Skip to Content

Deterministic Vs Random models

29 replies [Last post]
Anonymous

First of all, I'd like to say that I'm a little sad to see this forum getting so little attention lately... The topics here are probably my favorite (and the most useful) on the whole board...

Anyways, I thought I might try and start a new topic here to get the place going again... I apologize if the topic's been covered already, but I didn't see it anyplace else...

So, the topic here is deterministic (non-random) versus random systems. I'm mainly thinking of combat systems here, but it could just as easily be applied to any type of opposed check where two or more people are competing for a desired outcome or goal (so feel free to discuss those, too!).

It seems like there's a few people who aren't familiar with the terminology (I myself didn't know a non-random system was called "deterministic" before I came here), so I thought it might be a good idea to go ahead and define what these two systems are, and how they differ (In my opinion, anyways).

Deterministic models are systems in which the outcome is based on non-random numbers, whether it be comparing unit strength versus defense, or one player's auction bid against another's. This outcome may or may not be known to the players as they are in the process of determining it... Typically in the "unit strength versus defense" example, both factors are known to all players, so one may make an informed decision on whether or not an attack will strengthen one's position (or weaken the opponent's). This is sometimes also true in the auction example, if the bids are out loud, but there may also be secret bids, where the outcome is still not random, but the result is not known until everyone reveals their bids... Another example of the "unknown outcome" would be players selecting cards from their hand and playing them simultaneously... The outcome is determined by who played the "better" card, but neither player will know if his card is best until all the cards are shown.

Random systems, on the other hand, make use of some type of random number generator, whether it be dice, a spinner, drawing a random card, or any other system where the players do not have direct control over the outcome. I don't think there's too much need for examples here, since these are the types of systems we are exposed to most often.

With all of that out of the way, the discussion I'm hoping to have is on the merits and faults of both of these systems, which will hopefully help others to choose which type of system will best suit their next game.

As for deterministic models, I tend to feel that they put more emphasis on tactics (again, I'm thinking in mostly a combat-oriented game)... If you see your opponent making Cavalry, you know you'd better make some Pikemen and put them in the right spot to intercept. However, they can also take away (to some extent) the ability of a player who is lagging behind to catch up (the lead players will typically have more resources, and can more effectively counter the lagging players without harming their own position much), since there would be no "lucky rolls" to turn the tide of the battle.

Random models tend to favor luck over skill, though they differ in the degree to which they favor it. Games like Risk (while fun to play with the right group) could just as easily be played by rolling a hundred dice and seeing who got the most 6's (again, with the right group, alliances and other diplomatic factors can add more depth to the play; Since they aren't considered part of the rules, however, I'm looking past them for this example). This is due to there being only one type of unit, and no situational factors altering the rolls. A slightly more tactical random model would be that of Axis & Allies. Dice are still rolled to determine the outcome of battle, but each unit has its own "to hit" number, and the defending side typically has a higher chance to hit. This can essentially be boiled down to allowing players to "purchase" better rolls (the cheapest unit, Infantry, must roll a 1 to hit on the attack, while a more expensive unit, Tanks, hit on a roll of 3 or less). This gives the game the added facet of choosing between quality and quantity. A player might be able to fill up the board with his Infantry, but he'd never be able to go anyplace without buying some of the more expensive units to help on the attack.

To summarize, I feel that deterministic models typically favor tactical thinking (whoever makes the "best" moves will win), and random models run the gamut from simple luck games to games that attempt to equalize the balance between luck and skill.

Thoughts?

sedjtroll
sedjtroll's picture
Offline
Joined: 07/21/2008
Deterministic Vs Random models

This is a good topic to discuss. In addition to combat systems, many games have many systems that are either deterministic or random. Take Resource Production for example...

In Puerto Rico, you can look at the board and see what everyone will produce if you choose Craftsman, and you decide to choose Craftsman or not.

In Settlers of Catan, you can look at the board and think about past die rolls to see what people might have in hand or what's likely to be produced, but you cannot guarantee production or know exactly who will have what on your next turn.

When it comes to combat, I think it's nice to have some combination of the two systems. In a completely deterministic combat system, you know (or can figure out) exactly what the outcome will be, and if it's not favorable then you simply don't attack. To me that has a sort of anti-climactic feeling, spending more time thinking about the possible outcome than actually producing the outcome. I think it can lead to Analysis Paralysis problems as well if there's enough to think about.

On the other hand, in a random combat system you can't reasonably tell if you have an advantage or not. Depending on how random the combat, the best you can hope for is to pile up the most bonuses or advantages the system offers you and then roll the dice to see what happens. Herein lies the meat of it, as far as I'm concerned- piling up the advantages.

If the system offers little or no way to get an advatage in combat AND a random combat resolution, then it's probably too based in luck for many strategy games. However if there are ways to improve your chances of winning before undertaking the attack, to outweigh the luck factor of the random element, then the game is about setting up for the random combat so you are more likely to come out on top.

A simple example of this is Evo, where fighting is a straight die roll with a target number. You can 'purchase' an effective +1 modifier by buying Horns for your dinos (one of the things that are auctioned off during the game). So you can make the decision to fight based on weather or not you have more horns than the other dino, but the outcome is determined by the roll of 1d6.

I have been unable to find a better way to handle combt in a game. I would like to think of a deterministic model which doesn't break down into AP, but I can't seem to think of one. Until then, I am stuck with the idea that there needs to be some random element, some element of risk involved in instigating a fight.

I suppose Tigris and Euphrates has a good combat mechanic which is mostly deterministic and only a little bit random... it involves board position (which you can see) and the addition of tiles from each player's hand, (which you can't see). You know that each player has 6 tiles, and that there are 4 different colors of tiles, but you don't know how many of any given color an opponent has. So there is a definite risk involved in fighting. This system works very well, even if it is disheartening to find out your opponent happened to have 6 of the same colored tile in hand, and it happened to be the color you were fighting over. And that's where the luck comes in- the tiles are drawn randomly.

Scurra and I agonized over some way to make the combat in All For One, our pickup/deliver game about the Three Musketeers, more exciting. The main thing though was that we didn't want the combat to take over the game, it's a small side thing and needs to have quick resolution. what we ended up doing to try and make it interesting but still based on player choice, as well as to keep all the players involved, was to make it a sort of voting system. In a nutshell, when there's a fight between 2 characters, the players vote for who will win. They vote with cards in hand, so they may not have a card to vote for a particular character, or they may not want to use it for that if they do. Also, players aren't tied to a particular character and there are reasons that any given player might want any given character to win a duel. You can see the details of this combat system in my Game Journal entitled All For One

I am relatively happy with the combat mechanic we came up with, but there have been complaints that it's not 'fun enough'... which brings up another question:

Which is more "fun," Deterministic or random combat? Can they both be made fun? How?

- Seth

DSfan
Offline
Joined: 12/31/1969
Deterministic Vs Random models

sedjtroll wrote:
Which is more "fun," Deterministic or random combat? Can they both be made fun? How?

I like both types of combat, it just depends in what type of game, and if it fits into the game.

Take Diplomacy for example. There is a small board of Europe, and not a lot of troops on the board because of it. This game works well with a Deterministic model because if this small board where to have a Random example, there would be to many guys, and not enough space to fit them all.

A random example is Risk (Another good game) It has a large board of the World divided into territories. Now if you were to add a Deterministic model to it... the whole game would fall apart. It would be to hard to count a bunch of troops fighting another group of troops, which would also allow you to cheat, by saying "How many troops you ask.... um... I got 30" when the person didn't even count.

My game "Complete Hazard" with the rules in a Journal Entry, Games in Progress, is using a both. I am trying out both to decide which one fits well with the game, and adds more fun. Right now I'am going with Card Battling, which empasizes some-what on both Random, and Deterministic.

For example, players draw 2 more cards each turn, adding a Random "What am I going to get?" kind of thing, the Deterministic parts comes in when battling. Each player sees what kind of units they have and what their "strength" is, thus adding cards to your "fighting hand".

-Justin

Anonymous
Deterministic Vs Random models

All right... I had to read through your post a few times to get everything straight, and there were a lot of good points in there...

sedjtroll wrote:
When it comes to combat, I think it's nice to have some combination of the two systems. In a completely deterministic combat system, you know (or can figure out) exactly what the outcome will be, and if it's not favorable then you simply don't attack. To me that has a sort of anti-climactic feeling, spending more time thinking about the possible outcome than actually producing the outcome. I think it can lead to Analysis Paralysis problems as well if there's enough to think about.

I can agree with that partially, depending on the game situation... With a deterministic system, if you look at it from a single turn perspective, I agree that it's easy enough to see who will win any particular encounter... However, the point in such a game then becomes to put yourself in a position where your opponents will be unable to launch a successful attack against you, while you are simultaneously able to wear down their position.

Additionally, there can always be a hidden (but non-random) element so the outcome can't be predicted by the attacker. In this case, players might have some type of reserve points (numbered cards, chits, things like that) that they can commit to the battle depending on how important that particular encounter is (this would be fairly similar to secret auctions). So, there's still an element of risk, but no randomness.

sedjtroll wrote:
Depending on how random the combat, the best you can hope for is to pile up the most bonuses or advantages the system offers you and then roll the dice to see what happens. Herein lies the meat of it, as far as I'm concerned- piling up the advantages.

If the system offers little or no way to get an advatage in combat AND a random combat resolution, then it's probably too based in luck for many strategy games.

This is exactly what I was trying to get across (and I hope I did) with the Risk vs Axis & Allies example... In Axis & Allies, you can spend more money to get a better chance at success, so the randomness of the rolls is tempered by the ability to put your money into fewer, but more successful, units.

sedjtroll'}but the outcome is determined by the roll of 1d6.[/quote wrote:

I thought of another thing while reading this part... A big part of how random dice-based combat is depends on what percentage of the overall total the dice plays... Basically, the smaller the maximum dice roll is compared to the maximum level of modifiers (a d6 where the modifiers can get to +10, for example) the more a player will have a feeling of being able to control a large part of the outcome, while still having a bit of the tenseness of a random outcome.

sedjtroll wrote:
I would like to think of a deterministic model which doesn't break down into AP, but I can't seem to think of one. Until then, I am stuck with the idea that there needs to be some random element, some element of risk involved in instigating a fight.

This again would go back to the hidden (but non-random) elements I mentioned before... In your Tigris and Euphrates example, players had hidden tiles (drawn randomly) that would affect the outcome of combat... Depending on the rest of the gameplay, a mechanic like this could work with non-random draws (for example, starting with a certain amount of each color (to stay with the T&E example) and committing a particular amount based on how important you feel the battle is).

sedjtroll
sedjtroll's picture
Offline
Joined: 07/21/2008
Deterministic Vs Random models

knic wrote:
Depending on the rest of the gameplay, a mechanic like this could work with non-random draws (for example, starting with a certain amount of each color (to stay with the T&E example) and committing a particular amount based on how important you feel the battle is).

Hmm... that might be interesting. Say instead of drawing tiles at random for T&E, you begin the game by dividing ALL the tiles evenly among the players. So all starting positions are equal and you will never draw any tiles. Then you play the game normally.

If you commit a lot of Red tiles to fights, you will start to run low. If you run low, it's suddenly dangerous to attack since you probably need to save tiles for defense. This goes for all the colors, but Red in particular because it counts for both internal and external conflicts (hence there are more of them as well).

That would really make you think about how much the combat is worth, and how many tiles to invest in it. You KNOW how many tiles an opponent has, or you can try and track it anyway.

That could indeed be very interesting, and is entirely deterministic. It reminds me of an auction mechanic in that you are sort of bidding on winning the conflict.

- Seth

Anonymous
Deterministic Vs Random models

Without going on in great detail as I'm currently on a time crunch for the night and want to catch up with a few threads before heading out, I would say that I am absolutely of the mind that I have seen enough "randomness" in games and that anything that comes out that is random will have to be absolutely phenomenal to get my buying dollar (or even to get me to open up the package if I'm given one / sent one).

Randomness has been done to death in sooooo many games over the last batch of decades that unless something is being done that no other system anywhere has done -- good luck! -- I think people are best off designing non-random games where decisions make the difference, NOT a roll of the dice or the spin of a wheel.

We have Trouble ... we have Monopoly ... we have Sorry ... we have Chutes 'n' Ladders ... and on and on.

My thought: enough with the random.

phpbbadmin
Offline
Joined: 04/23/2013
Deterministic Vs Random models

BarronVangorToth wrote:

Randomness has been done to death in sooooo many games over the last batch of decades that unless something is being done that no other system anywhere has done -- good luck! -- I think people are best off designing non-random games where decisions make the difference, NOT a roll of the dice or the spin of a wheel.

...

My thought: enough with the random.

Or the shuffle of a card? Be careful about your statement! Randomness comes in many different forms! Often times, depending upon card distribution, decks of cards are MUCH, MUCH more random than a six sided die. I'm going to play the devil's advocate here and ask you where would your CCG, Raw Deal, be without the randomization provided by the luck of the card draw? Sure, you can design your own deck, but there is still a lot of randomness in such a mechanic.

-Darke

Anonymous
Deterministic Vs Random models

I tend to agree with BarronVangorToth that for the moment, I'm more interested in deterministic systems than random... Just as he said, at least from a development standpoint, I'd be hard-pressed to make my combat system stand out from all the other dice-rollers...

However, I can see where the random systems are useful... I wasn't as kind to Risk as I was to Axis & Allies in my original post, but Risk was probably the first large scale game I ever played... It was easy enough to pick up when I was still in grade school, and simple enough to explain to people to get them to join a game with me. To this day, it gets pulled out every now and then when I'm in the mood for a lighter, quicker war game.

It's a lot easier to get a casual gamer to play a game that can be explained, essentially, by saying "The more units you have, the more dice you roll. If your dice beat the other guys, you win" than it is to get them to read an extensive rulebook and try and memorize unit stats and matchups.

It does seem, though, that some of the fun is lost by removing the random element... If everything turns into an equation to find the best strategy and units, then we may as well just be theorizing with rulebooks alone in our rooms... On the other hand, it can be said that the most random element in the game is the human element... Everyone brings a different viewpoint and strategy to the table, and even in a game where there is nothing random at all, you'll still have the chance to see wildly varying playstyles.

All in all, I suppose, what I'd like to see in a game is a solid deterministic base with a little randomness sprinkled on top for spice... I have a game, for example, where most elements are deterministic, but a few actions that can be taken have random factors in their resolution.

sedjtroll
sedjtroll's picture
Offline
Joined: 07/21/2008
Deterministic Vs Random models

knic wrote:
I'd be hard-pressed to make my combat system stand out from all the other dice-rollers...

jwarrend has endeavered to do just that, and I think his idea has a lot of merit. The defined portion in his thread is the deterministic portion of the mechanic. The random portion would lie in the cards or whatever is used for the undefined portion.

Quote:
To this day, it gets pulled out every now and then when I'm in the mood for a lighter, quicker war game.

Quick? Did you and I play the same Risk? heh.

Quote:
On the other hand, it can be said that the most random element in the game is the human element... Everyone brings a different viewpoint and strategy to the table, and even in a game where there is nothing random at all, you'll still have the chance to see wildly varying playstyles.

As evidenced in Puerto Rico, a game that is applauded for having almost no randomness, and yet each time you play the game is different from the last.

- Seth

jwarrend
Offline
Joined: 08/03/2008
Deterministic Vs Random models

knic wrote:

It does seem, though, that some of the fun is lost by removing the random element... On the other hand, it can be said that the most random element in the game is the human element...

It sounds like there are really two different effects you're considering in what you're calling "random" -- the "aha!" factor and the "hand of fate" factor. The "aha!" factor relates to not knowing in advance what will be the outcome of combat. The "hand of fate" factor relates to the combat resolution being out of the players' hands to some extent.

A game can have neither of these (e.g., chess), both of them (e.g., Risk, where the die roll also has an "aha!" effect), but it can also have only the "aha!" factor; consider Stratego, for example. By hiding some information, the outcome of a combat may not be known in advance, so there's the surprise element, yet the outcome is deterministic. LotR:the Confrontation builds off of this by allowing players to add cards to supplement their pieces' strengths, which again is deterministic but still gives an outcome that is unknown.

Basically, if it's the element of surprise that you want in a combat resolution, there are ways of achieving that deterministically. That's not to say that there aren't interesting randomized combat resolution mechanics yet to discover. In that sense, I disagree with the Barron. I think, however, that he's right that decisions are key; it's just that, in a game with a strong random element, decisions can be motivated by a risk/reward factor. There's nothing more satisfying than a crucial win against long odds, and it's much harder to design a deterministic system that can allow for such outcomes. There's room for both kinds of games.

-J

Anonymous
Deterministic Vs Random models

Darkehorse wrote:
Or the shuffle of a card? Be careful about your statement! Randomness comes in many different forms! Often times, depending upon card distribution, decks of cards are MUCH, MUCH more random than a six sided die. I'm going to play the devil's advocate here and ask you where would your CCG, Raw Deal, be without the randomization provided by the luck of the card draw? Sure, you can design your own deck, but there is still a lot of randomness in such a mechanic.

-Darke

Randomness ("luck") plays almost no role whatsoever in Raw Deal. The same is true about many of the other top CCG's. I realize it seems counterintuitive, but with the various mechanics involved in Raw Deal---

I'll give an example to illustrate how much I believe Raw Deal has little luck involved:

Let's say two players sit down that I know. I know one to be a good player and the other to be an average player. I then look at each of their decks as well as their Backlash decks.

I will wager my $100 vs. your $10 that I can determine the winner -- and I'd be making a killing over time.

If luck were a huge factor, you wouldn't have players that on the tournament trail can win 95 out of 100 games.

SOME CCG's, yes, but even with games like Magic, VS., and L5R, there is very little luck involved -- some oddities with pairings in tournaments, sure, but even these obstacles are bested by the skilled player with the most carefully constructed decks who play well under the pressure.

Any time you can construct a deck and there are various cards that allow you to make tons of decisions in the course of the game (especially in ours with the Backlash deck which are fixed cards you start the game with) you completely minimize luck to such a degree that it's almost irrelevant.

HOWEVER, as to your underlying point, yes, card games are usually luck-based and whether you are taking a card from the pile of Chance or rolling a dice, I'm against the whole notion.

Anonymous
Deterministic Vs Random models

knic wrote:
I wasn't as kind to Risk as I was to Axis & Allies in my original post, but Risk was probably the first large scale game I ever played... .

Axis & Allies has the immediate design problem of Russia destroying Germany on the first turn, effectively making them irrelevant for the remainder of the game, barring some immediate pressure from Japan.

Risk is an interesting model that I am maybe more of a fan of than most, as there are so many decisions that are made -- and quickly -- especially when everyone picks their own countries out of the gate (vs. random dealing them out). It's problem comes in the rewards, where it should be:

Australia: 2
South America: 3
Africa: 4
North America: 5
Europe: 6
Asia: 7

I made a point, I think on this board, that dice are often problematic when a few are rolled. If I roll one dice and you roll one dice -- blah. However, in risk, you are oftentimes involved with decisions where it will you be rolling 30 dice against your opponent rolling 25 and then "luck" plays less of a roll as a whole due to the large sample size.

Risk has less luck than would seem likely due to all of the rolling, but it's evidenced by the best players almost always winning, the same you will say about Axis & Allies (especially when you let Germany alter their initial troop layout to overcome the "we can't hold up against Russia" factor).

Hamumu
Offline
Joined: 12/31/1969
Deterministic Vs Random models

It sounds to me like it's not randomness you disdain, but "luck". Your CCG isn't deterministic, no matter how much you say it isn't about luck (and I don't dispute your ability to guess who will win) - drawing cards from a deck IS random, even if you get to pick what's in the deck, as long as it's shuffled (and it's not all the same card...). A game doesn't have to be massively random to have random elements. Rolling 30 dice is still random, it just has lower probability of outlying values.

I'm a firm believer in randomness, it adds a lot of fun to games. Purely deterministic games are too stodgy for me - if it's possible to completely figure out all outcomes, invariably playing well entails spending a lot of time carefully thinking out all the angles. I'd rather just give it a shrug and go for it. I also like hidden elements (and I'd be okay with purely deterministic if it has hidden information - because it still comes down to "I'm just gonna try this and hope it works!"). I don't like things to be really overly random though. I kind of enjoy yahtzee, but it's really too random. But boy does rolling a full yahtzee feel good!

An interesting aspect of the deterministic vs. random argument is human behavior. It can be either... but you don't know which it will be in advance! And even in a purely deterministic game, you can't ever assume your opponent will always make optimal choices (in fact, he surely won't), so there's a bit of a random element in how opponents will behave. Not that that affects whether a game element should be deemed random or deterministic - the person isn't part of that equation!

But my main issue is that: it's not entire GAMES that are random or deterministic (a purely random game would be utterly pointless... e.g. War; though there are of course truly deterministic games), it's game elements. Just because a game isn't heavily based on luck doesn't mean it doesn't include random elements. A CCG definitely has a random element without a doubt. Doesn't mean skill is not a factor.

Anonymous
Deterministic Vs Random models

I realize the distinction you're going for, but just to make sure my point isn't missed: randomness leads to situations where luck becomes a factor, and if that is some percentage over a certain point (I'd say 10%) of determining the winner, that is too much. And I'd probably go with more of a 5% being the key -- i.e. the person who makes the best decisions should win far more often than the other. Obviously, yes, there will be slight elements of randomness / luck in almost anything -- but if it isn't completely minimized to the point of almost being inconsequential, that I believe is problematic.

Axis & Allies and Risk both are on the absolute brink of the amount of luck that I personally like to see in games (by the way, that's all of these differences come down to, personal choices and preferences), and I let them both slide probably because of random fond memories of yore when I was a little Barron blitzing through Europe in my tank OR hiding off in Australia wondering when I'd take over Asia.

I've literally bought or played 100's upon 100's of games in the last few years as I've tried to give myself and my friends a crash course in gaming, moreso than in the past, and far too many games have far too much luck and that is the basis of my point.

Differentiate how you will: decisions should be the primary (read: 90%+) element of a game in my book. Others may differ in their opinions and that's cool, but the first thing I think about when I'm thinking about a new game is whether happenstance will decide the winner. If it's yes -- I move along.

You can have some modicum of it in the game, but it should be minute.

Anonymous
Deterministic Vs Random models

Just a quick comment to add into this discussion. I remember a certain CCG which has a decent mechanic for adding a randomized element to combats. It was/is called Star Wars (the one by decipher), where every card that you put into a deck also had a random number generator on it. This random generator was the inverse of the power of the card in the deck, for example Darth Vader would have a random number generator of 1, while a card like disarmed had a random number generator of 5 or 6. So if you stacked your deck full of powerful cards, you would be losing some of your advantage in the randomly generated numbers.

I personally of course am not a big fan of straight up random, while I am a big fan of a deterministic system with hidden information, or a deterministic system with an opportunity cost associated with the different choices.

Personally, right now I am interested in the opportunity cost variety of spicing up combat. This reminds me of Magic: The Gathering, because every turn you have an opportunity to attack, but might have to give up various things based on board position, you open yourself up to conterattack, as well as possibly having to sacrifice a creature to your opponents monster that is sitting there ready to eat your little critter, etc.

Some more on this may come around later today, as for now, I must go.

Anonymous
Deterministic Vs Random models

Ideally, I think you'd like to aim for a game 50% strategy and 50% tactics, but that is VERY hard to balance and most games slant one way or the other (in the discussions of CCG's, Raw Deal falls more on the strategy side, whereas Magic falls more on the tactical side).

If you really want to make something truly unique, have something 50/50 -- that I'd love to see as usually everything is more of one than the other.

Anonymous
Deterministic Vs Random models

BarronVangorToth wrote:
Ideally, I think you'd like to aim for a game 50% strategy and 50% tactics, but that is VERY hard to balance and most games slant one way or the other (in the discussions of CCG's, Raw Deal falls more on the strategy side, whereas Magic falls more on the tactical side).

That brings up another interesting point (though I don't know if it ought to be the start of a new thread instead of a tangent here)... What, in your opinion, is strategy versus tactics as they relate to various game genres? I guess for the most part, I'd say strategy focused games require a clear, long term goal (for example, "I want to win through economic means, so I'll focus there instead of on my military" or "To have a chance, I'll need to take North America, then move across into Europe"), whereas tactical games are more based on reactions to the current situation ("He's got Cavalry so I need Pikemen" or "Looks like he's got a lot of red cards, I'd better focus on blue for now"). Just as you said, each type will bleed over to the other (strategy still needs to have some room to react to unexpected situations, and tactics need an overall goal so the game will eventually end).

Which type of system, random or deterministic, is better suited to each type of game? I tend to feel that randomness has more of a place on the large scale strategic games, where the outcome of a single roll won't (usually) make or break the game, and smaller tactical games are a better place for deterministic systems to allow players the chance to focus more on the tactical side rather than hedging their bets against a bad die roll.

Anonymous
Not sure about that CCG arguement...

but I've never played Raw Deal.

If you had an expert at a game shuffle their cards, and you let an okay player sift thru their deck and put them in the order of their choosing, I would guess that the okay player would kick butt.

I think that there is a perception that dice are "too random" by a lot of people, but these same people play card games and such.

Personally I like a game that combines more than just 1 element to them, luck included. The feeling of doing your best, and still losing is okay. Now if there are only elements of luck (which I don't think Risk or Axis & Allies is), such as The Game of Life I'm not so happy.

Anonymous
Re: Not sure about that CCG arguement...

jjacy1 wrote:
If you had an expert at a game shuffle their cards, and you let an okay player sift thru their deck and put them in the order of their choosing, I would guess that the okay player would kick butt..

This would be an interesting experiment and you'd have to see just how "okay" an "okay" player is and how "expert" the "expert" player is. This would play a far greater goal in any resource based CCG (say: Magic) where you could put the necessary land you required right on top and put all of the remaining land on the bottom.

While it is counterintuitive I realize to believe CCGs aren't very random -- it's nevertheless the fact of the matter, as excessive randomness would prevent people from being able to consistently win ... and win a vast majority of their games.

This is akin to thinking poker is any semblance in the long term to a game of chance -- an obviously incorrect thought as well.

Hedge-o-Matic
Hedge-o-Matic's picture
Offline
Joined: 07/30/2008
Deterministic Vs Random models

Well, I was facinated by random systems for years, and then started to move toward how random components to a system could create more organic systems where some parts were detirministic. Lately. though, I've didtched random factors to see what can be done with non-random elements, under the theory that the more players interact, the better. So I really like bidding and what have you. But a random effect in a player's mind isn't that difficult to achieve merely by adding a few factors that can't be calculated exactly (such as hidden information), or adding another element, such as time pressure or a built in distraction. If combat is the subject, then skilled warriors need to block out the tension they operate under.

zaiga
Offline
Joined: 12/31/1969
Re: Deterministic Vs Random models

knic wrote:
To summarize, I feel that deterministic models typically favor tactical thinking (whoever makes the "best" moves will win), and random models run the gamut from simple luck games to games that attempt to equalize the balance between luck and skill.

I don't think you can say that deterministic models typically favor tactical thinking. A good counterexample would be Chess. It's a purely deterministic game, but it's also very strategic. The other end of the spectrum would be a tactical game, with a lot of human interaction and randomness. Citadels and Scrabble come to mind in this regard: you basically optimize your own turn, and perhaps you can plan a bit for next turn, but mostly you are dependent on what you draw and what your opponents do.

A deterministic model itself doesn't provide any tension or "fun". If you are using a deterministic model, the tension should come from elsewhere. For example, the combat system in Vinci in purely deterministic, there are no unknown factors during your turn. The tension in Vinci comes from deciding when to "go in decline", hoping your opponents make moves that don't hurt you too much and from choosing a new combination of civilization tiles.

Adding a bit of randomness can be a good way to combat analysis paralysis. Imagine Axis and Allies with a purely deterministic combat resolution; the combat phase could last for ages with players trying to optimize their results. In a sense, players could do the same with a random combat system, optimizing their odds, but hardly anyone can figure out the exact probabilities for each attack on the spot, which forces players to go with a "more or less" approach and just do something that feels right.

Luck is often derided by gamers, but it is an essential component for a lot of games, and it can be a great source of tension when done right. However, dice and other randomizing devices are also often used as a cop out by game designers, sometimes used in places where a player decision would have been more appropriate or more effective. When I come up with a randomizing mechanic I often wonder whether it wouldn't be better to let a player just choose a result, perhaps adding a cost for certain choices, or making the mechanic deterministic. Sometimes a random system fits better with the game, sometimes a deterministic system; it really depends on many things.

I also feel that a human opponent is often the most satisfying randomizer. You have a general idea of what they are going for, but you can't really ever be sure. A good example of this is Puerto Rico, which has very little randomness, yet every game plays out differently because players make different choices each time they play.

Anonymous
Deterministic Vs Random models

I just read through this thread and want to add my 2 cents to it:
IMO it is not important how much randomness a game features. It is just important that players BELIEVE they can influence the outcome! :wink:
What do you think about this?

zaiga
Offline
Joined: 12/31/1969
Deterministic Vs Random models

PeterTheRat wrote:
IMO it is not important how much randomness a game features. It is just important that players BELIEVE they can influence the outcome! :wink: What do you think about this?

I wholeheartedly agree with that statement. Essentally, it boils down to this: do players feel that they play the game, or do they feel the game is playing them?

For example, suppose you have a game where at the start of your turn you roll a D6, and you get that many action points. I think a lot of players wouldn't like this mechanic, because the dice decide how much they can do during a turn, and they can't influence the dice. In essence, the game plays them.

This complaint is often leveled at Settlers of Catan. You decide where you place you initial settlements, but after that it's up to the dice. If you are unlucky with the dice, it could be that you don't get any resources for several turns. Luck in a game isn't bad, but having nothing to do is.

Another example, suppose you have a game where you have to roll dice and "beat" a certain number to be able to perform a certain action and where you may buy extra dice to increase your chances. Say you have to roll 5 or higher. You could buy just one die, but then your odds are pretty slim. If you buy four dice you are almost certain to roll 5 or higher. Of course, you could still get extremely unlucky and roll four "1"s. I think players will regard this mechanic as much less "random" as the one in the previous example, because players have some control over how much risk they are willing to take.

larienna
larienna's picture
Offline
Joined: 07/28/2008
Deterministic Vs Random models

Good topic and interesting comment.

I would say that both system are good and both system are bad. It's a matter of balance. This is the area where the game designer must judge if you must place a random or deterministic element. It's a game designer talent.

What I hate most about deterministic games is the fact that you know at some point in the game that you will die and there is no turning back. So you just get bored and surrender. You need some element that will make sure that you are not completely dead.

Making totally random game make the player have no control on the game. He just roll dice and see the result. it's like playing slot machine in casino, pretty boring. Normally, in board games, the randoms elements are influenced by different elements of the game which does not place the odd 50%/50%. So making decision in your game improve the odds but don't give you automatic success.

As Sun Tsu ( or one of his fans ) said : few army troops can win against many. The random element here represent the tactical ability of your commanders which might not be represented in the game.

Another example, in advance war ( GBA video game ) the combat results are deterministic. If you do not start really good or if you do not do the right things, you are doomed. I once added a simple random element that could reverse this strategy, it was some sort of initiative at the beginning of combat. Adding this element could make some surprises that would allow players to reconsider their position.

I think using random element where, in real life, success is not abolute is the key. The player must still feel that he can influence to flow of the game.

By the way, my farther and his friend, in the game "pay day", started making side bet on all dice rolls in the game. It make the game pretty much interesting and gave the player some control on the game which they did not have.

Anonymous
Deterministic Vs Random models

@Larienna

You are absolutely right. If a game gets too deterministic, it looses much of the appeal, to many players. (game example: Chess)
1. in a deterministic game analysis paralysis can happen more often than in a game with random element, where someone can take some risks and make choices from the gut
2. in a deterministic game a casual player is put off, because he sees no chance in winning against an experienced player
If a game gets to random it looses again much appeal. (game example: Snake and latters, sorry for this bad example)
1. in a random game players have no influence in the gameflow and feel they are played by the game

Still there is another moment, that is not pure random and by no way deterministic, that must be considered: When a game gets chaotic. (game example: Citadels)
I think also here it is important that people should FEEL they can somehow influence the outcome of the game. Chaotic game should always be shorter than less chaotic games.
Still I think chaos can be a nice element to a game. Bruno Faidutti does a great job in introducing Chaos in his games IMO.
Your opinions?

larienna
larienna's picture
Offline
Joined: 07/28/2008
Deterministic Vs Random models

I have not played citadel, so what do you mean exactly by chaotic games?

Anonymous
Deterministic Vs Random models

@Larienna

Hi, I will try to answer you later!
Right now I am here at work and don't have the time to go into detail!
Let me just propose this interesting article in the meantime:
http://www.westpark-gamers.de/index.html?/Artikel/bad_luck.html

Even Bruno Faidutti did an article about chaotic elements
http://www.faidutti.com/index.php?Module=listes&id=34

Bye by Peter

zaiga
Offline
Joined: 12/31/1969
Deterministic Vs Random models

Larienna wrote:
I have not played citadel, so what do you mean exactly by chaotic games?

Basically, a game can have no randomness, and still be very chaotic, because you cannot predict the actions of other players. A good example of this is Puerto Rico. There's only a small random element - the plantation draw and the initial seating position - yet it's very hard to see more than a few turns ahead, because you cannot really predict what other players will do, even though there is no hidden information.

Anonymous
Deterministic Vs Random models

I see zaiga has already answered Lariennas question!

Chaotic means some unpredictable actions that are not created by the game but by the players! Normally this happens in highly interactive games where it is literally impossible to predict the outcome of ones move.
Also games which have integrated some rules or cards to intercept a players move are by nature chaotic. For example MtG: There someone plays a card which has a certain function. But as long as the other player has some cards to use and some mana left to use them you never can be sure if your move will indeed finish as predicted.

I personally think the epitome for a chaotic game is "Hoax"! The whole game is build around this chaotic feature. There are no dice and no other randomness inducing mechanism, just the rules who allow the players to act and react to the behaviour of the other players.
Just check its rules at BGG!
So even if this game has a structure and no randomizers it's fairly unpredictable how a game of "hoax" will end!

So the problem in a standard game is the scaling of the game from 2 to4 or even to 6 players!
By nature a game that is quite calculable with 2 players may become very chaotic when more players take their turn before someones next turn! One game that suffers from this is for example "Clans"! I find it a VERY GOOD 2-player game, but when it comes to a 4 playergame I just play it for fun, because winning is matter of ..."chaos" or "Bad" or "good" luck if you prefer.

I think this pointis often forgotten when testing a game! Or when it is considered it is just tested with a certain number of gamers. It may still work with another number of players, but the whole feel becomes more "chaotic"!

So why doesn't Magic suffer? The answer is, that the whole game is BUILD AROUND this chaotic elements. And there is where most of the fun lies during a match of MtG, apart from the "Metagaming"!

What are your opinions?

CRasterImage
Offline
Joined: 12/31/1969
Deterministic Vs Random models

BarronVangorToth wrote:
I've literally bought or played 100's upon 100's of games in the last few years as I've tried to give myself and my friends a crash course in gaming, moreso than in the past, and far too many games have far too much luck and that is the basis of my point.

Barron, be careful not to mistake game systems as being luck-based purely on superficial inspection. Many games may appear to be heavily random upon first look.

Let me take an example: Space Hulk

At first glance at the rules and for the first couple games, it may seem to be a dice-fest. Terminators shoot at Genestealers and hit sometimes. Genestealers claw at Terminators and hit sometimes. Sometimes the Terminators achieve the mission goal, sometimes the Genestealers stop them.

It takes time to start to become aware of other patterns going on under the surface of the rule system. Patterns that are achieved deterministicly. When players discover these patterns they find that the game has deeper layers to it than they thought. Terminators start using Overwatch for proper covering fire. Genestealers start learning how to perform proper "rush" ambushes.

Pretty soon you find that you are playing a completely different game than you started out playing. The dicefest has turned into a tactical challenge for both players.

If you have been looking into hundreds of games recently, in order to gain a "crash course in gaming", then you may have missed some of the deeper challenges in those games and dismissed them prematurely.

Games are like Ogres, which are like onions, they have layers.

Syndicate content


forum | by Dr. Radut