A Diceless Resolution System using Resource Management

[Big Simon]

This is a repost from RPG.net, but I thought I'd run it by you guys (for those of you who don't browse that other site).  

Please keep in mind that this resolution system is currently systemless.  That is to say, I don't have specific stats/attributes/skills in mind, though I do explore some when I get to the combat example listed below.

Starting Out – Contests and Initiative

Each player has an assigned number of Action Points, as determined during character creation.  Action Points are spent for any action that might be opposed – from blaster combat with alien space pirates to a debate with a staunch political opponent, to hacking a popular video game's high score list.

Any such conflict is called a Contest.  When entering into a Contest, each player wishing to act first in the Contest must make a bid in order take the initiative.  The player with the highest initiative total (adding the initiative bid to the appropriate stat) pays the amount bid and declares his action.  From that point on, if another player wishes to take the initiative in the contest, he must wait until after the player with initiative acts at least once, and then bid higher than the current initiative amount for the next action.  Either that, or wait until the player with initiative either chooses not to act any longer, or cannot act due to lack of Action Points.

Any time a player either chooses not to act any longer, or cannot act due to lack of Action Points, any other player wishing to act may take part in another initiative bid.

An Example of Combat Resolution

Combat actions, like any opposed actions, are Contests.  At the beginning of a Contest, players make a grab for initiative (see above).  The player who wins the initiative bid sets reduces his available Action Points by the amount of the bid (the point bid, not AP + Stat), and then may use his remaining Action Points for combat actions.  A player may spend as many Action Points as he wishes on any combat action.

Only the player with initiative may make attack or movement actions.  Others may only act in defense.  

To make an attack, a player simply pays the number of Action Points he's willing to invest in that attack.  To defend against an attack, a player pays the number of Action Points she's willing to invest in that defense.  Any time an attack bid (skill + AP spent) is higher than the defense bid (skill + AP spent), the attack is successful.  An attack does damage based on the weapon type, plus the number of points the attack bid exceeds the defense bid (known as an attack's "Power").

For Example:  Bob's character, Zanthos, has 30 Action Points.  Melanie is playing Hythera, with 32 Action Points.  They have found themselves pursuing Margun (AP36), godfather of an intergalactic cadre of mobsters.  They've boards his private carrier and have cornered him on the bridge.  The GM has informed them that Margun has his blaster drawn in one hand, and a Karavian Battle Dagger in the other.  Looks like there's going to be a fight.

GM
Make your bid for initiative.  Margun bids six.

Bob
(Checks his character sheet)
I can go seven.

Melanie
I think I'll hold my action.

GM
Margun is going to up his bid to ten.

Bob
I'm out... I'll hold on to my points for later.

GM
(Initiative (4) + 6AP.  Total:  10.  Reduces available AP by 6.)
Margun levels his blaster at Hythera and lets a couple bolts sizzle through the air.
(Attack (5) + 5AP.  Total:  10.  Reduces available AP by 5.)

Melanie
(sets aside seven action points, reducing her total pool to 25)
Hythera leaps to the side, tucking into a somersault.  Hopefully a computer console will offer some protection.  That gives her a defense of twelve.
(Defense (5) + 7AP.  Total:  12.  Reduces available AP by 7.)

GM
That's enough to keep her safe... for now.  He takes aim at Zanthos, who is still standing out in the open, and fires.
(Attack (5) + 10AP.  Total: 15.  Reduces available AP by 10)

Bob
Zarthos spins to the left to avoid the shots.  I'm spending five points in his defense.  Hopefully that's enough.
(Defense (6) + 5AP.  Total: 11.  Reduces available AP by 5.)

GM
Not quite quick enough, Zanthos is struck by blaster fire for seven points of Serious damage.
(Blaster (3) + Power (4).  Total:  7)  

(Keep in mind that Margun now has only 15 Action Points remaining for this round.  Hythera has 25.  Zanthos has 25.)

Melanie
Okay... it's time for Hythera to step up.  I bid eleven points to take initiative.
(Initiative (4) + 7AP.  Total:  11.  Reduces available AP by 7.)
I pop up from behind the command console and toss a concussion grenade in his direction.

And so on. A combat round continues until all characters have exhausted their Action Points, or until all players have decided to hold action until their characters' next round. Then a new round begins and Action Point pools refresh. A character with AP remaining at the end of the round may carry up to half his full Action Point pool over into the new round. The character that has initiative when a round ends carries it over into the new round. Rounds continue to refresh until the Contest has ended.

Well... that's about all I have for now.

What do you think?  Good?  Bad?  Too complex?  Not complex enough?

[Ben Lehman]

This sounds very similar to some of my work on Cradle, which should be going up on my website very shortly.  Also, it is pretty similar to Nobilis.

What was a big problem for me was the manner of setting the base difficulty numbers and expenditures for the GM.  The GM is never going to have other characters fully statted, and when everything is a comparison scale, the GM can easily just throw out a number that they know the player will win or lose.  This is a little sad, for various reasons.  Fixing this problem can be done in one of several ways.  Nobilis fixes point totals to an absolute scale, so you know exactly what seven points can do, which gives some authority to the player.  Cradle limits the GM's ability to set traits.  You might not care (although then you might as well play a GM-fiat based game) or you might want to find your own solution.

Have you considered capping the AP expenditure on any given challenge?   Have you considered how to limit skills so that people don't try to apply their one uber-skill to everything?  Have you considered dropping less important challenges like defense, and replacing them with challenges where there are stakes for both sides (for instance, if the defender wins by a large MoS, they do damage back)?  Have you considered (and this is super-important) how points refresh?

yrs--
--Ben

[Big Simon]

I'm hoping, first off, that GMs will be honest enough with their games to stat out NPCs that matter.  Mook rules will take care of lesser opponents.  They'll be limited to certain traits, and certain scores in certain traits... as well as certain levels of AP expenditure.

One of the reasons I didn't want to cap AP expenditure for PCs on indivudual actions is that I wanted to give players a chance to succeed in huge ways.  But these tremendous actions will lead to a character being able to do less later on in that round, and leaving him open to attacks from other opponents (since we're talking about combat).  Of course, I've considered power levels, as well, where characters of lesser power levels will be limited to a given amount of AP per action (say, equal to the character's power level).  Another option I've pondered is giving actions a point range - both minimum and maximum.

I've also been thinking about allowing tremendous successes in actions like defense have some added effects, but haven't added them in yet.  I'd hate to think that people who make outrageous defensive moves would be penalized for it.

Under the curreent system, APs refresh after every combat round.  Players will be able to carry left-over APs to the next round (up to half their total AP), so those who don't use them won't necessarily lose them.

Like I said... work in progress.  Thanks for looking it over, and the advice!

[TonyLB]

The GM Fiat isn't only involved in giving the NPCs high traits.  You have to choose how many APs to spend for them on a particular action, right?  Well then you're deciding (on GM Fiat) whether to make something easy as pie or impossibly difficult.  And since you aren't telling them (and thus they can't reply to a very difficult situation with extra effort), your fiat is very likely to turn into deciding success and failure, straight up.

I had a lot of that in Castle Falkenstein and Nobilis.  These days I leave that job to my impartial little scape-goats, the dice.  I can't be blamed... I just roll them and follow their imperious commands.

[dredd_funk]

I think you can eliminate the problem of GM-fiat in action point allocation simply by ensuring that the process is double-blind (or completely transparent with defenders allocating last), one of which I'd be surprised if bigsimon didn't have in mind (I've worked out something very similar in my own game).

If we'd gotten to the part in his example in which the characters fired back at the NPC we reach the crux of a decision: does the GM get to allocate the NPC's defense action points before or after the players have announced their allocation of offensive action points?  If the GM allocates after the players, then GM-fiat is facilitated.  If the GM allocates before the players, but in secret, then GM-fiat, as far as the allocation of action points goes, is prevented. The process has to be double-blind so that neither the GM nor the players derive an advantage from the other side declaring first.  Or, conversely, that the mechanic dictates that the defensive player is always allowed to allocate after the offensive player has announced their allocation.  In these situations, both the GM and players are treated equally by the mechanic and the risk/rewards that it builds in.

As Ben has said, there are a number of good ways to limit GM-fiat in selecting the challenge level.  Codifying the precise limits in a textual 'authority' (i.e. the 'rules') upon which both the GM and the players can draw for credibility is probably the best thing to do.  The old D&D 'Monster Manual' was a pretty good example of creating a textual authority for setting creature abilities, though it didn't address how to properly select those creatures.  That goes for any type of challenge really, not just combat.

How you go about doing it mechanically in your own rules is pretty much up to you.  In this case, since action points dictate how much each NPC or character can achieve, setting limits on the maximum difference between NPC and player(s) action points would be one way.  That way, the GM can't simply create an ad-hoc NPC that has three times as many action points as the entire party of players--in effect an example of GM-fiat by stat selection.

It's good to be aware of the question, though, so thanks Ben, you've given me a number of good questions to think about for my own game!

Oh, and hello, my name is Chris.
Cheers.

[TonyLB]

If the GM allocates before the players, but in secret, then GM-fiat, as far as the allocation of action points goes, is prevented.

Good in theory, not so good in practice.  Again, I reference both Nobilis and Castle Falkenstein, which do have these double-blinds in place.

If your base skill is five, and you can allocate anything from 1 to (say) 10, then allocating a 1 is deliberately giving the victory to the player, and allocating a 10 is deliberately preventing them.  The number of times that you will be that surprised by their action bid is very, very few.  It happens, but not enough to undermine the overall feeling.

If the AP spends center around five then you might still be out of control if you bid 4 or 6, but the strength of your fiat rises radically if your bid goes even as far as 3 or 7.  My experience is that after a very short time it feels indistinguishable from just deciding what's going to happen, about 90% of the time.  If your experience substantially differs then maybe we should figure out where the difference arises.

[andy]

While the crunchy gamist in me would dearly miss the dice, I think that you're on the right track with a bidding mechanic. In my dice-centric game (which includes an allocation/bidding mechanism of sorts), I solved the GM fiat issue by having the players allocate/bid last. This gave them an advantage, but then they're the heroes.

Have you checked out the excellent (and free) Active Exploits? It is close to what you are envisioning.

Happy gaming.

Andy

[dredd_funk]

I haven't played the two games you've mentioned and thus we're probably imagining slightly different dynamics.  In my mind I was not thinking of 'one-shot' questions of skill that are completely isolated from other in-game events.

In this respect, I think we're speaking to cross purposes concerning fiat (player or GM) for a particular action, and fiat for the resolution of a challenge.  I wouldn't disagree that, if the resolution of a singular action was--by itself--the entire challenge, then fiat definitely enters into play.

This seems to be the situation you are envisioning, no?  E.g.  GM--I know that the player has a Disarm skill of 5 and can allocate from 1-10 action points to the attempt to disarm this particular trap, if I, as the GM, set the trap difficulty to 16 or greater, the attempt will automatically fail.  If I set the difficulty to 5 or lower, the attempt will automatically succeed.

The key question for the mechanic under discussion is how it could effectively handle such 'one-shot' situations without fiat entering in.  It probably couldn't without modification--I agree there, no doubt.  Specifically, in the case of the 'disarm trap' example, the questions are: how do you prevent the player from automatically allocating all of their action points (especially if they are not in combat), and is there a counter-balancing mechanic in place that causes player success if the GM sets the difficulty too high?  That does have to be addressed if one wants to limit GM fiat.  The trouble to me appears to be not having these questions simply devolve into a 'fortune' system of mere guessing as opposed to the 'fortune' system of random dice.  The gamist/simulationist side of me would rather see a system where some sort of logical analysis is possible.

However, as a mechanic for multiple-task challenges, such as combat, I think the current system could work without fiat entering into it.  Primarily this is because there is a built-in counter-balance to simply using all of one's action points in a single, massive endeavor: viz. it may--and probably will--take multiple singular successes to successfully win the entire challenge and points used for initiative are lost, thus a de facto 'bonus' is granted to the loser of the initiative to counter-balance the fact that the winner has the right to act first.

This means that a) it isn't clear what it takes to 'win' the current incremental unit of the challenge, and, b) winning a single incremental unit probably won't constitute a 'win' for the entire challenge itself.  In fact, sometimes it may be the immediate cause of defeat.  I'll see if I can explain relative to the example in question.

In the example provided, if the GM were to expend all of Margun's action points in a single effort, surely it would succeed.  The question remains if Margun, now without any action points left, would survive the challenge with no defense.  I'll try to detail what I understand of the mechanic in this  situation:

If the GM has Margun bid 15 of his action points for initiative, leaving only 21 remaining, Margun wins initiative and puts all 21 remaining into a blaster shot at Hythera (21+5=26) but Hythera, with 32 points remaining, selects to defend with 21 action points (total defense of 21+5=26), resulting in a miss (26-26=0), now, however, Margun has 0 action points remaining while Hythera has 11; the counter-attack will be devastating--Hythera wins initiative with a 1, attacks Margun with her remaining 10 action points, and attacks at (10+herWeaponSkill) versus (0+MargunDefenseSkill)--and that provided the designer intends Margun to be able to defend despite having 0 action points.

Margun's in a lot of trouble, even though, by 'fiat' he 'won' the initiative and put everything into his first action.  In a double-blind situation with multiple factors involved like the one in question (I don't know anything about Nobilis or Castle Falkenstein), I'm not seeing how the GM could create a fiat through allocation of action points.  The GM can, of course, simply decide that Margun has 63 action points (1 more than the combination of Hythera and Zanthos), prior to the combat ensuing, but that isn't what we're talking about right now.

I think the designer has come up with an elegant mechanic in that respect: spending action points to determine initiative means that there is a definitive 'cost' to grasping initiative.  The GM and players must 'spend' their action points wisely throughout the round or risk leaving themselves defenseless.

Am I misunderstanding you?  Is there something I'm not taking into consideration?  I'm sure you have more experience game-playing these types of mechanics because I really don't have any.  Do Nobilis and Castle Fralkenstein have 'one-shot' challenge resolution for non-combat skills?  Do they not take into account some counter-balance to ensure, during combat, that one course of action is always best--i.e. you do get penalized for allocating 10 out of 10 sometimes?

Your experience and perspective are appreciated.

Cheers
Chris

[LordSmerf]

If the GM has Margun bid 15 of his action points for initiative, leaving only 21 remaining, Margun wins initiative and puts all 21 remaining into a blaster shot at Hythera (21+5=26) but Hythera, with 32 points remaining, selects to defend with 21 action points (total defense of 21+5=26), resulting in a miss (26-26=0), now, however, Margun has 0 action points remaining while Hythera has 11; the counter-attack will be devastating--Hythera wins initiative with a 1, attacks Margun with her remaining 10 action points, and attacks at (10+herWeaponSkill) versus (0+MargunDefenseSkill)--and that provided the designer intends Margun to be able to defend despite having 0 action points.

Margun's in a lot of trouble, even though, by 'fiat' he 'won' the initiative and put everything into his first action.  In a double-blind situation with multiple factors involved like the one in question (I don't know anything about Nobilis or Castle Falkenstein), I'm not seeing how the GM could create a fiat through allocation of action points.  The GM can, of course, simply decide that Margun has 63 action points (1 more than the combination of Hythera and Zanthos), prior to the combat ensuing, but that isn't what we're talking about right now.

Assuming that the situation is double-blind then it actually becomes a guessing game.  Did the GM allocate all available points to that attack, or just one point?  In your above example the GM has the option of taking Initiative, then attacking with one point to draw a massive defense.  By drawing the defense first, and then spending for an all out attack you pretty much win.  If the player manages to outguess you (i.e. figure out where you actually are going to use your big guns) then they win.  However, that is no longer a case of resource allocation...

Thomas

[TonyLB]

I agree completely with you (EDIT for clarity:  Chris) that this system (and the systems I discussed) are fiat-free on the level of whole encounters.  So we're on the same page, cool!

My concern was that the social problems of GM-fiat arise from its inclusion at the micro-level.  Simply put, if you've got three actions against three different players, you'll probably end up deciding to defeat one of them, softball another and give the third a fair chance.  And then, if your social contract has some expectation of "fairness" then you have to keep careful track of what you've done, and make sure that you don't defeat or softball the same people consistently.

Does that clarify the emphasis I was approaching the question with?

[neelk]

The GM Fiat isn't only involved in giving the NPCs high traits.  You have to choose how many APs to spend for them on a particular action, right?  Well then you're deciding (on GM Fiat) whether to make something easy as pie or impossibly difficult.  And since you aren't telling them (and thus they can't reply to a very difficult situation with extra effort), your fiat is very likely to turn into deciding success and failure, straight up. I had a lot of that in Castle Falkenstein and Nobilis.  These days I leave that job to my impartial little scape-goats, the dice.  I can't be blamed... I just roll them and follow their imperious commands.

When I ran Nobilis, I usually just ended up just flat-out telling the players what the stats of their opponents were, and how many MPs they were spending. At that point, the players could decide whether the conflict was important enough to them that they were willing to spend the MPs to win.

I did this because it was never a real double-blind -- as a GM, I had a pretty good idea what the PCs' stats and MP totals were, and the players usually didn't come in knowing what the NPCs stats and MP totals were. The easiest way to resolve the asymmetry was just to give the players all the relevant information; that way they didn't have to try to "read' me, and could just work on how important the conflict was to their PCs.

[Big Simon]

If your base skill is five, and you can allocate anything from 1 to (say) 10, then allocating a 1 is deliberately giving the victory to the player, and allocating a 10 is deliberately preventing them.  The number of times that you will be that surprised by their action bid is very, very few.  It happens, but not enough to undermine the overall feeling.

If the AP spends center around five then you might still be out of control if you bid 4 or 6, but the strength of your fiat rises radically if your bid goes even as far as 3 or 7.  My experience is that after a very short time it feels indistinguishable from just deciding what's going to happen, about 90% of the time.  If your experience substantially differs then maybe we should figure out where the difference arises.

This is, in truth, one of my biggest concerns when I look at the direction I'm going.  I think that I've figured out multiple-conflict encounters, but I haven't worked out the (rather huge) kinks for individual actions that aren't directly tied to time.

For instance, if a character is attempting to pick a lock, what's to stop the player from spending all of her character's AP on that action, knowing that they will immediately refresh?  

Fixing this problem might simply be a matter of offering a set range of difficulties for the GM to use as a guideline when creating static, one-shot encounters (picking a lock, searching for a secret passage), coupled with partial AP degradation with a slower refresh rate for the lost AP.

For instance, the GM has set a lock at Formidable (20).  In order to open that lock, Hythera will have to use her Security Systems (6) skill, and pay at least 14AP.  Of course, Hythera doesn't know how much it's going to cost to open the lock, so she goes to the extreme: she spends 20AP for a total of 26.  She opens the lock, and since there are no other AP-draining activites going on, her AP pool refreshes... shy three points.  Those three points will refresh after ten minutes (in-game time) of not doing anything with her Action Points.

After opening the door, she slips inside and starts getting to her work.  She's been hired to hack into a shipping corporation's computers and copy their requisition spreadsheets.  To do this, she'll have to use her Hacking (7) skill, against a Challenging (10) computer system.  She spends 5AP, bringing her total to 12, which makes her successful.  Since there are no more AP-draining encounters going on, her AP pool refreshes, minus another 3AP, meaning that she's now down a total of 6AP from her original total of 32.  She has only 26AP from which to perform future actions, and must wait 10 minutes per 3AP for the drained AP to refresh.

Unfortunately for her, a discreet security camera in the corner of the room caught her entering, and the corporation's security forces arrive just as she's finishing up her hack job.  She's either got to submit to them, or fight her way out... and she's not in the mood to be arrested.  She enters the combat with 26AP.

Combat could work much the same way; each round, the Action Points pool refreshes, sans three points.  Once you've gone through enough encounters to run out of Action Points, you can only perform actions at your base skill -1.

Hrm.

Doesn't completely get rid of GM fiat... but it's a start.

[Ben Lehman]

Chris and Simon -- Welcome to the Forge! (Can't believe I forgot that.)

This stuff about GM fiats and double-blinds is all very good.  Just to summarize, the double-blind problem can be solved by one or more of the following --

1) Absolute scales for the abilities (so we know that a four is enough to climb a slick, steep wall, say)

2) Transparency in base levels + randomization by hidden bidding.  In other words, you have 10 points in security, and the lock is 7 points, and we both know this going into point bidding  (This is Cradle's solution)

3) GM bids first, announces total -- NeelK's solution.  This looks like "cheating" but it isn't.  It simply allows for informed decision making.

4) A double-blind.  I think that we all agree that the double-blind breaks down in practice, because the GM will know the player's stats.  There are solutions to this, but they are unwieldy.

Also, though, I want to talk about refresh times.

The thing is that, outside of a combat or combat-like situation (where you have multiple important actions in a short period of time), your AP bidding system breaks down.  Say that I need to climb a wall.  There is no reason for me to not wait a turn, accumulate 1.5 times my AP, and then just spend it all the climb the wall.  Suddenly, the game becomes deterministic.  Whoa!  We didn't account for that.

Different systems handle this in different ways.  Cradle ties point refresh to either scenes or the challenges themselves, but only gives you a little bit.  Nobilis ties refresh to dramatic confrontation.  I believe Active Exploits does it by session.  The thing that all of these systems have in common is that, no matter when you spend the points, they are a real and genuine cost to them.

Right now, outside of combat or combat-like situations, your system doesn't.  You might want to think about this.

Also, I'd like to talk about statting of things.

It is poor design to assume that everything the game world will be statted.  Unless the game is heavily rail-roaded, and even then, the PCs will do things that you didn't expect them to -- attempt to blow open the door rather than bypass the lock, say, or attempt to sneak into the castle rather than weaseling an invitation from the steward.  The thing is that you cannot account for all of these things.  Ever.  It is just impossible.  So you are going, as a GM, to be doing some statting on the fly.  Most likely a lot of statting on the fly.

And you don't have fortune to fall back on.

Solving this problem is unique and interesting.

yrs--
--Ben

[dredd_funk]

Simon - I like the mechanic, btw!  The initiative system produces benefits for both the winner and loser, which is cool.  Even though the comments here have some good ideas for ironing out the kinks, don't get discouraged!!

Tony - yep, I totally see where you're coming from and I hadn't really thought about that (thanks!)...though taking the concept of fiat to that level, while productive in some sense, makes me think that 'GM fiat' might pertain to every decision a GM makes if it doesn't refer to an immediate, 'objective', external authority.  I've got more to say but I want to make sure that this thread stays on topic and addresses Simon's needs.  It's his thread.

Ben - thanks for the welcome!  The game-design bug hit but, unfortunately for me, it hit someone that hasn't played in a long, long time.  It hit one of those people that merely was going to take advantage of the internet's ability to start playing again with a few old friends.  The utter nostalgia of it all made me want to design (frankly, the design bug wouldn't have hit had I found the Forge first and gotten access to a lot of the designs here--it's easy to see how much thought, work and effort has been placed into many of them and realize, "well, let's be honest, I'll never do something that good.").  For good or ill, however, I discovered the Forge after the bug hit, so I'll just have to keep going.

I'd love to respond to the issues both you and Tony have raised but--as I said above--this may not be the best place.  I don't want to get in the way of the advice/help you can give Simon.  When the time comes, I can ask for advice and clarification in a separate thread.

Cheers.
Chris

[M. J. Young]

I've got a couple ideas reading this thread over the past couple days, which I don't think are precisely duplicated.

The game uses stat plus points to determine outcomes. If we assume that the stats are set in ways that make sense--that a man isn't going to have the strength of a horse, for example, unless he's clearly a very powerful man--then the players have a rough guess at the stat of their adversary. The referee, of course, knows the stats of the player characters, and the number of points they have available. If the referee goes first, stating the adversary's expenditure, the players can make a stab at how many points they need to spend to oppose that. The referee informs them whether they are successful or not, and from this they work back to hone their estimation of the abilities of this adversary (in some ways rather realistic). The longer they engage the opponent, the more accurate is their assessment of his abilities.

As far as inanimate obstacles, there seems to be a notion floating around that if I see a lock I have no idea whether it has a value of three or thirty. As someone who knows a few locksmiths, this makes no sense to me. Let me suggest that anyone can force a lock if he spends enough action points to do so, and the easier the lock the fewer points are needed to force it. Someone with skill at locks has the advantage not only that he gets a few points lead, but that he can look at a lock and say, "Oh, that's a Skillcraft double-tumbler security special; that's something like a fifteen to twenty point lock" versus "I've seen tougher locks on luggage, and this will be automatic for a person of my skill." That is, the skill isn't only the ability to perform the task; it's also the ability to accurately estimate the difficulty involved in doing so.

Hope that helps.