Good Gamist Deisgn

[GreatWolf]

Your last post is interesting.  Would a gamist game be more strategic and tactical if the GM also had only partial information about the characters?

Just a brief observation.  I don't think that full or partial information necessarily makes for a better Gamist game.  Chess has no hidden information, while Tigris & Euphrates has a much higher amount of hidden information.  Both are excellent strategic games.  

However, I do know that the question of information control does need to be considered and addressed when designing a Gamist RPG.  There's probably a lot of room for further work in this area, since most RPGs do not use or encourage hidden information.

Seth Ben-Ezra
Great Wolf

[xiombarg]

However, I do know that the question of information control does need to be considered and addressed when designing a Gamist RPG.  There's probably a lot of room for further work in this area, since most RPGs do not use or encourage hidden information.

As a possibly off-topic aside: Is this really your experience?

In my experience, I've been in a lot of Sim-heavy games where you're not supposed to know much OOC than you know IC about what's going on.

There's a lot of information-hiding in RPGs. It's just that it's very Sim-oriented, not Gamist-oriented. For example, in those same games I usually know the most optimal way to engage in combat.

[quozl]

In my experience, I've been in a lot of Sim-heavy games where you're not supposed to know much OOC than you know IC about what's going on.

But what about the other side?  Is the GM not supposed to know what the characters know?

[xiombarg]

But what about the other side?  Is the GM not supposed to know what the characters know?

I've seen this a lot, too. Ususually it's where the GM doesn't have any idea what the players are planning until they do it.

[Mike Holmes]

Oy, my head hurts.

Hidden information isn't the only way to make a good game, either, no. But it's one good potential option. To be really more precise, what's neccessary for a good game is that the player has to use some skill to discern between multiple strategies in order to win. That choice has to neither be too simple - if there's an easily discerned dominant strategy it's not much of a game - and not too hard. Because if its too hard, then players choose randomly instead of trying to figure out something they can't.

How to allow for those strategies to be present in sufficient quantity and quality is a matter of determining the cross product of the maneuvers available in a "move" and how difficult it is to evaluate each maneuver relative to each other. This difficulty can be created through lots of information, or through information which is known to exist, but the nature of which is unknown. Or even the potential that something comepletely unknown exists.

You can't make any statements about what's optimal here, because it's all on a case by case basis. As I said, you can actually overdo it.

TROS does have hidden information as a potentially important part of strategy. You don't know how many dice your opponent has, neccessarily, and he doesn't have to use them all. Perfect example.

It may turn out that TROS is a RPS sort of game in the end. But to it's credit, none of the many players who play it have discovered what the "cycle" is yet. As such, it's still not a "known" game, and nobody knows a best strategy. Yet people can still point out tactics. This means that TROS is at the optimal level for a game in terms of strategy in general terms. There are gradations of optimal, however. But compared to, say, D&D, which is almost completely a "known" game, it's way ahead of the curve.

Mike

[quozl]

Oy, my head hurts.

Mike

Mine too.  It's hard to pin down what strategy and tactics are and not just say "I know it when I see it."  For example, is Dominoes more or less strategic if played with all player's tiles face-up?  I don't know and my head hurts thinking about it too long.

But let's get specific now, specificially RPG physical combat systems.

In your opinion, does TROS offer the best strategic and tactical physical comabt system for an RPG?  What parts could improve?  Are those parts done better by another RPG?

[epweissengruber]

So is it as I feared that there are no rpg rules that implement strategy and tactics in play?

Most RPG rules are heavy on boring strategy: chosing powers and classes that will ensure long term success.  I dropped out of the hobby because no one executed fun or interesting tactics during play.  We sat around rolling dice in a desultory fashion and making Monty Python references.  Those games that facilitate in-play tactics and concentrated fun are Sorcerer and Hero Quest.  

I will give a shout out to Deadlands.  The system is inelegant and clunky, but I loved the interplay of strategy-influencing chance (The odds are high I will get 3 chances to act) and tactical response to the outcomes of chance (Only got 1 card -- can't fire off this clunky old revolver this round).  All the special options for duels made for some nail-bitingly tense play.

[Mike Holmes]

It's hard to pin down what strategy and tactics are and not just say "I know it when I see it."  For example, is Dominoes more or less strategic if played with all player's tiles face-up?  I don't know and my head hurts thinking about it too long.

But, see, there's this whole field of mathematics that does exactly this called, you guessed it, Game Theory. This is why your head hurts, you're trying to intuitively determine the principles that people like Von Neumann (also invented the modern computer) and John Nash (of Beautiful Mind fame) came up with in the middle of the last century. If you're really, really interested in this, do your reading there. I can only claim to have a cursory knowledge of how it works, but I know enough to say that there's a body of work there that at the very least makes all this much easier to discuss. Oh, if only I could use a phrase like "minmaxed mixed strategy equilibrium" and have people get what I was saying. :-)

In your opinion, does TROS offer the best strategic and tactical physical comabt system for an RPG?  What parts could improve?  Are those parts done better by another RPG?

Best? You're kidding, right? Without criteria? Let's say you've got some players who really want a slight challenge, but mostly easy recreation - what some people would call the Beer & Pretzels level of challenge. Then you have another group who, tired of playing Go, want to find an equal or more challenging game. What's best for one is not best for the other group.

TROS provides a level of challenge that'll probably get your character killed with a Beer & Pretzels attitude eventually. So it's somewhere in between. A nice happy medium. Does that mean that I think that more challenging games shouldn't be created? Nope. But I leave the specific level of challenge to the designer.

I'm going to get into some Game Theory here. I have to assume that people don't understand the terms, so if you do, please understand that I'm not trying to talk down to anyone. I'm using the simplest and most obvious examples just to be clear, not to insult anyone's intelligence.

What is important is that, if the Game is really to have some sort of Gamist challenge, that no "dominant" strategy be readily apparent from the get-go. That is, at the very least, players should have to consider their current positions at all times to do well. A dominant strategy, as you might guess, is one that gives a better result than any other (strategy, again, being a plan of moves). If you've played tic-tac-toe as an adult, then you're probably aware that there's a dominant strategy about where to place your first X, and where to counter with the first O. With both players playing these strategies, nobody can win (if you believe otherwise, then you haven't played enough with players really trying to win).

(For those who've seen the movie Wargames, this is the point of the movie. It's essentially a parable about the work actually done by people like John Nash in game theory at the Rand thinktank in regards to nuclear war.)

In fact, I'd go so far as to say that, if possible without exceeding your Beer/Go level of intensity, that the game should have no easily discerned dominant strategy at any particular moment. That is, there may be one, potentially, but the player should at least have to think a little to come up with it. And given faulty analysis, they should fail.

A possibly better goal, is not to have dominant strategies available at all, and instead have a game where Nash Equilibria of some sort exist. That means that the player may not be able to optimize things perfectly to his advantage, but he can make the best with what he has in some fashion that allows that the opponent may also be reaching some level of benefit (that's a poor explanation, but it's a difficult concept - Nash one the Nobel for it, remember).

Thus, for instance, if the game only considers weapon damage, then, all else being equal, you select the weapon that does the most damage. This is such an obvious case of a dominant strategy that games go to all manner of lengths to mitigate this sort of thing. The problem is that they only muddle the analysis a little in most cases, and in the end there's just a harder to discern dominant strategy. And further, they usually aren't affected by tactical considerations later on. Thus, a player can determine the best weapon at the outset. For example, often weapon weight comes into play in theory as a potential problem for encumberance. But usually the player quickly discerns that the more potent weapon's extra weight easily is worth the few coins that it means he will not be able to carry out of the dungeon. Weapon cost is similar (lots of games will have exponentially more expensive "fine" weapons, as slightly more of a challenge).

The typical example of a solvable, but potentially difficult weapon stat to minmax (minmax is a technical term, which means to determine the dominant stratgy through number crunching, sorta), is weapon speed. The player has to consider what the deleterious effects of a slower weapon are in terms of the comparison to the damage produced (usually these are made inversely proportional in some way). These often result in some calculation of damage-per-unit-time or somesuch that often involve character statistics (and as such may involve character minmaxing as part of the overall strategy: e.g. the Shortbow archer phenomenon in TFT).

This sort of analysis is still all up front, however (in some really complicated cases, the player has to account for the potential loss of damage producing capability over time due to wounding effects, but even that's a predictive chore). What would make weapons of more interest is if there were different options in their use that made difference given the tactical situation. For instance, half-swording in TROS. The same sort of minmax analysis occurs on the spot in TROS when a character decides in which manner to use the sword against an armored foe. Is the gain in penatration ability worth the other penalties that come with the option? Not a hard decision by itself, but combined with the other tactical considerations in TROS, the weapon with this option does provide for more tactical options, and hence contributes to why, at least till now, TROS has not become a "known" game.

A known game is one in which a solution as to the dominant strategy, or equilibria have been determined (and are presumably known to the players).

In fact, in TROS, there are likely no dominant strategies. Even given statistical analysis of the outcomes, the player still has to guess to an extent what the opponent will do, hence making the first choice difficult. This is wonderful, BTW, because it makes the Master Sensei Samurai phenomenon true - often the first to attack will lose. AFAICT, the closest thing to a dominant strategy in TROS starts with goading your opponent into attacking first. But even that's dubious from practical results I've seen.

So, now is my appreciation for TROS more clear? I think that there's a lot more excitement in Gamist RPGs if the player decisions (Step On Up, as Ron would put it) are tactically fateful, meaning that they have to be determined on the spot, and not merely be the result of some overall winning strategy. Otherwise combat becomes just going through the motions to do the best that you're aware is possible. That's not to say that strategic elements should be ignored - that's a valid choice, but it's refusing to use an entire second set of potential decision-making possibilities. Just that, often, the tactical elements are irrelevant in light of the strategic in many games.

Are we getting anywhere?

Note that dice rolls are a godsend for these purposes. Because any such decision to be made becomes an order of magnitude more difficult to analyze with randomizers. Because instead of looking at one particular value for a decision, randomizers mean that you have to look at potentially many statistical outcomes. What matters more, the mean of the weapon's damage, or the range (min and max), or some other statistic like the standard deviation for many dice rolled together?

Also, there's that whole "odds transparency" issue. All odds are calculable, but some methods require such problematic math that (unless we're those bored Go players), people are just not likely to do the math. Instead, they'll estimate, which itself becomes a good Gamist skill. OTOH, if estimation becomes impossible, then the dice fall out of the equation in terms of what players will do to determine best tactics. For those who like to be able to use the odds as part of the determination of strategy, they should be more estimable, or even calculable.

This, then is where aesthetic issues come into play. If a designer feels that they want to emulate the character experience more, he may decide that having easily calculable odds is bad for that feel. So he may reasonably disclude them. The point is that, again, what's "best" for a game isn't easily decided. If the game becomes so divorced from feel that the player doesn't think in terms of character and situation any longer, but rather in terms of numbers only, then is it a RPG any longer?

So, this is just the surface of this issue. I could go on at length. What I hope to show is why there's no simple way to discuss this subject. What should be done is to disect things into parts and to look at them in digestible chunks. Beyond that, the only thing that I can say is that one should understand the principles of Game Theory so that you can attack the specific problems that occur. Interestingly, Game Theory really can't address something as complicated as an RPG overall in one swoop. The best it can do is to break it down into smaller parts to look at each one so that they can be assembled back together again (where another overall round can occur). So even the theory can't give you a direct answer, if you see what I mean.

Mike

[Mike Holmes]

Following on the above...

Those games that facilitate in-play tactics and concentrated fun are Sorcerer and Hero Quest.

I disagree strongly. Rather, I think we must be talking about different things. Because the only strategy in either of these games is to convince the GM to give you more bonus dice or augments. That's a simple dominant strategy.

Don't get me wrong, these are amongst my favorite games. They just have very little in the way of tactical decision making in terms of Gamism. What they do have is decision making in terms of Narrativism. For example, I have Love's Alana 7w, but Hate's Alana's Father 10w. Same augment from each, but I have to ask myself, is it worth sacrificing my love for Alana to kill her father? Great Narrativism support, almost no Gamist support. The book even enumerates the dominant bidding strategies so that you don't have to come up with them on your own.

Mike

[GreatWolf]

I don't remember if I've posted this link before, but I think that Shannon Appelcline's Thinking Virtually column is quite helpful in dealing with these issues.  Start with column #56 (Strategic Introductions).  He is discussing boardgame design (with a specific eye towards web-based "board" games), but the principles extend straight into good Gamist design.

And if I'm repeating myself, I apologize.

Seth Ben-Ezra
Great Wolf

[quozl]

But, see, there's this whole field of mathematics that does exactly this called, you guessed it, Game Theory. This is why your head hurts, you're trying to intuitively determine the principles that people like Von Neumann (also invented the modern computer) and John Nash (of Beautiful Mind fame) came up with in the middle of the last century. If you're really, really interested in this, do your reading there.
Mike

Thank you, Mike, that really helped.  I've been reading a little about game theory in the books I have about designing computer games but have a lot left to read.  Are there any books that you recommend for an introduction to game theory?

Also, if you're willing, I would love it if you analyzed part of an RPG with game theory in threads here on The Forge just like you did with weapon damage (and weight and cost, etc.)

Thanks again!

[Mike Holmes]

What game? And just as an exercise? I'd prefer to apply it to an actual design. I was tempted to get in on the "tactics" game discussion as that seemed appropriate.

Gads how I wish we had an expert in this here. I worry that I'm making myself look foolish by trying to teach something that I only dabble in myself. Which is my way of saying that I've learned everything I know of Game Theory from articles and the like. You can probably do as well with a search on the internet as with anything that I can suggest.

http://www.gametheory.net/

This site works as a decent primer, IME.

Mike

[quozl]

What game? And just as an exercise? I'd prefer to apply it to an actual design.

Mike

Please do.  I don't have a game in mind but any that analysis of any existing game should be a good start.  Now I need to start reading....

[xiombarg]

BTW, as a quick aside to this thread: As I mentioned elsewhere, old-school (little booklets) Traveller had a lot of rules for determining prices and general economics, making an excellent non-Combat Gamist minigame sort of thing.

[Mike Holmes]

Sorta. I'd say that, yes it's a non-combat mini-game. But I wouldn't say excellent. That is, it's a simple matter of looking at the map, and the world price modifiers to see where the profits are. Then it's just a matter of how much you're willing to gamble on the outcome die roll per commodity.

To whit, we played that game to death, some sessions being nothing but that. The result? Even with a cruddy tramp merchant, you can pay off your debts and become millionairs in very short order. Basically, in the context of the overall game, the "Starship Economics" game is too easy to win as written. Too few variables, too few risks, and little strategy besides "buy low, sell high".

So, I wouldn't say it's great. Not a bad start, however, and I really like the idea.

Mike