rabbit warrens or against Paper RPS Tigers.

2) Talking about RPS type systems your RPS type systems no longer need to be built in nice circular/opposed balanced ways. You can totally have "UFO" powers that Trump rock, paper AND scissors, or you can have "Rabbit" powers that apply less often than normal, and CPR just accounts for them only when and if they apply.

3) Similarly "Spotlight Balancing" is now something you can do (or chose not to do) manually and with significantly greater accuracy, instead of being simply assumed and again relying on average encounters.

3) "Wealth = Power" is not a significant problem any more. If you have wealth that equates to (immediately relevant) power then your CPR is higher, if you don't, it isn't.

4) Complex character building mechanics are no longer part of your balance mechanism. CPR measures character power, so you don't specifically need another mechanic like character level or character building points to determine the value of a character. You CAN use such a character building system as well, or you can make gains on the complexity front and just drop it and use CPR for everything.

5) You can use CPR to create some truly "organic" characters. Character abilities can just be given out with things like training montages, item looting, and background fluff. CPR then keeps a running tally of the impact and lets you do more of the same.

6) CPR is a system that not only allows but favors on the fly character generation and even extension. If your NPC Sailors in a Sailing encounter have a CPR of some specific value that CPR tells us nothing, and limits you in no mechanical way from simply adding abilities that only count in Farming encounters should you ever need to expand on the NPCs in that way.

But can this crazy sexy bitch of a mechanism ever actually be implemented?
I ask this question not because I doubt it can, indeed I think I've got most of the basics figured for using it with my own preferred home brew rules system.

No I mostly mention it because RC is going to come in here and be all like "That sounds impossibly massively complex inside my care bear gut, that's how people measure complexity right?"

Well. It potentially IS complex, like any sweeping mechanic, if implemented poorly. But it can do something many such proposals don't by delivering significant complexity benefits by replacing entire OTHER sweeping potentially complex mechanics.

The current stuff I'm looking at for my own implementation is a CPR system that is somewhere equal to or less than the complexity of resolving a single round of combat for the characters involved in order to determine their CPR before the encounter.

I'm going to post the basic mechanical context, my proposed plans for making that into a CPR based system and maybe some other stuff. But I'm not going to throw it in all at once because this is already a long post with plenty I want to discus, and anyway I'm only just now downloading Open Office to get access to my old rules documents and the relevant "Cunning CPR Implementation Plan" document I wrote up a little while ago.

So yeah. Discuss, poo poo the idea, push it in various directions, whatever, once I see some interest, and get my act together I already have more material around here somewhere to add to the bonfire.

Fuchs · Post by **Fuchs** » Sat Apr 18, 2009 9:21 am

I almost always eyeball my encounters based upon the specific party composition, gear and circumstances, adjusting and tweaking when needed, instead of CR, so I'd guess I do this already.

PhoneLobster · Post by **PhoneLobster** » Sat Apr 18, 2009 10:09 am

Fuchs wrote:I almost always eyeball my encounters based upon the specific party composition, gear and circumstances, adjusting and tweaking when needed, instead of CR, so I'd guess I do this already.

It is essentially the same thing. But the idea here is to build a whole system around it, making it a formalized, easier to use and more accurate mechanic than just "eyeball it".

But the fact that you, and I and every fool already spends time making context based judgments of applicable character power on an individual encounter by encounter basis (with no damn assistance from the system being used) essentially just supports the position that there is a need for such a mechanic.

It also suggests that yes, as a means of balancing individual encounters and therefore individual games "average encounter" based balancing paradigms like RPS, Spotlight balance, and even "The Same Game Test" are a failure.

Draco_Argentum · Post by **Draco_Argentum** » Sat Apr 18, 2009 10:10 am

In the interests of moving an idea that sounds good forward:

1) How are CPR totals created? For example what if I have a passive ability that does quadruple damage to undead and an active ability that does some damage and stuns undead. Thats very different from having the damage and stun ability and an other active ability.

2) I think you overstate the benefit re wealth. If I want to beat up some necromancer lord and wealth increases my CPR to the point I can do so I'm still incentivised to horde cash. If any CPR can take the necromancer lord why haven't the local peasants done so?

PhoneLobster · Post by **PhoneLobster** » Sat Apr 18, 2009 10:35 am

Draco_Argentum wrote:1) How are CPR totals created? For example what if I have a passive ability that does quadruple damage to undead and an active ability that does some damage and stuns undead. Thats very different from having the damage and stun ability and an other active ability.

You are asking how do you weight the value of two bonuses when one has an action cost and another doesn't?

I'm looking at two issues on that.

1) Can you give a firm estimate of value to bonuses that aren't a simple numeric bonus?

The answer has to simply be "yes" if you can't then we can't balance ANY system that has things like variable action costs or special abilities that do anything more than "always add X".

2) When determining CPR how do you determine what choices with limited resources (like actions) the players will make?

I'm leaning towards an attackers favor + party like no tomorrow set of assumptions about how choices in combat will be made.

My intended implementation of CPR already plans to assume that attackers will always attack their best possible target with their best possible abilities and if they have any sort of limited use options they will always opt to use them.

Part of this means that calculating CPR isn't JUST totaling all relevant abilities, but is also, potentially, determining which combination of abilities gives the biggest relevant total. However this is effectively the SAME choice / operation players make during every individual turn of combat anyway and any limits on per combat turn options (which would be a sane thing due to limited complexity of choice issues made much of around here) would effectively carry over their benefits in the most part to CPR.

2) I think you overstate the benefit re wealth. If I want to beat up some necromancer lord and wealth increases my CPR to the point I can do so I'm still incentivised to horde cash. If any CPR can take the necromancer lord why haven't the local peasants done so?

If you horde cash and go face the necromancer with a fancy new whatever the best thing you can buy is (and we don't care if it's a kitchen knife or the hand of vecna) then the GM (and the players) know your net CPR compared to the necromancer.

If you spend all your cash on black jack and hookers we still know your net CPR.

The GM can use this as a tool to re-balance the encounter, adding additional mooks, the odd item or something. If for some reason the NPC/encounter is known ahead of time to the players and the GM is unwilling or unable to modify it in any way while the PCs effectively spend game time modifying their own attributes then it becomes a game of "how much prep work can we do to lean this encounter to our favor" for the players.

And presumably if you are allowing that, or encouraging it by having NPC villians waiting around saying "neener neener you need to raise your CPR to take me on" then I can only assume you WANT to do that and that CPR is going to help you in the execution of that goal by tracking your advancement towards it.

In the mean time the ability to pick up some item or rent a swanky hotel room without playing pretty please with wealth by level guidelines is a significant advancement forward compared to most wealth vs power balancing mechanisms.

PhoneLobster · Post by **PhoneLobster** » Sat Apr 18, 2009 11:29 am

OK anyway, as a sort of "proof of concept" on this I'm going to implement it into my own home brew system.

To do that in a way that makes any sense I'm going to need to describe it a little and give some, well, context.

Elements of my basic combat system
1) Big Squares
It uses my favored big squares system. I've described that elsewhere, it is somewhat less directly relevant to this discussion in comparison to other aspects of the system but it's worth mentioning.

The simple nature of big squares and the way in which a particular layout of zones can rather more transparently influence a battle could mean that CPR conditions like "Don't count this in CPR if combat starts in the same zone as opponents" and so on will be both easier to determine and more predictable in outcome.

2) Simultaneous resolution and Damage per turn Caps
The turn system is simultaneous. Declaration and resolution of actions happens in a specific order, but ultimately is the order is not especially important and all effects from individual actions only apply at the change over to the new round.

There is a limit to how many injuries a character can sustain in a single turn and that limit is ONE. The character only suffers the most damaging injury dealt to them in the same turn. All other attacks that would have dealt damage are then described simply to have set the character up to take the fall on the big hit.

This somewhat reduces the impact of multiple attackers and focused fire tactics. Such things still have an influence and the attackers favor assumption of CPR will generally count/predict casualties based vaguely on focused fire tactics.

3) Bonuses and attack rolls
I've abandoned attributes, skills, levels, and classes.

Characters and their abilities are just keywords, bonuses against keywords and keywords that grant special benefits like longer range attacks or attacks against multiple targets.

So you have characters running around with the "Slow" keyword on their attacks and/or defenses and other characters running around with bonuses like "+4 vs Slow" on their attacks and defenses. And that's really most of the system there.

Attacks are an action/combat turn resolved as follows. Attacker rolls 1d20+relevant bonuses and compares to 10+defenders relevant bonuses. If the attack is equal or greater the attack succeeds.

The implications for CPR are beneficial, this is a pretty simple mechanic and the whole keywords thing means you are already dealing with transparent conditional statements telling you when and against whom bonuses apply instead of having a whole pile of less transparent attributes, skills, special abilities AND keywords that need to be worked through to derive that information.

4) Damage as a penalty to defense
A successful attack deals an injury.

Injuries are either Critical (removes a character from combat), Severe (applies -10 to defense) or Light (applies -5 to defense).

The injury dealt from an attack is normally the next worst injury than the target currently has (note that any injury dealt THIS turn isn't yet counted due to the simultaneous resolution thing). So you go, no injury-> light-> serious-> critical.

For a voluntary -5 attack penalty before rolling you can skip one step in the injury progression and deal a serious injury to an uninjured target or a critical to a lightly injured one. If you still succeed with the penalty, otherwise you miss.

For a voluntary -10 attack penalty before rolling you skip straight to a critical from uninjured.

The implications of this for CPR is at first fairly good (predictable character life span over the basic progression with basic attacks).

It is also good because if a character walks into combat with an existing injury it can be measured by CPR as a simple direct defense penalty.

And then maybe not so good, the option to gamble reliability for bigger damage means that CPR may become less accurate with assumptions being made about when or if players will take that gamble and the calculation required to value abilities and build the CPR system may become more complex to account for the existence of that option.

For now the skipping ahead on the injury progression remains but it might need work later.

5) Spare Lives
When a character is injured they may have an ability called a Block.

At the change over to the next Round when an injury from the prior turn would normally be applied a character may expend a Block and negate the injury.

Blocks are interesting in their impact on CPR. Effectively being of variable value depending on what sort of average damage per turn is being thrown around.

So for now the CPR system is going to adjudicate blocks by counting a character's available blocks (if any) and considering them near the end of the process when opponents attacks and defenses are compared to determine some idea of average damage per turn.

Why CPR is so attractive to this system
I lacked a balance and character building/advancement mechanism. It even lacked the most rudimentary encounter challenge rating methodology.

For the last three or more iterations it has been in the too hard basket and left pretty much as "Wing It".

Being as it was a system where nearly all the bonuses were conditional vaguely RPS sorts of things it was pretty damn clear that it was all about power varying on an individual encounter by encounter basis and traditional "average encounter" methods were just plain not going to help.

CPR gives this system a more formalized methodology to eyeball relative power of characters and also sets a practical standard by which powers can be evaluated for character building that isn't based in an "average encounter" assumption that will fail in most campaigns, even the ones I the god damn designer will be running.

And so...
That is enough of a background I think to support a later post explaining how I intend to implement a CPR system on top of those mechanics.

Which I'll do in my next major non-reply type post here.

Username17 · Post by **Username17** » Sat Apr 18, 2009 12:10 pm

Problems:

Character Generation: I'm sticking to my guns on this one. You only have two options; that of building characters in advance of the game and reworking characters Final Fantasy Tactics style at the beginning of encounters. If it is your intent to do the first, CPR is right out the window during actual character creation and we're back to average power levels and same game tests. If it's the second your character generation system has to be incredibly simple because you're asking people to do it several times a night.
Puzzle Monsters: Or really any challenge that includes any amount of mystery whatsoever. If the players do not know which abilities they should be using, then the value of their abilities is right back to being an average power based on your speculation of what abilities they could use. If the players don't know ahead of time that the Baron is in fact a werewolf than the fact that the party white mage has moon ray is something that may or may not radically alter the party's effectiveness in the upcoming situation.

Fundamentally, to make CPR work the way you're claiming it does (where it actually does away with the need to assign average encounter utility assumptions to abilities while people make their characters) you need to have a system with incredibly complete information and a dynamic (and therefore simplistic) character creation system. As in, even Strange Synergies is both too complex and includes too many unknowns for a CPR system to actually deliver on your incredibly inflated promises.

Now I'm not exactly saying I'm against CPR. There are things about it that seem like they would be good. Having a CPR system in place to give decent guidelines about when you were making a challenge that was too easy or too difficult for the team at the table could be invaluable. Especially to novice game masters.

But it's not going to replace making people purchase their powers ahead of time based on a theoretical average future utility of those powers based on a best guess of campaign direction. It can't. And beyond that, it probably shouldn't be used to scale difficulty based rewards, because in doing so you would end up making conditional abilities feel worthless. If making a challenge easy by having the right tool for the job ends up giving you challenge rewards for an easy challenge i makes having the right tools pointless.

Good tactics and having the right tools on hand should be rewarded. So if you actually scale the encounters so that they are always the same difficulty that's actually bad. So using CPR as a proscriptive rather than descriptive tool would actually make the game worse and I don't suggest doing it.

But as a formalized system to keep GMs from accidentally TPKing the players it could be good. Very helpful even. Just not as revolutionary as you seem to hope.

-Username17

PhoneLobster · Post by **PhoneLobster** » Sat Apr 18, 2009 1:05 pm

FrankTrollman wrote: Character Generation: I'm sticking to my guns on this one. You only have two options; that of building characters in advance of the game and reworking characters Final Fantasy Tactics style at the beginning of encounters.

This is an interesting point and one I have already considered.

It boils down to several issues.

1) CPR is something that responds to a rather specific set of requirements and has arisen from a specific argument. You and RC simple were having none of this "here is something I prepared earlier" approach to NPCs, it was on the fly or nothing. CPR leans toward supporting on the fly rather than prepared earlier for much the same reason.

2) CPR may lack support for prepared earlier character creation. But it is no worse off than an "average encounter" balanced system. All your abilities as a basic requirement of CPR have valuations in a unified scale plus information about situations when that valuation may change or not apply.

An average encounter balanced system has a nearly identical set of ability valuations on a shared scale and simply lacks the additional information.

You could potentially take a well balanced traditional points based system and simply include CPR conditional statements beside the points values. That doesn't make the system ANY less usable for "prepared earlier" character generation. In fact that additional information can only be considered to be an additional tool in the GM's toolkit for prepared earlier material.

I will make this clear, average encounter based balance points do NOTHING special to assist in providing balance to prepared earlier material. The fact that that prepared earlier material is in fact LESS balanced for actual specific uses means if anything it provides LESS support.

3) Can CPR be implemented with an achievable complexity level
I am happy with a simple character generation method. You might note that the context of the home brew system I'm describing is at least at its foundation VERY simple indeed.

But as I mentioned if your CPR system substitutes for character generation methodologies you can quiet simply stack CPR contributing items, skills or bonuses into a pile until you reach your preferred total. No out of context skill balancing required, so no class mechanics, no leveling system required so no special limits to expenditure like feat slots or skill ranks, etc...

That is a LOT of savings in on the fly character generation complexity.

And anyway, if CPR is of any use in any form then you can do it on the fly, if you can't do that it does NOTHING. If you can create a CPR mechanic that gives ratings of character power in a reasonable manner on the fly then you can use it to build characters of a reasonable power rating on the fly.

Puzzle Monsters: Or really any challenge that includes any amount of mystery whatsoever. If the players do not know which abilities they should be using, then the value of their abilities is right back to being an average power based on your speculation of what abilities they could use. If the players don't know ahead of time that the Baron is in fact a werewolf than the fact that the party white mage has moon ray is something that may or may not radically alter the party's effectiveness in the upcoming situation.[/list]

Two things.

1) I'm very much pro-transparency anti mystery. If you need to pull Gygaxian secret rules and puzzle monsters out to support your argument you are gaining no traction with the likes of me.

2) CPR is a tool for balancing character generation and encounter generation. The vast majority of character and encounter generation is performed by the GM, and the GM is expected to have access to all the relevant information unless you want to do double blind Gygaxian and have the players keeping secret information

But it's not going to replace making people purchase their powers ahead of time based on a theoretical average future utility of those powers based on a best guess of campaign direction. It can't.

Can't indeed.

Certainly won't.

But what it does do is act as a tool to facilitate that very activity. And does so on a campaign and game specific level.

If you are purchasing abilities that have CPR guidelines about "Fire Resistance" and your GM tells you you are playing in an adventure with a large number "Fire Resistant" enemies you are better off than in a non CPR character building system.

Even if you ignore that information and go ahead and create a Fire Mage who only uses Fire anyway the GM can measure your CPR and tone down fire encounters or give out extra character build options accordingly.

And beyond that, it probably shouldn't be used to scale difficulty based rewards, because in doing so you would end up making conditional abilities feel worthless. If making a challenge easy by having the right tool for the job ends up giving you challenge rewards for an easy challenge i makes having the right tools pointless.

The right tools ARE pointless.

This is RPGs we are talking. They are story driven games with story driven events and abilities. But then whatever those events and abilities are there is a requirement for a certain level of balance, fairness, PC survivability and PC victory.

And that doesn't change by using an "average encounter" type rubric to judge where characters fall within those acceptable ranges.

Indeed within an "average encounter" rubric that actually works you have two solutions.

1) Story cannot in fact drive events or character abilities (or even player choices!) at all. Events, abilities, and choices MUST adhere to a standard spread across the assumed average or the game is imbalanced, unfair and not meeting its goals.

2) Story and choices cannot meaningfully impact the expected outcomes. The extremes of the average must be as close to the mean as possible. Because any extremes that do exist within the control of story or choice based events could be the extreme that the game spends 100% of its time implementing.

If you instead use a CPR system you still ultimately are bound by your balance points of survivability and success, but choices can matter MORE because there is no specific requirement to drive events by balance policy rather than player choices, and the extremes of the range can be wider because CPR provides a mechanism for the players and GM to know where in the range they are falling and control that on a case by case basis rather than just HOPING they are hitting the "average" sweet spots.

Effectively a CPR system provides the greatest possible freedom for rewarding player tactics and good character builds. It is also very forgiving of bad character builds, but not especially forgiving of bad player tactics.

Just not as revolutionary as you seem to hope.

Before this there just plain wasn't a rubric I was prepared to use for relative character power for my own rules. So any at all IS pretty revolutionary.

But this is a very exciting mechanic in a lot of ways. I'm liking this one a lot, and indeed a lot MORE than I liked abstract positioning systems and damage as a penalty to defense vs death systems.

And both of those seem to have caught on just a bit around here since I started talking about them.

Heck I liked social combat, segregation of game phases, and formalized chase resolution even less, but enough to be an advocate early enough that I might have been the first around here to mention them and those seem pretty popular these days.

So I'm just saying, I'm a trend setter, and I'm calling this one the next big trend.

Username17 · Post by **Username17** » Sat Apr 18, 2009 1:55 pm

PL wrote:The right tools ARE pointless.

OK, you seriously just lost me. In a persistent world there should be persistent challenges. Characters should be encouraged and capable of seeking out actions that their abilities are applicable to.

If you want to do away with that, your system is not relevant to my interests.

-Username17

RandomCasualty2 · Post by **RandomCasualty2** » Sat Apr 18, 2009 5:08 pm

PhoneLobster wrote: 1) CPR is something that responds to a rather specific set of requirements and has arisen from a specific argument. You and RC simple were having none of this "here is something I prepared earlier" approach to NPCs, it was on the fly or nothing. CPR leans toward supporting on the fly rather than prepared earlier for much the same reason.

Here's the problem I'm seeing. If you're doing this as an on the fly system, I imagine it may have a lot of problems, given that to get your CPR baseline you have to calculate the CPR for each monster and each PC based on the opposition. You'd even have to pretty much do it for each spell the PC casters chose to prepare for that day.

And that's a lot of calculating, since every spell for a 10th level wizard and 10th level cleric is going to have to have its own CPR rating.

If you're planning on calculating that thing on the fly, think again.

This is another one of those ideas you had that seem like it would work for a video game RPG (because computers do math fast), but would impractical for tabletop.

Kaelik · Post by **Kaelik** » Sat Apr 18, 2009 6:05 pm

In fact, it's even worse then that, because CPR itself depends on what you are facing.

If you have a Beguiler, you can't calculate his CPR until after you know what enemies he is facing. What if they have good will saves? What if they have immunity.

This might change based on system, but for example, say you calculate CPR, then you start picking enemies, then when you realize some abilities are less useful against those enemies, so you recalculate CPR again, and then repick enemies.

So basically, iterative recalculations. even if it's only 2 or 3, that's still not going to happen in PnP.

Ice9 · Post by **Ice9** » Sat Apr 18, 2009 6:43 pm

There's more than one type of balance. CPR handles the first type (balancing an encounter against the PCs) just fine - better than a static CR system could, although with a lot more on the fly calculation needed.

But as for the second type - balancing the PCs against each-other - CPR doesn't seem to help, except in a "hindsight" kind of way. That is, you can look back and see who's been more effective in the majority of combats, but that information is only valid as long as the type of foes remains the same, so even if you're willing to do retroactive character rebuilding, it may just push things further out of wack against the next foes you face.

Now one thing you could do is to, at the beginning of the campaign, compare the characters against a range of challenges that they'll most likely face in future, average the results out, and balance character creation based on that ... which is exactly what CPR is supposed to be an alternative to.

Also, while it's always good to know how tough an encounter will be, and scaling an encounter to fit sometimes makes sense, always scaling your encounters runs into the problem that tactics and preparation become counterproductive, or at least useless.
"So we're going to face the dragon - should we prepare ourselves against such a foe?"
"Nah, that would just make it tougher, or maybe give it more minions."
"What about sneaking up on it?"
"Again, that would just make it tougher."
"Well we may as well march in playing trumpets and juggling pies then!"
"Hmm, you may be on to something - that would probably weaken the dragon significantly, especially if we left all our weapons outside and beat each-other up first."

Roy · Post by **Roy** » Sat Apr 18, 2009 6:46 pm

Exactly why retroactive bullshit like that = Fail, fuck you DM, I'm playing Smash Brothers.

PhoneLobster · Post by **PhoneLobster** » Sat Apr 18, 2009 10:54 pm

FrankTrollman wrote:In a persistent world there should be persistent challenges. Characters should be encouraged and capable of seeking out actions that their abilities are applicable to.

You again advocate that the game must have predesigned material that cannot be changed by the GM to fit his personal game.

That is unacceptable.

If you want to keep on living in a fantasy land where everyone's game is a drab and perfectly predictable clone of the same game test then good for you.

Ice9 wrote:Also, while it's always good to know how tough an encounter will be, and scaling an encounter to fit sometimes makes sense, always scaling your encounters runs into the problem that tactics and preparation become counterproductive, or at least useless.

I think you fail on understanding the system here.

CPR just measures how much your collected powers are applicable to an encounter. This effectively provides the option for on the fly rebalancing with measured adjustments.

It by no means enforces a system where every challenge always changes to meet an exact CPR. As for every challenge adapting to meet some vague range of CPR. THAT IS WHAT STANDARD AVERAGE ENCOUNTER SYSTEMS ALREADY DO. Only with them you don't know exactly where in the range each encounter falls.

Simply put in EITHER system your choices and tactics are permitted to have an effect and weigh an encounter one way or another. But in both systems break if you permit choices to go outside of the bounds of the designed balance points of the system, that is an unavoidable aspect of RPG design.

It is utterly invalid to present it as Frank is as a criticism of CPR because it applies to "average encounter" systems as well, and they are less good at measuring when and how it happens and supporting the extreme ends of their assumed difficulty ranges.

RC wrote:"My care bear complexity gut hurts!"

I've told you what I'm doing about that now shut the fuck up and wait your turn rather than doing your "feelings of complexity" wanking all over my thread.

Murtak · Post by **Murtak** » Sat Apr 18, 2009 11:28 pm

PhoneLobster wrote:
FrankTrollman wrote:In a persistent world there should be persistent challenges. Characters should be encouraged and capable of seeking out actions that their abilities are applicable to.
You again advocate that the game must have predesigned material that cannot be changed by the GM to fit his personal game.

That is unacceptable.

I read this as "If I want to play a Hunter of the Dead I damn well expect to smash undead to pieces and I'm going to actively seek them out".

PhoneLobster · Post by **PhoneLobster** » Sat Apr 18, 2009 11:49 pm

Murtak wrote:I read this as "If I want to play a Hunter of the Dead I damn well expect to smash undead to pieces and I'm going to actively seek them out".

Good, you should do that. And in an average encounter balanced system you just broke the fucking game if you succeed. Because you were balanced on the assumption of a specific proportion of undead encounters only.

CPR says "sure no problem" it measures you vs all those undead and if need be provides the option to rebalance accordingly.

And this "option" bit is important. CPR is just a measuring stick, it provides information and how you (at a GM and player level) use it is up to you.

Indeed if the GM wants to reward you, even unfairly reward you for pushing game events in your favor CPR provides the measuring stick that lets him know just how much he is doing so.

All this "You is taking my choice away! Systems which regard my choices as conforming to a vast sweeping average give me more choice!" is like declaring that the existence of the Tape Measure means that all pieces of wood automatically become the same length.

In fact I'm going to take that metaphor and run with it.

Imagine that carpenters cannot measure the length of pieces of wood by any other means than an eyeballed gut feeling.

A company has a method for sorting pieces of wood that guarantees that the pieces of wood fall within a known average length of 3m, but makes no specific guarantees about any single piece of wood's length.

Some guy instead proposes the use of a tape measure. Allowing him to know the actual length of any given piece of wood, so he can get wood from the sorting company, or elsewhere and know exactly how long it is.

Then everyone says "That's a stupid idea because all your pieces of wood are going to be 3m long forever! The average sorting company is better because there is (unknown) variation in their wood!".

Yeah sure. I think they are missing something in their analysis. Indeed I think their analysis is still skewed by the assumptions they are carrying over from the failings of the prior system.

Murtak · Post by **Murtak** » Sun Apr 19, 2009 12:03 am

PhoneLobster wrote:
Murtak wrote:I read this as "If I want to play a Hunter of the Dead I damn well expect to smash undead to pieces and I'm going to actively seek them out".
Good, you should do that. And in an average encounter balanced system you just broke the fucking game if you succeed. Because you were balanced on the assumption of a specific proportion of undead encounters only.

CPR says "sure no problem" it measures you vs all those undead and if need be provides the option to rebalance accordingly.

Fine. Let us assume you have a working, accurate CPR model. We have the above situation. Now what? Do the monsters get tougher? Do the characters get less experience? If the monsters get tougher it is impossible to specialize, so I guess that one is out. So I guess the characters get less experience. I guess that sort of works, but it seems like an awful complicated way of doing things.

If that is all you are using CPR for you can get an absolutely perfect system without any calculations at all by using the following system:
- The characters level up every x sessions.
Because that is what handing out experience by CPR amounts to. Mind you, there is nothing wrong with that goal.

PhoneLobster · Post by **PhoneLobster** » Sun Apr 19, 2009 1:14 am

Righto. I've discussed what CPR is and some of it's benefits. And I've explained the basic context of the mechanics I want to apply it to.

So now for a proposal on how to apply a CPR implementation to those mechanics.

The Cunning CPR Plan
So now I need to take that system and create some sort of CPR methodology that allows a moderately simplified means of rating some set of opponents in whatever environmental context against each other in a measurable way that can predict combat outcomes.

Plan A)
The system would basically calculate net numeric bonuses to attack and defense prior to combat. These are the same numbers you add to your combat rolls anyway so that's both no big deal and kind of handy to know in advance anyway.

Every non numeric special ability would apply an additional “extra” bonus to attack or defense totaled separately and then considered in combination with the numeric bonus. This bonus would be measured in terms of the value of the special ability if it was converted to a flat numeric bonus, effectively trying to measure non numeric bonuses on the same scale as numeric ones.

You then compare the net attack and defense values of the opposing characters take the difference and refer to a very simple table that tells you what average damage per turn for a character with X advantage or disadvantage to their attack is and you can use that to determine which side is likely to win, and even in the process potentially gain information about how many casualties are likely in the process.

Plan B)
I have a rival plan to plan A where the numeric bonuses are used straight up to determine an average damage per turn and everything else is calculated on the scale of how many extra turns worth of damage output it provides or removes. It is a potentially more accurate way of dealing with abilities that do things like provide extra attacks, but also more complex.

For now I'm sticking with plan A for the most part and only applying something resembling plan B for the accounting of "Block" abilities.

So blocks are totaled up for each character and then are considered to provide an additional number of turns of likely life span for that character based on the average damage the opposing side is dealing per turn that the full CPR comparison of attack and defense abilities has already given you.

The C bit in CPR
OK so up until now both Plan A and Plan B are basically little more than points based systems with a mildly well defined basis for what constitutes a “point”.

So you need to introduce the “Context” bit.

And that bit is EASY, especially in this system. Your bonuses and abilities mostly run off the presence keywords on your targets. So when you get +4 Defence vs Giants (like a D&D dwarf) then it comes with the built in CPR proviso “This ability only counts when the encounter is against Giants”

And once again, this is exactly the same “Does this bonus apply?” test and part of the same calculation that you use whenever making an attack anyway. So it's not exactly HARD. And it IS very much associated directly with the mechanical impact of the ability, which makes calculating it as a point of comparison even more useful.

So anyway. All your abilities either will apply in a given encounter or not. Any which are somehow less directly clear than “+4 Defence vs Giants” in whether they apply include some conditional statement to determine if they do apply like “this ability applies +5 extra to attack rating if facing more than 1 opponent”. And you can do that with MUCH simpler notation and wording once you apply a little terminology and some standardization.

And THAT is the “Contextual” bit. That is the bit that makes it different to a standard points based system. That is the bit that the entire system from here on out gets built (or in my case rebuilt) around.

Assumptions I'm Making to make this work
I'm making a number of assumptions to make this work.

1) Dealing with multiple Opponents
OK so you are likely to deal with multiple opponents and at least the NPC side of the equation is likely to be going up against diverse groups of PC opponents.

The assumption being made is that you calculate the attack component for CPR for each type of character based on their strongest attack against their best possible target on the opposing side.

You calculate the Defense component for each type of character based on their best defenses against their strongest possible attacker.

All in all this means that the error in CPR estimates that this introduces slightly favors the more diverse group (usually the PCs) over the more homogeneous groups (usually NPCs). And begins to lose accuracy starting from the fall of the weakest target/s in any diverse group.

Of course the CPR method described here provides the tools to actually calculate additional CPR comparisons of things like the next weakest target and so on. So if you wanted to you CAN add additional working time and complexity and gain direct benefits in additional information and accuracy.

1a) Attacking characters will always be able to attack their best possible target.
Just restating part of the first assumption clearly and as it's own statement. CPR is weighted towards determining maximum fatality of an encounter, because it is that extreme that we are most concerned about predicting and managing.

1b) Defending characters use their best possible defenses against their strongest possible attacker.
Basically the other half of 1a) just so it's clear.

1c) Abilities with bonuses and penalties will always be activated favoring Offensive bonuses
So if you have an ability that says "-4 defense +4 attack" CPR assumes you will use it, while it won't assume you activate the reverse ability.

Not that I intend to have any such abilities in my rules set, but that's the guideline I will use for CPR accounting for them should I suddenly include something like that.

2) It determines all ratings based on “start of combat” conditions.
Since CPR is being used primarily to determine if a combat is an appropriate challenge for a party of characters I have decided that it will run all its estimates off the resources and the condition of characters at the start of combat.

If a character walks into combat with a wound or status effect that reduces their abilities CPR will account for it. Which is a good thing.

But if someone pulls off a lucky shot first round of combat and throws down a wound or status effect that would have altered CPR had it only applied before combat it is frankly a bit late for it to seriously impact CPR. Effectively how can you use a system designed to determine if a combat is appropriate after you have already committed to a combat?

Of course to some degree the estimates that CPR gives you account for such cases as part of an average, and if you DID want to recalculate CPR the tools are provided by it's very nature (though what you intend to do with that information once already in combat I don't know)

3) It still only gives average estimates of damage output/character life span in combat
I mean characters can still walk in and roll all 20s or all 1s and characters estimated to last 6 rounds of combat by CPR can go down in round 1. But the idea is that CPR accounts as much as possible and in detail for everything BUT the actual dice rolls.

So CPR can give you a firm idea of the average results of a specific encounter, while the “average encounter” balance point for other challenge rating system can only give the average result for an non specific “average encounter”.

4) It doesn't account entirely well for limited use abilities
I'm rebuilding my homebrew from scratch to accommodate CPR, so right now I'm not sure if abilities with “limited ammunition” are going to be in it. But frankly they, or something vaguely like them are pretty likely.

At the moment this is falling under “start of combat” condition, plus “best options/best target” and the assumption that if you have limited use powers that would beneficially apply you WILL use them.

Further it is assumed that if you don't use them it was because you happened to be doing fairly well anyway. Effectively if you don't feel the need to use some big gun your character must be performing sufficiently better than average for the result to be basically similar.

And So...
All in all these assumptions ARE assumptions and points where accuracy of CPR can potentially degrade if or when the assumptions are invalid to whatever degree.

But considering these assumptions remove and replace the “average encounter” assumption that all encounters WILL add up to a perfect pre assumed average of some form they are certainly the LESSER of two evils. By a long shot.

In addition these are fairly specifically defined assumptions. It is somewhat clearer when they are breaking and what to do about it. CPR even provides the basic measuring sticks you need to do further work towards fixing it. Should you be so inclined.

The complexity of this system
OK so these are the sorts of operations players are expected to go through in a single round of combat in this system...

Determine movement, ranges, optimal numbers of targets, and biggest attack.

Declare and resolve "special" actions like movement and other stuff.

Total relevant attack bonuses and defense bonuses

Roll and read dice, add to total.

Compare totals

Apply results such as injuries.

Apply blocks.

And this is an example of the operations players are expected to perform once before combat to calculate CPR.

Determine weakest targets/best attacks.

Total relevant numeric bonuses to attack and defense and "extra" bonuses representing relevant special abilities.

(all the movement, special abilities, multiple target etc... abilities just changed from "determine and resolve" to refer to CPR value and add to your totals instead).

Do NOT roll and further total, simply compare to determine the difference between attack and defense for each attacker.

Refer that difference to a table to get average damage per turn for each character.

Total average damage per turn on each side and compare to opposing numbers of characters/blocks to determine who is likely to win, how many casualties are likely to occur and even how long it will likely take.

So, CPR is coming in at a series of operations fairly similar in complexity to resolving a round of combat. That is the complexity goal I set as fairly acceptable and for that cost I'm calling the information that CPR is delivering in this system a bargain.

Using CPR as a character building system
OK so the idea is now you have all your abilities stated out with CPR values either equal to their flat numeric bonuses or to an estimate of their value converted into such.

You have these values effectively measured in attack bonus, defense bonus and number of blocks.

And since you have this notation anyway you could use your CPR values as points values for a points based character building system.

The point where that falls down is in an ability like say...

"Immune to all attacks with the Weak keyword".

Because that ability is worth ZERO defense and attack bonus for calculating CPR. It simply acts as a conditional statement that effects an opposing characters CPR as they become unable to add Weak keyword stuff to their total attack. And indeed by the end of resolution could even end up with zero average damage per turn if they can't make any attacks without "Weak" attached.

Well. Lets say I'm working on this one and throw in a good measure of, well that's blanket immunities for you, stuffing things up like always...

Where to from here
OK so from here on out my work with my home brew begins to depart from direct relevance to CPR policy wonk.

I've got to redefine all sorts of non numeric abilities I've been using or would like to implement and give them values on the same scale as attack and defense bonuses.

That IS a hard process with plenty of potential pitfalls but it is also the same process that is gone through in every RPG design that attempts to measure the value of abilities in any remotely uniform way. If that is impossible then remotely balanced RPG design in general is impossible.

I'm planning to attempt to value and define abilities in the smallest possible components with information on their combination and effects on each other's value.

The core rules then instead of saying "Silver Sword= +4 attack vs Slow, attack gains ArmorPiercing and Fast keywords" Will simply define those keywords, and bonuses, their effects and their values. The actual Silver Sword thing will be an item that is constructed from those core components and presented in campaign specific rules sets.

So then you can treat the core either as a points based system that players and GMs could interact directly with building custom items and powers, OR you can treat it as I intend to, as basically all the "working" behind the derived powers and items that will actually be used in the game setting so that GMs and Players can easily back engineer, understand and if need be modify the campaign material for their own game.

But that is running more into extensibility territory than CPR. CPR helps because now I have SOME sort of evaluation metric such a component based core has some actual meaning for back engineering and modification instead of merely defining what abilities do and requiring a big fat "eye ball it" as the valuation methodology. But a traditional points based system would have had vaguely similar impact.

Anyway, I think I've defined most of what I need to to explain CPR and how I'm implementing it in my specific rules set. RC can now go and have his complexity emotion fest if he wants as he has material he can directly address or ignore with real considerations of complexity instead of his care bear gut.

I mean there is plenty more to discuss including addressing specific special ability types and maybe a bit extra on CPR as a non specific encounter metric for character building, but this seems pretty sufficient as a start.

CatharzGodfoot · Post by **CatharzGodfoot** » Sun Apr 19, 2009 1:42 am

PhoneLobster wrote: "Immune to all attacks with the Weak keyword".

Because that ability is worth ZERO defense and attack bonus for calculating CPR. It simply acts as a conditional statement that effects an opposing characters CPR as they become unable to add Weak keyword stuff to their total attack. And indeed by the end of resolution could even end up with zero average damage per turn if they can't make any attacks without "Weak" attached.

Well. Lets say I'm working on this one and throw in a good measure of, well that's blanket immunities for you, stuffing things up like always...

How is that a problem? If an attacker is going to be weak against that creature, then the ability counts as total immunity. If an attacker isn't the ability doesn't matter.

And shouldn't a character's ability to attack a certain target alter CPR? I mean, if you have the above monster, and give it flight and laser eyes, then it should matter if the only character with a bow is weak against it. If you don't factor that it, it kinda kills the entire concept...

PhoneLobster · Post by **PhoneLobster** » Sun Apr 19, 2009 1:47 am

Murtak wrote:Fine. Let us assume you have a working, accurate CPR model. We have the above situation. Now what? Do the monsters get tougher? Do the characters get less experience? If the monsters get tougher it is impossible to specialize, so I guess that one is out. So I guess the characters get less experience. I guess that sort of works, but it seems like an awful complicated way of doing things.

I'm not planning to use an experience and leveling system. I'm planning on using CPR to permit me to give out (or freely allow collection of) items and training montages, effectively it's the same as "level up based on fluff events".

Because CPR lets me measure the impact of that and adjust game play (as or when appropriate only) it helps me to get away with just saying "yeah sure, that guy can teach you the cool trick he just did with his axe" as an advancement methodology.

You see experience and level systems, from a balance stand point, are exclusively about measuring character power and keeping it within certain limits. CPR lets me do that tailored to flavor with greater accuracy, so I don't need experience or level systems.

But on all your other points... what the heck?

I mean what exactly do you imagine are the "average encounter" system's answers to those questions?

And what part of having some actual metric by which to measure the information you need to ask the question "is this combat too easy/too hard" is in any way worse than having less information but still needing to ask the same question?

CPR lets me know how hard an encounter is. I can change the encounter, I can not change the encounter, but that is damn important information.

Without it an average encounter system leaves you unaware as to whether an encounter is too hard or easy, and what impact any changes you decide to make might be.

Really what possible benefit is there to an "average encounter" balance point in the case of the specialist who always faces his specialty?

That's the definition of the broken balance point right there. And without CPR you have no mechanic to approach it and interact with the breakage. Where the heck are you going with this line of argument?

PhoneLobster · Post by **PhoneLobster** » Sun Apr 19, 2009 1:52 am

CatharzGodfoot wrote:How is that a problem? If an attacker is going to be weak against that creature, then the ability counts as total immunity. If an attacker isn't the ability doesn't matter.

And that is exactly how CPR will work.

The problem there is that I was discussing that example in the CPR as a character building metric bit.

So if you want to also use CPR to build characters without any specific encounter in mind you have CPR points ratings for stuff and that mostly works just like a points system.

Until you encounter an ability that has 0 CPR value to the character it is on and just modifies CPR on other characters. Like immunity, or any keyword that doesn't specifically do anything but might be targeted by an opposing bonus.

They work fine for the primary use of CPR as an encounter challenge rating tool, but are potentially an issue with it's secondary use to build "prepared earlier" characters with no specific combat encounter in mind.

Draco_Argentum · Post by **Draco_Argentum** » Sun Apr 19, 2009 10:50 am

PhoneLobster wrote:You are asking how do you weight the value of two bonuses when one has an action cost and another doesn't?

Not really. I'm asking how you will weight the value of synnergistic abilities. Active ability C does quad damage to a stunned target. C's value is entirely dependent on what the party can do to cause stuns.

PhoneLobster · Post by **PhoneLobster** » Sun Apr 19, 2009 11:08 am

Draco_Argentum wrote:Not really. I'm asking how you will weight the value of synnergistic abilities. Active ability C does quad damage to a stunned target. C's value is entirely dependent on what the party can do to cause stuns.

I've put some thought into that one already.

It's pretty simple. Any ability that requires such a setup (and I intend to have some) counts as worth zero CPR unless an allied character has an ability that can provide the setup.

Somewhere along the line I was considering a CPR weighting based on the expected frequency of use. So if the setup can only go off once every 2 or three turns the finishing move only gets rated at about 1/2 or 1/3 value. But for something as simple as "attack 1 deals stuns, attack 2 is cooler but requires stuns" that could happen basically every turn of combat it's basically attack 2 gets rated at full value as long as attack 1 is present.

There is the additional proviso however that the CPR rating system and basic combat mechanics I have implemented means that if your basic numeric attack bonuses don't put you on the RNG or put you in the poor end of it your setup attack is going to go off very infrequently and so your finishing move is going to go off very infrequently and be of less value. And as you gain basic attack advantage your finishing move becomes more valuable.

Right now the method I'm using to account for that is that finishing moves and the like add to the "extra bonus" to attack that you add to your actual same as in combat attack bonus when calculating CPR.

This extra bonus to attack is labeled separately and totaled up separately to the regular flat numeric bonuses to attack and only added at the last step just before comparing it to target defense totals.

The idea being that a net attack value relying on a large margin of "extra bonus" from things like finishing moves, extra attacks, long range attacks etc... is a CPR total that indicates the character might not be as reliably effective as one with a total derived entirely from basic numeric bonuses.

It's a questionable aspect of the system and is one of the reasons I am contemplating a conversion of a larger part of the system to a sort of "Plan B" style accounting for such abilities which may well deliver greater accuracy when dealing with additional abilities that change in effectiveness depending on how much of an advantage or disadvantage a character has in basic attack/defense.

Username17 · Post by **Username17** » Sun Apr 19, 2009 11:38 am

Ugh. Yeah, you do that.

Without the aid of a computer, this system is completely unworkable. And I still don't see any actual advantages because players still have to initially purchase their abilities based on non-contextual assumptions of ability utility; and any specific reactions by the game to the CPR effects actively and specifically counteract the effects of intelligent play.

Do not want.

-Username17

PhoneLobster · Post by **PhoneLobster** » Sun Apr 19, 2009 11:50 am

FrankTrollman wrote:Ugh.

You manage to raise not one single point I haven't already addressed and/or debunked.

I am especially offended by the "counteracts intelligent play thing". I addressed that in fucking detail and you are talking out your fucking ass by harping on over something I've debunked without for a second addressing my material debunking it.

If you are going to be like that then fuck off out of my thread.