Tactics: Antagonist Bidding

[Ben Lehman]

Previous Threads on this game:
http://www.indie-rpgs.com/viewtopic.php?t=8551
http://www.indie-rpgs.com/viewtopic.php?t=9041

So, after some long agonizing over the virtues of varying levels of "GMless" play, I have decided the following things:

1) If it is possible to have non antagonistic tactical play, it's hard enough that I don't want to design it.
2) The most important thing for Tactics, as a design, is egalitarianism and non-centralized play (no one GM, can play with players missing), rather than strictly non-competitive play.

To that end, I've come up with this system bit, and am fishing for opinions and advice about it.  Keep in mind that Tactics is designed to be High Step on Up, reasonably low Competition Gamism, and also to have *very* clearly defined rules.  Like automative manual clear.

In Some Manner (see: Results, below) the following things are set up:
1) a situation
2) a set of maps corresponding to that situation, including areas of control.
3) a set of goals corresponding to that situation (ideally after everything else)
4) any neutrals that may be involved in that situation
5) any special rules that may be involved in that situations (reinforcements show up in 10 rounds, say...)

Choose Characters:

Everyone except the loser of the last session chooses one character that they will control for this situation, in this order:
Winner of last session
Host (if not loser of last session)
Scenario Designer (if not loser of last session)
Any "Survivors" of last session
Any "Escapees" of last session
Any "Casualties" of last session
Loser of last session

Ties are resolved via one dice roll off.

Bidding:
Then, going around the table starting to the left of the host, the players "bid up" what sort of opposition forces they think that the party can defeat.  If a player does not think that the party can defeat those forces, they can "call the bid."  They take on the role of the antagonist for this scenario.

Option: Each "basic unit" has a Difficulty Number.  If you like, you can bid Difficulty Numbers (three fives and a two) rather than units, allowing the antagonist to choose his troops.

Option: As long as the bid is sizably "up" the bidder can readjust the units (we won't take the student, but we will take two ninjas.)

Equip:
The loser of the last session, if not the antagonist, takes on the character he chose.  Each player chooses abilities for her character.

Play:  Commences as per rules, with antagonist controlling all the enemies.

Winning:

If the Party Wins:  The last person who "bid up" has won the day (along with everyone else.)

If the Enemies Win: (prevents the party from achieving its goals), the Antagonist has won.

A Marginal Victory, where some goals were accomplished, but not all, is considered a win for the Antagonist, although the consequences for the Party and overall storyline may be less dire.

Forfeits:
Any number of forfeits may be decided upon by the group.
1) Loser provides (or at least pays for) food and drink for next session.
2) Winner designates someone to establish next situation.
3) Winner chooses one troop, who gets two bonus XP dice and a bonus roll (particularly good for the antagonist.)
4) Winner gets to set up a "cut scene" (assign parts) to be played out between games, either between party members or important NPCs.  The winner's control over the scene can be anything from full Narration to "establish the end goals, and let loose" to "set up the parts, and see what happens" in which case it is nearly a new Situation.

Questions:
Should there be some other way that the maps are designed?
What needs clearer definition (this is a pretty vague piece, all toled)?
Should some bidding system other than "go to the left" be designed?  I worry that people might frequently get left out of the bidding.
Should the forfeits be more clearly defined?
Any other concerns?

Thanks for your comments.

yrs--
--Ben

[Mike Holmes]

Questions:
Should there be some other way that the maps are designed?

Other than what? I couldn't find how the maps were designed.

What needs clearer definition (this is a pretty vague piece, all toled)?

Yep, pretty much everything. Fortunately some of that can wait until it's tested.

Should some bidding system other than "go to the left" be designed?  I worry that people might frequently get left out of the bidding.

Huh, the "Name That Tune" method. I think it is problematic in this case. Because the difficulty of the task is known. If you're not familiar, the TV show Name That Tune involved players bidding down the number of notes that they could name a particular tune in (often the bidding would get down to as low as four or even less). The thing is that the players didn't know what the tune was, so they had no idea how hard it would be to name the tune in X notes.

In this game, the player will know pretty well after a while, the danger present. So they can pretty accurately gauge the value of their bid. So, what's their motive? If it's beneficial to be the Antagonist, and will proceed with that assumption (because if it's not, then nobody will ever call). If they can call their own bid, the first bidder will just go straight for what they think they can take out on the first bid, and call themselves, so I'll assume that this is also not the case. If they can't call themselves, but anyone else can call any bid, then they'll bid just below what they think they can do so that they can call the next bid (assuming it has to be higher). So I'll work on the assumption that only the bidder to the left can call.

In that case, the the bidder tries to set the bid so that it will reach the appropriate level on reaching the bid of the player on their right. Meaning, in practice that every player is doing this and so they'll upbid as little as possible so that it is they who end up in the catbird seat. If there are discrete units of upbid, then a player will be able to determine just exactly what to bid so that his desired bid comes around to him. In which case the only tactics of the others are to either upbid more in order to thwart the first bidder by giving it to somebody earlier, or to take the bid at a level lower than the first bidder thought would be the right bid.

That makes for very slight tactics. If the upbid units are small, then maybe you'll get players calling "early" - but this will tend to hose the player on the first bidders left badly because he'll never see a reasonable bid. If you have Large units of bid, then I think it all would be pointless - the first bidder would just get their planned bid unless "thwarted" as I mention above. Not much going on there.

If the player can make any bid, not neccessarily one that's an upbid, then you'll see players tempting the player on their left with as low a bid as they think that the player will bite on. This is fine, except that it may go on indefinitely as there's pressure on the bidder to keep the bid low enough that they can win. It also means that since there's no escallation that the bid can be called at any time including the first. If all players can call, then this isn't problematic particularly.

So none of these options are striking me as particluarly good. It all hinges on the fact that you're bidding a known quantity. In markets bidding is based around the gamble that you don't know the sale price at a later date, and can only estimate. Yes, if the game has dice, then it's true that there's some uncertainty, but typically it's minimalized in terms of analysis by the fact that many rolls have to be made. Hence expected value analysis tends to be fairly effective in determining the value of an opponent.

If you allow "recombination" of elements at each bid (that is, if I bid one orc, you can bid two goblins instead of having to include my orc), then there's at least the challenge of quickly estimating the force that each particular creature brings. Still, good players will be able to do this quickly.

So the solution seems to me to be a matter of somehow making the opposition at least partially unknown. This can't just be randomization, because that would be similarly irrellevant to the way that dice are (players will just bid on what they know). There has to be some skill in predicting the result, meaning that there has to be partial data.

Here's what's coming to me as an example. Each character draws X number of cards (or maybe the whole deck). Then "bids" are a matter of putting one (or maybe more, but limited in any case) card face down into a blind in the middle. The cards have creatures or other forms of opposition on them. When someone calls, that player gets the blind to work with as the opposition. (A neat side effect of this is that if you don't allow "table talk" making players keep their donations secret then the opposition will be a surprise to every player to some extent).  

That may be too random - with large groups you'd know so little of the composition that your guess would be just a vague stab. So what I'm thinking is that you make it so that only the player to the left can call (which is good because that's rigorous - he either calls or bids and then we know he's done), and then you have bidding players show their cards to the players on their right. Probably half round down or something. That way you know some reasonable percentage of the cards going in, but have to guestimate on the rest.

I think something like that would lead to a very fun bidding process full of strategy and deception. Was that three kobolds, or three giants that just went in? Can you make it if it turns out to be three kobolds given what you know of the rest of the blind?

Should the forfeits be more clearly defined?

Very much so, most importantly on how they're determined.

Mike

[Ben Lehman]

Thanks very much for your response!

What needs clearer definition (this is a pretty vague piece, all toled)?

Yep, pretty much everything. Fortunately some of that can wait until it's tested.

BL>  To some degree.  I feel that offering playtesters a method of scenario design and role distribution which is not hand-wavy "GM vs. Players with no-rules challenge" is very key to the basic goals of Tactics.

 Of course, I'm also trying to nail my core rules before I hit setting, and both are absolutely necessary for playtest.

Should some bidding system other than "go to the left" be designed?  I worry that people might frequently get left out of the bidding.

In this game, the player will know pretty well after a while, the danger present. So they can pretty accurately gauge the value of their bid. So, what's their motive? If it's beneficial to be the Antagonist, and will proceed with that assumption (because if it's not, then nobody will ever call). If they can call their own bid, the first bidder will just go straight for what they think they can take out on the first bid, and call themselves, so I'll assume that this is also not the case. If they can't call themselves, but anyone else can call any bid, then they'll bid just below what they think they can do so that they can call the next bid (assuming it has to be higher). So I'll work on the assumption that only the bidder to the left can call.

BL> You would be correct in this assumption (only the next player can "raise" or call.)  But you are very incorrect in your other assumption.

Assuming the goal is the be victor of the scenario (note that things are a lot more complex and, I would say, the overall play-goal is most likely a "good game.")  So, therefore, it is best to be the antagonist if and only if the opposing forces outweigh the PC forces.  If the PCs outnumber the antagonistic forces, it's much better to be the "last raise," which I'm thinking of renaming the "captian" and giving certain command and coordination duties among the PCs.

So essentially we have people going around the table saying "we can take out 4 Soldiers and an Artillerist," "We can take that out plus two Students," "We can take that out plus a Ninja!"  It's a big old boasting contest, until someone calls "bullshit" and then the group must step up to take out the forces last bid.

Interesting thought:  What if the "captain" had no PC, and could only advise?  I'd worry that this would be boring in actual play, but it is an interesting scenario.

The goal of this system bit is to set up scenarios which are reasonably "balanced," by which I mean provide interesting and difficult challenge for all the play-group.

So the solution seems to me to be a matter of somehow making the opposition at least partially unknown. This can't just be randomization, because that would be similarly irrellevant to the way that dice are (players will just bid on what they know). There has to be some skill in predicting the result, meaning that there has to be partial data.

Here's what's coming to me as an example. Each character draws X number of cards (or maybe the whole deck). Then "bids" are a matter of putting one (or maybe more, but limited in any case) card face down into a blind in the middle. The cards have creatures or other forms of opposition on them. When someone calls, that player gets the blind to work with as the opposition. (A neat side effect of this is that if you don't allow "table talk" making players keep their donations secret then the opposition will be a surprise to every player to some extent).  

That may be too random - with large groups you'd know so little of the composition that your guess would be just a vague stab. So what I'm thinking is that you make it so that only the player to the left can call (which is good because that's rigorous - he either calls or bids and then we know he's done), and then you have bidding players show their cards to the players on their right. Probably half round down or something. That way you know some reasonable percentage of the cards going in, but have to guestimate on the rest.

I think something like that would lead to a very fun bidding process full of strategy and deception. Was that three kobolds, or three giants that just went in? Can you make it if it turns out to be three kobolds given what you know of the rest of the blind?

BL>  The difficult with any sort of hidden information, to my eyes, is that it works against the essential goal of this system bit:  It is not necessarily intended to be a fun game in and of itself (although that would be a bonus), but rather provide and design a fun game for the remainder of an evening's play.  I forsee a very small amount of time actually being used with this system, and a very large amount of time being spent playing out the actual scenario.
 If the scenario presented is vastly underpowered or overpowered (and, given the card system you mention, it would end up that way with unpleasant frequency) then the entire resulting game is *no fun at all*  This is a bad, bad outcome, because it means wasted hours or bypassing the mission design system, both of which are failures.
 Your point about bidding the known quantity is well taken.  I'm just trying to think about how to work it out without too much rejiggering.

Here's a thought -- NPC characters have a "threat value" of 1-6.  Everyone, at the beginning of bidding, rolls a certain number of d6, and can only bid up by those units.  You can add as many dice as you want, but you can't take any out.

Should the forfeits be more clearly defined?

Very much so, most importantly on how they're determined.

BL>  Yes.  I think it's best to hardwire certain forfeits into the rules, and have other ones added on as "optional."

yrs--
--Ben

[DevP]

Each character draws X number of cards...

Incidentally, maybe this how to integrate Texas Hold-'em into the RPG. ("2 Students and a Goblin on the Flop, Ninja on the River...") But I digress.

So essentially we have people going around the table saying "we can take out 4 Soldiers and an Artillerist," "We can take that out plus two Students," "We can take that out plus a Ninja!" It's a big old boasting contest, until someone calls "bullshit" and then the group must step up to take out the forces last bid.

Whereas I see the point, this sort of negotiating is very, very intuitive, and thus advantageous. (Incidentally, I'd say switching of elements, i.e. 1 Ninja for 4 Pirates, can be done via consensus, but would not count as an upbid; or maybe subsitution must be itself an upbid, again approved by consensus.)

The random element is helpful; how about a blind bid of SITUATIONAL elements? There is a deck of scenio affects (either generated by the players or perhaps pre-defined by the system), and that this is part of the bidding; each Condition is a negative for the players, and the bid can involve blindly adding more Conditions to the fight. These condition will only be revealed at the start of play. So the key currency of the fight - the opposition - is negotiatied, and while this auction has picked a near equal point for Antagonist and the PCs, the Conditions are an unknowable edge to the bid.

A side-question, Ben: are the Antagonist-play and Protangonist-play different enough that the antagonist isn't just juggling extra characters? (I'd understand if the Protagonists had lots of character elements to min-max and track, while Antagonist-elements are essentially simplistic mooks.)

[Ben Lehman]

So essentially we have people going around the table saying "we can take out 4 Soldiers and an Artillerist," "We can take that out plus two Students," "We can take that out plus a Ninja!" It's a big old boasting contest, until someone calls "bullshit" and then the group must step up to take out the forces last bid.

Whereas I see the point, this sort of negotiating is very, very intuitive, and thus advantageous. (Incidentally, I'd say switching of elements, i.e. 1 Ninja for 4 Pirates, can be done via consensus, but would not count as an upbid; or maybe subsitution must be itself an upbid, again approved by consensus.)

BL>  Okay, time for more textual clarification -- you can only "upbid" or "call."  So, in that case, all rearrangements would have to be upbids :-)

The random element is helpful; how about a blind bid of SITUATIONAL elements? There is a deck of scenio affects (either generated by the players or perhaps pre-defined by the system), and that this is part of the bidding; each Condition is a negative for the players, and the bid can involve blindly adding more Conditions to the fight. These condition will only be revealed at the start of play. So the key currency of the fight - the opposition - is negotiatied, and while this auction has picked a near equal point for Antagonist and the PCs, the Conditions are an unknowable edge to the bid.

BL>  To some degree, I fail to see how this is helpful.  It doesn't seem to add step on up possibilities, and it adds a very high chance of "oh, that was too easy" or "oops, that was too hard."  Which is disadvantageous.

A side-question, Ben: are the Antagonist-play and Protangonist-play different enough that the antagonist isn't just juggling extra characters? (I'd understand if the Protagonists had lots of character elements to min-max and track, while Antagonist-elements are essentially simplistic mooks.)

BL>  That is a good question.

In short -- at present, no.  Antagonist NPCs are defined no differently than player's PCs.  Hmm...

yrs--
--Ben

[Mike Holmes]

Assuming the goal is the be victor of the scenario (note that things are a lot more complex and, I would say, the overall play-goal is most likely a "good game.")  So, therefore, it is best to be the antagonist if and only if the opposing forces outweigh the PC forces.  If the PCs outnumber the antagonistic forces, it's much better to be the "last raise," which I'm thinking of renaming the "captian" and giving certain command and coordination duties among the PCs.

Did I give the impression that I didn't understand this? Because I did, and this is precisely what my post is talking about throughout. I get that it's about establishing a balanced scenario. My points are all predicated on the idea that at some point the bid will be "juicy" enough for the next guy to call. The thing is that there are still game elements, and I'm applying Game Theory to tell you how it's going to turn out.

All the problems that I pointed out stemming from this are still there. Try a test and you'll see what I mean. You intuitively saw that the bidding will "instantly" get to the equilibrium point, and be bid. Now that you point out the "captain" thing, most often the first bidder will be the captain, and the guy to his left will be the opposition. Followed second by the next pair. Later players won't get to bid. The reason for this is that there's no reason to "underbid." In most bidding games, you start out low so that you might catch the opponent off guard. In this case, the opponent can easily determine his opposition, and so will be able to accurately judge whether or not to take the bid. The only "skill" will be the simple data analysis that applies to each opponent. After a while a shorthand will appear, and this will become very simple.

BL>  The difficult with any sort of hidden information, to my eyes, is that it works against the essential goal of this system bit:  It is not necessarily intended to be a fun game in and of itself (although that would be a bonus), but rather provide and design a fun game for the remainder of an evening's play.

Then you ought to simply determine the opposition at random. Drop the game elements entirely if you think they're going to mess with creating balance.

This is boring, however - you end up with the Warhammer point system, essentially.

I forsee a very small amount of time actually being used with this system, and a very large amount of time being spent playing out the actual scenario.

I think that the method that I enumerated would take about five minutes. Probably longer than hashing things out "by the numbers". Because estimation is required.

If the scenario presented is vastly underpowered or overpowered (and, given the card system you mention, it would end up that way with unpleasant frequency) then the entire resulting game is *no fun at all*  This is a bad, bad outcome, because it means wasted hours or bypassing the mission design system, both of which are failures.

I think you're not seeing it properly. If the variance on the cards is relatively small, then there's little likelihood that things will be out of whack. It would just need to be tuned. A big question is the one of "granularity," that you keep dodging. What can you "bid"? Do you have to bid discrete creatures? Or can I bid a 4th level Orc vs. a 10th level Orc? Armor? How detailed can the bid get? Do you have to keep all the elements already bid, or can you reorganize them? That is, if Player A bids 3 Orcs, can I bid 4 goblins? Or do I have to bid 3 Orcs and a goblin?

Once I have a better idea of how this would work, then I could propose a system that would result in some balance. In any case, you would accept that players are prone to making mistakes, and therefore misestimating the value of the opponents, right? Things aren't perfect, and the whole bidding thing implies that someone could bid too early or late. Meaning that the balance would be off anyhow. I posit that a system with unknown information can be made just as "balanced" in the outcome as one in which everything is known. In any case, you have to accept that occasionally things won't be precisely on. (Unless you want to just go to the Warhammer method).

Further, I'd posit that it's not fun to play "balanced" battles. These are neither realistic (who looks for a pitched battle), nor all that dramatic, and tend to be the longest battles possible. So I think that if you allowed a little more flex into the game in terms of what's encountered, that it would be more fun. "Balance" problems can be picked up in the metagame somewhere instead. Total Party Kill? Then maybe the last bidder loses more points or something. Too easy this time? Then the antagonist loses points. That's the risk you take calling. Fun gambling stuff.

Here's a thought -- NPC characters have a "threat value" of 1-6.  Everyone, at the beginning of bidding, rolls a certain number of d6, and can only bid up by those units.  You can add as many dice as you want, but you can't take any out.

That adds a skill, and speeds things up a little, but doesn't answer all the questions. Can I only bid one die at a time, or more? Can I see the other player's dice? What this would do is to create a strategy around trying to get the optimum to come to you as an individual. Which would be ancillary to the whole process. Might as well roll a die to see who gets to be the opposition. What it does promote is taking early as a gamble, which could be life-threatening to the PCs.

Worse, if you have "ratings" for things, unless the system is very simple, similarly rated things will have different advantages over different groups. Making the creature selection process another area of variance for the outcome (less balance) and making it likely that for a particular group of PCs that optimal opponents will appear. If these get selected repeatedly, that's dull. I really like the idea of bidding whole creatures.

BTW, what's a "win" for the PCs? That is, how often is one of the goals that all of the PCs remain alive? What happens if a PC dies? If PC survival is desired, then you'd want it to be a scenario goal. But if that's the case, then one of the best ways to win as the opposition is to pick on a particular PC until dead. Problem?

Also, do the PCs encounter alll of the opposition at once? If not, then how is the "staging" determined? Because it makes a huge difference to how effective the creatures are. I agree with Dev that situation should be part of the bidding. For example, lets say that the map is a dungeon. Then "bidding" should also involve which rooms the creatures are in. Traps should definitely be in the mix as well as chasms to jump, etc. These can be on the map to start as a sort of "handicap" (or it's opposite), but I think that making it part of the bidding is potentially more interesting. I mean, if the traps are hidden cleverly, and the party may not all be aware of their existence, then it adds that element back into the game.

Apollogies if you've answered this on a previous thread, but, have you seen Rune? What do you think of how it handles these things?

Interesting thought:  What if the "captain" had no PC, and could only advise?  I'd worry that this would be boring in actual play, but it is an interesting scenario.

This seems tangential to the discussion, but, I'll bite...why? What would be the point of this?

Mike