Musings On... The Role of Chance in a Board Game (#18)
Richard Young: We hear people complain about games being too random and I've had lots of discussion variously on the design of combat resolution mechanisms that either have no chance elements (deterministic) or some random element (non-deterministic).The issue of general randomness (Settlers of Catan), vice a particular aspect of the game (combat), is almost a separate topic but I'll lead off by stating that I have no particular quarrel with games that have an intrinsic random aspect to them, like Settlers, but I get a little upset at games, particularly war games, that include combat routines that are more than a little random (to the point of being totally unrealistic). One of my beefs in this regard involves "dice-baskets" which I tend to dislike. Other pet peeves will be uncovered as we get into this I'm sure...
Alfred Wallace: Let's divvy this up between wargames and euros.
Wargames first:
"Dice Baskets" (sometimes known as "buckets of dice") may not be very theoretically elegant, but they're way better than every CRT I've ever seen. For one thing, they do something that no CRT does: For every increase in force, there's an increased chance you'll hit something. There are ways you can jimmy CRTs to do that, but none do it automatically: Most of the time, a 6/5 rolls on the same table as a 5/5.
I'm guessing that this conversation is eventually going to veer over to a discussion of whether battles are random ("rolls of the dice") or deterministic, so let's get that out of the way: They're deterministic. Everything hinges on some sort of choice or predetermined situation. Officers put men into position, men break or fight, guns fire or misfire (depending on how well they're built and maintained, the quality of the cartridge, and much else), things go where they're aimed dependent on umpteen atmospheric and chemical properties, and so on and so forth. In some sense there's not a die roll or coin flip to be seen.
That said, wargames don't cover all this, not even close. You're not everything that made up the battle, ranging from the person who made the bullets up to the grand strategic leader. You play somebody with a very limited range of things you can control. Most wargames ignore this and give the player extraordinary powers, mostly for the sake of
simplicity. Some Civil War games make it harder to give orders to McClellan, say, but once you spend what it takes the McClellan counter tends to snap to attention and move wherever you want him to.
Lincoln would have thought that was *totally* awesome.
A lot of what happens in battles isn't due to any decision that the commanding general made. At Gettysburg, Lee didn't order Stuart to stay away, Early and Ewell to sit around on July 1, Longstreet et al to sit around the 2nd and 3rd, and so on. (The potential efficacy of any of that happening differently probably shouldn't be discussed here, but ponder the illustration.) For Lee (aka "The Confederate Player" in a Gettysburg game), not some outside frame of reference, whole swaths of the battle were outside his control and thus essentially up to chance. From an independent, essentially omniscient frame of reference (say, the author of a book), then maybe the chance element disappears, and all is up to choices, but for a player of a game, representing a severely circumscribed point of view, that's just not the case. I don't think this lack of control (over your own side!) can be reasonably modeled deterministically.
Now then:
One can argue that games shouldn't assume that the player represents one person or, perhaps more precisely, one Decision-Responsible Party (i.e., you're not Lincoln or Davis, but the respective bodies responsible for directing the war, who act as one body (usually)), but should represent the entire side--Union or Confederacy, continuing the example--and therefore should assume a much more widely-encompassing point of view. In my opinion and experience, these can be great, interesting games but oversimplify the
situation.
This seems like a good time to bring up Bonaparte at Marengo, one of the very, very few (nearly) totally deterministic wargames out there that seriously attempts to be a simulation.
It's a great game. I love it. It's innovative, beautiful, and provides an interesting story and "looks" like Marengo. That said, I don't feel like Napoleon or Melas. It's a historical in its own way; there's too much control held by the player.
I'll say right now that my "dream wargame" would be pretty chaotic for the tastes and expectations of most wargamers--never mind complicated. Again, the ACW, writ large:
The player submits orders for each unit on the board (say, "move to hex Y13," "attack the guy in front of you," etc), which one then checks to see whether they're accepted or rejected by the commander of each. If they're accepted, then the orders are carried out. If rejected (which, historically speaking, could represent a lot of different things), then it generates its own orders, which could be anything from sitting in place, retreating towards its own "lines" or friendly units, attacking the closest enemy unit, striking off towards the nearest objective hex--whatever. Combat would be fairly random, modified by terrain, size, commander abilities and whatever else. Lincoln and Davis didn't get to decide what happened at Gettysburg, neither should you.
That'll never happen in a board game, I don't think, at least not a big one.
To recap, I think the biggest difference that lies between "deterministic" and "non-deterministic" types in wargame preferences has to do with what, ultimately, wargames should try to model. Should they model the war or battle in the abstract, like a book, and give the player a lot of control (over the player's own forces) and leave relatively little to chance? Or should they give the player a "role," putting the player in the shoes of someone in history and allowing that player to influence only what the historical model could influence? I think the former could be done deterministically, but not the latter. I don't think this difference is made very explicit most of the time. I, personally, feel that players should usually play historical roles.
Now then--for those few, proud, and strong of you still reading--non-wargames.
It's interesting that I can think of several games with too much luck, but not with too little. I mean, does anyone--and I mean anyone--really like how the shields work in Louis XIV? It's like a little lottery at the end. Roads and Boats (one of my favorites) is totally deterministic...except for mines in non-solo games, and that just irks the heck out of me. (I'm easily irked. Why would one little thing be random and none of the rest of it?)
An interesting idea that I just had, but I'm not sure how valid it is, is to ask whether good luck (or, perhaps, the avoidance of bad luck), for a given game, is a necessary, sufficient, or irrelevant condition for winning a game. A game where luck was a sufficient condition of victory--LCR, Which Witch?, Candyland, some hypothetical game that lasts seven hours, and then you roll 1d6, high roll wins the game, etc--would probably be considered by everyone to be bad, or maybe only good for children or possibly drunk adults. (The examples given are also games where good luck is a necessary condition for victory.) A game where at least avoiding abysmal luck is a necessary, but not sufficient, condition (a great many games) would probably hinge on how likely it is to avoid game-wrecking runs of bad luck. Games where luck is an irrelevant factor would probably be, I dunno, mostly abstract games of perfect information.
(There could well be games where luck is a sufficient but not necessary condition of victory--I can't think of one right off the top of my head, but that was the extreme position of the early criticisms of Battle Cry's "All-Out Assault" card, so I'm sure there are real examples.)
Probably something wrong with that somewhere.
Larry Levy: Amazingly, the answer is yes. In a recent user group thread about this very topic, several gamers came out in favor of the mechanic. Joe Huber is one such person. I believe his exact quote was that he wouldn't like the game nearly as much without the endgame lottery.
One thing I've found through bitter experience is that there are just about no universal opinions in this hobby. It can make it difficult to make any authoritative statements about anything. On the other hand, at least it keeps things interesting.
Jay Little: Regarding luck... In general, the longer the gameplay, the less luck I want determining the outcome. Luck/chance/randomness is still welcome, as it can throw unexpected curves into the game or help equalize things, but it all boils down to a few personal preferences:
1) In general, when I lose, I would prefer the loss be due to exceptional play by my opponents or poor play on my part.
2) In general, when I win, I would prefer the win be due to exceptional play on my part or poor play by my opponents.
3) The longer the game, the less I want the role of luck to determine the winner.
In a game of Memoir '44, with a single battle lasting 20-30 minutes, I really don't mind rolling horribly while my opponent rolls far above average, or getting nothing but Left Flank cards when I've got no troops in the Left Flank. Luck as an overriding factor in this case doesn't bother me. It's still a fast-paced, fun experience, regardless of the outcome of a single turn (even if it is pivotal to the game).
In a game of Hammer of the Scots, on the other hand, with a game taking 2-2.5 hours, I start to get irked when the turning point in the game is determined by luck.
For example, in my last game of HotS I played the English. I had two consecutive years where I drew four 1s and a single 2 for activation, while the Scottish player had all 3s and 2s. In a pivotal battle, I proceeded to roll 18-20 dice at B2 and missed with all of the dice,
while my opponent rolls 8 dice at B2 and hits with 7 of the 8. When my tactical decisions and plans are foiled simply by luck, rather than poor/exceptional planning by the players for a longer game, it bothers me.
In games that take longer than 3 hours, my tolerance for luck as an overriding factor diminishes quickly. There reaches a point where I'd rather concede a game to luck, then move on to something where all players start back on a level playing ground.
Alfred Wallace: I stand corrected!
Good to be reminded, once in a while, that there is no universally-held belief out there about gaming. (There are probably people out there who would disagree even about that!)
Or, put another way, "People even rated Wadjet a 9."
Seth Owens: I don't have a single point of view on chance in games. My personal tolerance for luck or chance in a game design turns on several factors.
A) Commitment. I don't mind luck so much in a quick, light game. Good examples are Memoir'44, Nuclear War, Cosmic Encounter and most traditional playing card games. Chance in this case adds variety to the game and there's usually time to play again if the dice or cards messed with you. Plus everybody knows luck plays a big role in the game so it's easy on the ego.
B) Modeling. I don't mind luck or chance if it's necessary for an authentic model. This mostly applies to wargames, although it certainly can apply to some extent in other genres (such as economic and business games and sports games). Fortune is such an integral part of the battlefield that it's hard for a serious wargame to avoid it. While one can get into deep philosophical discussion over whether life is ultimately "deterministic" or not, for all practical purposes the battlefield is not. Besides all the physical factors, tactics, doctrines, weather (macro and micro), etc. there is also the contest of wills involving each and every single participant. So many intangible and tangible factors are interacting that are beyond the control of any individual actor (general, king or sergeant) or group of actors (army, nation, or squad) that the most realistic choice is to include some randomization. While Bonaparte at Marengo is a most fine game, and a serious wargame, in the end it's an abstract wargame, not a simulation. (Bowen has a great discussion on chance on the BaM site, by the way, although I don't agree with his conclusions). The single biggest compromise wargames have to make is the fact that they
must give players far more control than any historical commander could have, At Waterloo Napoleon sent less than a dozen orders all day. So much of what happened was determined by the decisions of subordinates. The player always represents both the commander and several ranks of subordinates. After all, somebody has to move the counters.
C) Replayability. Even BaM has a certain random element to it, due to the variable French setup. Many games need some randomness in order to avoid stereotyped play.
I'm much less tolerant of randomness in "serious" nonwargames. While it may be necessary, if regrettable, that a 4-hour historical war game turns on a run of good or bad luck, that's war. But if the game is just a game, then I'd prefer not to lose a 4-hour game on a die roll. While it's not exactly a serious game, one reason I really don't like the "Killer Bunnies" card game is that it's really, in the end, a lottery.
So the bottom line is that I say let the dice roll if it's a wargame, but otherwise leave the decision-making to players, not the dice if it's not.
Morgan Dontanville: I agree Jason, I think the length of the game certainly shapes my willingness to tolerate luck. I also feel that the theme and the mood also help me to tolerate a luck-heavy game.
One of my biggest soft spots in games is the offbeat theme-heavy filler. I tend to find these games much more entertaining than most. I think, perhaps I am a part of an exclusive few that would put Der Plumpsack in their top 50 games (which is a contrary example as it has no luck, but you get the point).
In my quest for these weird little gems, I find a striking trend for game designers to want to create a longer game in order to mitigate luck. I have a number of games that expect you to play the entire thing over again three times to even everything out (Jump! and Fjords immediately come to mind). My feeling is that the element of luck will be the same, perhaps it will swing around and perhaps it won't, lengthening the game seems superfluous. I think this is an archived remnant of card games that expect you to play multiple hands, but this seems silly to me if you are playing a filler game.
The nature of the filler game is to be short, in between longer games. And there is no shame in having a short, beautiful, luck filled game. If you like it and want to play it again fine, but don't expect the game to get "better" or to even out in luck if you make the game longer.
Some games' themes well make up for the absurdity of the dice. Das Faultier/Le Paresseux is a silly roll and move game about moving sloths and trying to move the slowest. Most of the time you are moving other players' sloths. A couple times during the game you put a sawing wood token on your little guy to indicate that you are asleep. Sometimes, everyone goes to sleep and the only way to wake people up is to roll the wake up roll on the die. Often it can take a long time to get the roll. The absurdity of the game and the overwhelming size of the die just makes the game funnier and funnier as everyone sleeps. There are certainly some decisions to be made in this game, but
often this game comes down to luck (dice rolls and cards). Who cares though you are sloths, in a race for last.
Larry Levy: I've always preferred board games with less luck. However, I also love card games and many dice games, both of which are almost always high-luck by their nature. It was hard for me to explain why I liked luck in some instances and not in others. Then, I read a posting by Allen Doum which made the whole issue much clearer to me.
Allen divides luck in games into two categories that he calls Resolution luck and Situation luck. With Resolution luck, the player makes his moves and then uses some randomizing device to determine if his actions succeed or not. The classic example is the dice-driven CRT used to resolve combat in most classic wargames. With Situation luck, the chance elements come first, and then you have time to react to them. Most card games are like this; in Bridge, for example, the 13 cards you begin with are generated purely by chance, but the way you bid and play them is then entirely up to you.
I dislike Resolution luck, but find Situation luck much more acceptable. My problem with Resolution luck is that there's no way to react to it. You roll the dice and if you lose, that's that-- there's no appeal to the governor. This is not only frustrating, but it means that bad luck of this sort is usually fatal.
With Situation luck, though, there are often things you can do. For example, if I'm dealt a bad hand in a card game, I usually have some options. Depending on the game, I can play defensively, or try to interfere with my opponent's play, or try to hasten the end of the hand. Dealing with bad breaks in such games is frequently a very enjoyable challenge. I still usually prefer games with less luck, but if the luck present is Situation luck, I'm much more likely to rate it highly.
Naturally, many games have both forms of luck, which is why I'll object to some luck elements and not to others. For example, in Louis XIV, I don't mind the fact that the Influence cards you get each turn are dealt out randomly (there's lots of ways to mitigate against this, since there are many different ways of playing each hand), but I dislike the game-end lottery (which is the ultimate in Resolution luck, as it occurs after the game is over!). Actually, deciding how to play each hand is so enjoyable that Louis is a big favorite and I'm willing to overlook the shield mechanic (although I wish another game-end method had been chosen). Wyatt Earp is another example. The card draws are pure luck, but they also present a fun challenge in how to play them. However, resolving "bullet hole" Gavel cards is Resolution luck and its effects are so annoying to me that Wyatt Earp is my least favorite of the Mystery Rummy-type games.
Dice games are practically a different category entirely. For me, the fun in push-your-luck games like Can't Stop or Pickomino is calculating the probabilities and doing the action that maximizes my chances of success. The luck is SO prevalent in these games that you can't get too caught up with winning and losing; after all, your best move may only increase your chance to succeed by 20% or so. Thus, I tend not to worry whether or not I win these games and play them more for their intellectual challenge (actually, with Can't Stop, it's more like experienced-based intuition, but the concept is the same).
Jay Little: Larry's comments on Resolution vs. Situational chance is spot on with my personal tastes. The longer the game, the more I favor Situational over Resolution luck. The shorter the game, the less I mind Resolution luck determining the outcome over my own decisions.
Tom Vasel: A lot of my tolerance for luck in game depends on the claims of the game itself. It the game is heralded as a game of great tactics and strategy, then by all that's holy, it better not be based on luck! But if the game is advertised as a fun, light game, then I expect that luck will probably play a large role.
I also think that there is a perception problem. Can't Stop probably has as much luck as other games. But because it gives me a choice to make, I don't mind as much. Other games, such as Lucky Loop, seemed to force me into a beaten path, and therefore just seemed less fun as a result. If a game makes me THINK that I have strategy, then I'll
enjoy it more - regardless of luck.
Morgan Dontanville: Tom, I agree with you that luck should be properly represented to meet the depth of design. But, you seem to rate E&T an 8 and it is certainly heralded as a game of great tactics and strategy. For me it is far too luck based when matched with the level of intensity and the overwhelming weight of decisions.
Nothing like drawing all blue. In a game where every action is vital, losing the actions to draw again is crippling (especially when the second draw yields poor results) and trying to work with what you have feels like being a poodle in the bear baiting ring.
For me I don't mind luck all that much if the game allows for you to mitigate it. Struggle of Empires is a good example of this. There are a number of elements to this game that ultimately are determined by the dice. I notoriously have horrific die rolls, but Struggle allows me to modify the results as you mold your civilization. If I've invested heavily in the new world then I'm most certainly going to get Navigation to aid in my journeys back and forth. Even though, my movement still sometimes falls victim to the whims of lady luck, I feel that I have made choices that help me to advance, and I'm better off because of it.
In many cases, I enjoy getting a bad hand or an unfortunate piece of luck that I have to figure my way out of. When I succeed, I feel satisfied by the journey. When I don't, I know that I had a crap hand or was dealt a bum deal. That's why I don't mind the TransAmerica starting cities (although it is fun to complain about it).
I will say though, that I don't like waiting around to get lucky. Ticket to Ride annoys the hell out of me mostly for this reason, and T2R:E is even worse (I once literally had half the deck in my hand and could not connect my cities as I was waiting on the required wild cards for 1/2 the game).
To get to that whole shield thing in Louis, it makes me feel like I'm reading a What if... comic. What if you would have gotten more points, this would have been the result! In a game where points are precious, and working out how you can pull off the win in the case of a tie seems vital, the random point allocation feels pasted on and absurd. It makes me think that someone said, "Man, this game is too hard. I went for the shield strategy for the third time in a row and didn't win. Can we do something about that?"
Larry Levy: With regard to E&T and Struggle of Empires, Morgan, my feelings are the exact opposite of yours. While I have downgraded E&T a bit because of the luck factor, I think you definitely can mitigate against bad draws in that game. If I were better at the
game, I could tell you how in some detail, but the consensus among good players is that there's always something you can do with your hand.
SoE, though, is classic Resolution luck. What you call mitigation is actually taking out insurance. Certainly, if you pursue certain strategies, acquiring specific tiles is optimal strategy. But there are times when, in spite of proper play, you get hammered with the dice. I saw my good buddy Anye Sellers completely destroyed by dice rolls in one game. She didn't stand a chance; it was an amazing thing to see.
Supporters of this kind of action resolution always say things like, "Well, if you had just done this and that, that roll wouldn't have been a problem." Sure, but if I gotten a more likely result with the dice, then I would have wasted precious actions on those insurance policies, which can be just as bad. The fact remains that I can't adjust to the randomizing factors after they happen, so I am at their mercy. I think SoE is a very good game, but I'd like it more if the combat resolution didn't have such a large variance (and I'd really love it if Wallace had found a less luck-driven way to resolve actions, just like he's done with some of his train games).
The shield mechanic in Louis is just plain stupid. Fortunately, it doesn't affect things too often, so I can live with it. But it really sucks when you lose (or win) because of some cockeyed distribution. I like the idea of some sort of bonus points for lots of shields, but it wouldn't have been hard to make this bonus deterministic.
Frank Branham:
Regarding Resolution Luck:
I do not think that it is always to be avoided in strategic game design. Many wargames are prime examples of how Resolution Luck can be used well. In most games, there are a lot of conflicts requiring die rolls that form the basis of the game. The slower moving game structure and things like zones of control allow you to plan for possible failures, and fix any problems on following turns. Struggle of Empires, with fluid movement and fewer battles, does nicely highlight how Resolution Luck can hurt a game.
Whimsy:
There is another term for chance that I like to think about, which Bruno Faidutti labeled as whimsy. A lot of chance features in a game determine resource allocation or merely resolve the simple success of failure of an action. Event cards would be the most obvious example but the random movement of items on the board (Valley of the Mammoth's rampaging animals would be a prime example), or the patient deck in Quacksalbe-add more to the game than just chance.
If the element of whimsy is presented effectively, a game can really exist almost entirely upon its capriciousness. Tales of the Arabian Nights, for example, has almost arbitrary choices, and is mostly a massive pile of random elements.
Rick Young: I too would like to thank Larry (and Allan) for giving us the "Situational" versus "Resolution" luck references, as they help us to focus on what we are talking about with some useful shorthand. But, as always I guess, there are potential problems with any generalization. I'm left wondering what you would call the die rolls in Settlers of Catan.
Larry Levy: Rick, I think I agree with you that this is a more complicated case. I'd say they're about 80% Situational and 20% Resolution. The former is because you can certainly adjust to the resources you get, through trades and different building strategies. But as you say, if you don't get any resources, there ain't nothing you can do. That aspect feels more like Resolution luck. Those dry spells are sufficiently dull and frustrating that, like you, I'm not a big fan of Settlers. However, I'm pushing the luck toward the Situational camp because in most games I've witnessed, players can recover from bad rolls and find some way to win.
Rick Young: Everything in Settlers is deterministic except for which hexes are going to produce each turn. I think a case could be made for that mechanic to be either situational or resolutional. But given the term resolutional probably was meant to refer to the resolution of a battle in a wargame, then let us say for the sake of argument that it is situational. I'm not sure what your ability to mitigate is in an extended series of bad rolls in Settlers, as resources are required to do anything at all; so, when you are frozen out, I guess your only mitigation is patience (and this can happen regardless of how astute you were in picking your intersections, btw). I don't find that particularly comforting and is one of the reasons why I'm not a big fan of Settlers. Perhaps if it was a shorter game this wouldn't bother me so much.
My reason for dwelling on that for a bit is to introduce the notion of "luck management" which has been referred to here already but maybe not by that term. I don't feel I have a lot I can do in terms of luck management in Settlers in those games where I appear to have chosen all the wrong intersections (despite basic statistics telling me that my choices should have been alright). By the time the dice "straighten out," I'm so far behind that I just want the game to be over. A good game should give you ways to play your way out of a bad situation (as many do, like Bridge for example where your partnership's defensive skills can come to the fore while you wait for the "bathtub to swing your way").
Larry Levy: That's a very good term, which means I'm quite likely to steal it! :-) Dice games like Can't Stop and Pickomino are *all* about luck management. In the former case, it's just dealing with probabilities. In games like Pickomino (or even Yahtzee), there's another element: mapping out contingency strategies, where you can switch the direction you go in depending on how the dice come up. "Luck management" sounds like a very good way of describing that. It also describes how a good Bridge player plans to play a hand; combining chances in the proper order is good luck management (e.g., "I'll do this squeeze first, and if that doesn't work, I can still try that finesse").
Rick Young: So, situational luck can be a good thing or bad depending on the game's use of it and its duration. I think I'm in general agreement with most here in stating that shorter lighter games can be capricious and get away with it, but in longer games there had better be some real thought put into this aspect of the design or it will never rise above "second tier" for me.
I would also like to take this time to echo my support for the "rules of game design" put forward by Jay and Seth. I have to believe they express the feelings of most of us but yet games abound which appear not to reelect these basic truths.
I think Seth particularly made a good case for the need for some non-determinism in battle resolution which I intend so say a good bit more about in a subsequent post, and is why I find myself unsatisfied by the Bonaparte at Marengo. There needs to be an element of surprise in the possibilities when units collide which is absent there. The challenge is how best to incorporate this need...
Larry Levy: Well, it's needed if you're hung up on realism in the design. I'm not (as you can probably tell), so I wonder how often it's included through laziness or because the players expect it. There are some very good combat systems which are deterministic. The combat in Eon's Borderlands has no random elements, yet there's sufficient scope for clever play (and diplomacy) that surprises abound. I also think the purely deterministic combat in Vinci works fine. I just tend to think that designers are capable of being a bit more clever than giving us the same Resolution-type combat systems over and over again.
Rick Young: Larry sees where I'm going here when he reacted to my statement about the need for some element of chance in battle resolution systems: "...if you're hung up on realism in the design."
I suspect my background as a war gamer forms a bit of a contextual bias when it comes to resolutional "chance." Particularly with regard to war games, luck is not the first word that springs to mind when I consider battle resolution mechanisms. There is however, a great deal of room for the workings of chance.
I refer to Clausewitz who, when describing the human act of combat, used the terms, "fog and friction." Fog means that all the factors that could affect the outcome cannot be known at the outset, and friction means that even the simplest task becomes enormously difficult in the face of the chaos on the battlefield. Hence, while producing a thorough battle plan is the goal of every field commander (and his staff), even the best constructed plan, "does not survive the first shot!" Troops renowned for their steadiness suddenly become demoralized; the wind changes direction and the smoke from the canons hampers the attackers rather than the defenders; a regiment finds their avenue of approach blocked and miss crossing the start line on schedule; a key element of a coordinated battle plan is lost when a dispatch rider's horse throws a shoe and he winds up dead in a ditch with a broken neck, along with his dispatches; and on, and on...
There is no good way to simulate all of this, but there are tried and true ways of inserting such elements in war game design: command rolls (for reflecting the superior initiative of some subordinates); morale rolls (for the somewhat unpredictable nature of the human trait of "morale"); and, of course, the resolution of the battle itself with all of the myriad of ways that things can go either right or wrong.
For all these reasons, a strictly deterministic system of battle resolution does not sit well with me (Napoleon at Marengo being a good example) as the result seems too chess-like or abstract. Other games that have deterministic systems for conflict resolution do not offend to the same degree (Vinci, Game of Thrones, Diplomacy, Age of Renaissance, The Dutch Revolution for example), since I do not think of them as primarily war games, but games that involve conflict (be it military/economic/diplomatic or some combination) that need a resolution system. But when military battles need to be resolved, the way they get resolved should "feel" right or at least be consistent to the theme of the game. This requires at least some level of realism, in my opinion. And here, Larry is right...a simple coin toss just isn't appropriate.
Which brings me back to "luck management" in war game design. The system used must have some chance element but not to the point of being arbitrary or capricious. This is still a game of wits after all, but one that at the same time should reflect some of the realities of combat. When Napoleon said that, "God is on the side with the biggest battalions," he was referring perhaps to the God of War whose favour he wished to court in battle and thus avoid mere luck. Often, brute force can overcome subtlety. It is also what "mass and maneuver" is all about. One maneuvers one's forces so as to mass at a critical point and exploit a breakthrough. To a point, the more mass the better. Empirically, the minimum needed to ensure success was found to be about three-to-one odds at the critical salient. It is no accident that the 3-1 column of a classical CRT is where you begin to see all favorable results regardless of the spot on the die, and is where the luck management comes into play. You play to replicate the indicator for success on the battlefield which the CRT is designed to deliver. If you choose not to plan accordingly and try your "luck" with less (or are forced by circumstance), then you go in with the knowledge that "luck" indeed will be a factor. It doesn't mean you will lose that individual battle, but you realize that it could happen and why. The combat results table is, in the end, a very elegant way of ensuring that the "friction" of combat is reflected in a relatively realistic way given that we are talking about a board game here not a computer simulation. Yet, at the same time, it does not yield results that make you just shake your head in disbelief as I find too often happens with dice basket systems.
Is the CRT the only way to retain some element of chance without offending common sense? So far, I'm afraid so. I'd be happy to embrace a new idea here but I'm unable to think of one that doesn't offend something of what I know about combat. Maybe the infamous "cube tower" that I associate with Wallenstein but I understand is also used in at least one other game? Fog and friction all right...literally! It turns out to be a more elegant system for resolving battles than many give it credit. Your "luck" there actually can be managed after a few sessions of watching how it works and, in general, it rewards mass and maneuver but can yield the occasional surprise. But then, Wallenstein is not a pure wargame, but a game with a particularly strong war-like theme for a Euro, and thus may be why I like it more than I expected to.
Dice baskets arise from games like Risk and Axis and Allies (which are not war games either, but rather games with a strong war-like theme superimposed upon them). The idea of "hit dice" I believe comes from the miniatures side of the hobby, and pencil and paper RPGs. So melding the notion of hit dice and applying baskets of them to the business of resolving military conflict involving large quantities of troops is importing a mechanic foreign to the genre, and I have seen no treatise that attempts to claim any authenticity to its application. It's just more fun to roll a bunch of dice and look in the basket than it is to add and subtract a bunch of factors to establish a drm and then roll a single D6. Other than an intuitive realization that you need to way outnumber the enemy in a single dice basket encounter, there is no good way to estimate or manage your luck with that system.
CRTs have been criticized for being too "gamey" in that one deliberately combines strengths to achieve the best possible odds before rolling the dice (one strength point can make the difference between a 3-1 or a 2-1 battle which appears to give one little point too large a lever). I would argue that a game which allows for luck management will invariably have something you are "gaming" to maximize your chance of success. Whether it is cards, dice, tokens, or whatever resources you are given to work with, if they are resources that can be managed, a good player will be managing them to the best of his ability.
Besides, in a standard hex and counter wargame, simply do not allow sorting through stacks and you have a rough equivalent of a Columbia block game when it comes to force strengths and some of the time consuming "gameyness" goes away.
Another means of resolving combat involves the use of cards rather than dice (a pox on both your houses?) but I've monopolized this thread far too long already to get into that now...}
Dave Kohr: Actually, combat in Napoleon at Marengo is not, as RickY claimed, "deterministic". Nor is it in Game of Thrones or Diplomacy. In all those games, the result of combat depends on hidden moves or card selections made by opponents. In many battles in those games, there is no way to be sure of the outcome.
Also, I doubt the "dice baskets" in A&A, Titan, Columbia block games, and many others come from RPGs. "Hit dice" in the original RPG, D&D, refers to the number of dice a character gets to roll to determine his total "hit points" (points of damage he can take). If anything, the "dice basket' combat systems most likely derive from Risk, because it's such an old and well-known game.
Finally, there is a very good reason why many wargames use "dice baskets" to resolve combat, and why it often works much better than a CRT. The reason is that "dice baskets" can produce so many more different possible results over a much wider range, while still generating something like a bell curve centered around a "typical" result. CRTs with only a small number of dice as input to the combat resolution process cannot produce so many different possible results, and if the range of results is wide, then the chances of an extreme result will be too high. If the range of results in a CRT is too narrow, then combat resolution will become "deterministic" (i.e. more predictable), which I would agree with you is not desirable if taken to an extreme.
Alfred Wallace: How are we defining "deterministic" here? I've always thought that it meant that the outcome is determined solely by foregoing actions. Whether there's hidden information or not is of no consequence. Just because you're not absolutely sure how a battle in BaM is going to turn out doesn't mean that it isn't already determined on the board.
Larry Levy: This is an old dispute, and I doubt we'll resolve it here. For the purposes of clarity, it's probably better to divide situations into at least three categories:
1. Perfect information resolution - The player doing the action knows exactly how it will turn out before he does it. Examples include Chess, Vinci, and Borderlands.
2. Deterministic, but imperfect information resolution - There is hidden information, but it is selected by the participants, so there are no random factors. Examples include Stratego, Diplomacy, A Game of Thrones, and LOTR: The Confrontation.
3. Random resolution - A randomizing device is used to assist in determining the outcome. Examples include Risk, Axis & Allies, Wallenstein, and most classic wargames.
Dave Kohr: Larry, I think those 3 categories ("perfect information", "imperfect information", and "random") are a step toward better clarity in this thread. To continue along those lines, a resolution mechanism can have as input both imperfect (hidden) information and some random element, so clearly there is a 4th category also.
Larry Levy: I'll buy that.
Dave Kohr: I was calling "deterministic" a resolution mechanism that has no hidden information and no random element. My reason for doing so is that from the player's perspective, the outcome is certain (and in that sense "deterministic"), whereas if there is either hidden information, or a random element, a player in general cannot be certain of the outcome. It sounds like others in this thread are using "deterministic" to refer to mechanisms that have no random element, but may or may not have hidden information.
Larry Levy: Mathematically, "deterministic" means no random elements, therefore it could include hidden information known or selected by opponents. This, to me, makes the term less useful. I think the four categories we've come up with, generated by the intersection of deterministic vs. random and perfect vs. hidden information is much better.
Dave Kohr: Given all that, my own preference for resolution mechanisms in Eurogames is for mechanisms that rely just on hidden information. This is because I think such mechanisms make for better strategy games because they tend to reward good moves by players more than they reward good luck.
Larry Levy: I suspect many gamers would agree with this (some, of course, just like rolling dice). I, however, am not at all fond of games with hidden information. Bluffing bores me, as does double-guessing. I usually view blind bidding as a weakness in a game. I have no interest in detecting tells or other aspects that so many other players get into. It's just a personal preference. I make exceptions in games where the hidden information can be inferred (Aladdin's Dragons), or where some information is known (Ys), or where the hidden info has only a limited effect on resolution (Game
of Thrones). But for the most part, hidden info is a negative to me. In the most extreme cases, it seems exactly like a random element from my point of view. This isn't to say that I must have perfect information in a game. As I mentioned earlier, I very often enjoy games with Situation luck, which frequently involves hidden elements.
Dave Kohr: In wargames, my preference is for resolution mechanisms that rely on both some hidden information, and some random element, but where the random element rarely dominates the outcome. Probably the Columbia block games come closest to my preferences: there is hidden information in the form of hidden units, and a "buckets of dice" combat resolution mechanism that produces a typical range of results that experienced players can predict, while retaining the possibility of producing a more extreme result. I find that the possibility of those extreme results tends to encourage more realistic strategies (usually, greater caution and more planning for contingencies by keeping units in reserve, etc.). It also increases the fun factor for me.
Shin Yoo: My thoughts…
1. CRT, chess or dice bucket: To me, it seems impossible to deny the fact that on the whole, dice represent an even luck (unless they are physically defect in some ways). CRT is a way of imposing certain ranges of possible outcomes to the possible dice rolls, using specific statistical data. It IS a high level abstraction, but so are all games, wargames or not. This method itself sounds logical enough to me. What we can debate is whether a certain CRT is historically precise enough or not.
Chess-like resolution is interesting for a game's sake, but certainly this is not always the way things happened in the real world. If the game is leaning toward simulation aspect, the application of chess-like battle resolution will be highly limited, i.e. only to particular events where the outcome was always determined by sizes of armies and such.
Dice bucket is the wildest beast among the bunch. It is downright pure chance. Well, CRT is pure chance too, but you can always add more dice to enlarge the range of outcomes and raise the degree of precision. Same thing is possible with dice, but if it gets too complicated it's like you're using an imaginary (and simplified) CRT.
Personally, I don't have any fixed preference. It really depends on a given game. But I will admit that I've never played any historic simulation game using deterministic battle resolution (e.g. Bonaparte at Marengo), so I do not have clear idea how it actually feels. But then again, each game can aim different defree of abstraction, and this results in different user experience. I guess you can totally feel like a general at Marengo, in a certain level, without throwing any dice.
2. Man, Play & Games by Roger Caillois: This argument over luck brought to my mind this wonderful book. This can feel a bit too academic, but once you start reading it you will be surprised how well the writer's arguments fit to modern commercial games. Basically, Caillois departs from the definition of game by Huizinga, and goes deeper into actual games. At the heart of the book lies his categorization of games: Agon(skill), Alea(luck - yes, that's the meaning of the original Latin), Mimicry(simulation), Illinx(vertigo). Agon is a game of skill with no luck, whereas Alea is purely based on luck (lottery). Agon rewards the idea that more skilled men will succeed in life. On the other hand, Alea encourages the idea that men are all equal under the pure luck, and it compensates the lack of skill in some ways. Agon and Alea are thought to be two opposite extremes, which essentially forms the basis of the arguments we're having everyday at BGG.
Mimicry is the instinct to mimic someone else (RPG), and Illinx
is the thrill of physical movement(rollercoaster). I'm quoting him in a terribly rough way, but according to the writer, Alea teaches us to obey our fate. Well, in general, playing games teaches us to try our best and obey the result. So, we can try hard to obtain a balance between chance and skill in a game, but I'd like to think that what is random is random after all.
There, my two cents. I again highly recommend the above book - if you're serious gamer, it's a great read.
Ryan Bretsch: I like luck in a game.
In fact, I LOVE luck in a game and it makes little difference to me what type of luck it is. I see luck as an adaptive factor to strategy that randomizes the game to keep it "fresh". When done correctly, luck in a game is something a player must be able to account for intelligently... in order to win the game and to keep things interesting along the way. When done correctly, luck gives a lesser skilled or lesser experienced player a shot at winning. And that's important. Remember when you had your old Nintendo video game system from the 90's? You're playing Madden 98 with one of your best friends and he is clobbering you because he has mastered every nuance of the game? And then he brags to you that it is his 53rd victory in a row against you?
A lot of my friends are not big boardgaming fans in their own right. But they play my boardgames because they know it will be fun... a good time had by all, if you will. And they play because there is a good chance they can win. That makes it fun for them. And why not? Finally, when "luck" is done correctly, it is a GREAT socializing factor in a game. That is an important factor to consider too.
Rick Young: I'm sorry if I didn't explain the terms correctly (for years I had the def'ns backward in fact). Deterministic systems are fixed with regard to the result, there is no reference to some mechanic to find out what happens whether it be cards, dice, cube tower, etc. In Chess, a piece in the correct position takes another piece. Non-deterministic systems also have a defined process, but before the outcome is known, some randomizing process is added to yield the final result.
Larry has summed it up nicely and I would offer the ideas of fog and friction again as well. In Chess there is neither fog nor friction. Both players know where all the pieces are and what they can do. It is purely deterministic and quite abstract. A lot of abstract games also fall into that category such as Go, Checkers and Othello. Some conquest themed games such as Vinci, Civilization and Age of Renaissance also work that way.
In Napoleon at Marengo there is fog but no friction. You do not have a clear picture of the nature of the forces you are up against but when units do battle, the results are pre-determined. Stratego also works this way. Other examples include Diplomacy, Dune and Game of Thrones. The addition of "event" cards or other devices to modify the result is part of the fog but not the friction, if I can put it that way. Classic hex and counter war games had little or no fog, but some friction in the way battles were resolved usually by employing the dreaded combat results table (CRT). Examples abound.
Some games employ both fog and friction, particularly good examples of which are the Columbia block games. Your knowledge of the enemy's disposition is shrouded by uncertainty and the results of a battle are further subjected to a randomizing mechanic (hit dice in this case).
While I appreciate the need for some chance to be at work in battle resolution for war games, what I don't like is a mechanic that tends to dominate the exercise and certainly can be a question of degree and personal preference. I hated Risk because my experience was that it seemed to always come down to one monster army trying to wear down another in an endless dice-fest that could only inflict casualties two at a time. The mechanic got in the way of the fun, realistic or not. As Shin pointed out, the CRT placed boundaries on the workings of chance based on strength ratio comparisons, which hits my "sweet spot" and, if for no other reason than it was what I grew up with, just seemed to make sense. If the more chaotic nature of dice-baskets hits your sweet spot, and makes a game more fun then more power to you. However, for all of the chaos, fog and friction, there is a sort of calculus of the battlefield that should be respected and I tend toward games that do that for me.
David Kohr: Rick, I disagree with you on two points.
First, event cards most certainly can function as "friction". When my units attack another player's, I may not know whether they hold an event card that can bolster their defense, and I often don't know for sure whether the other player will play the event card. The effect of this hidden information on my strategic planning is often very similar to the effect of a random element in combat resolution.
Second, the "dice basket" mechanic where every unit gets to fire is often *less* chaotic than a typical CRT. The reason is that extreme results are usually less likely. Take the example of the classic 3-1 odds attack. On a typical CRT, with a single d6 as the random element, there is nearly always at least one result where the attack fails. This is an extreme result. Yet it comes up 1 time in 6, which is relatively frequently. An analogous attack in Risk or a typical Columbia block game has a smaller deviation of results--the extremes come up less frequently, meaning there is less chaos.
Wargames with CRTs were among the first hobby games I ever played, but over time I've come to appreciate that a CRT is just a crude approximation, and the "dice basket" mechanic of, say, the Columbia games is better in just about every way, except that battles take longer to play out.
Seth Owen: I think there's general agreement that chance is an appropriate element for wargames, particularly those that attempt fidelity to some historical situations. While the degree that chance can vary considerably, wargames that lack chance elements tend to feel too chess-like to be authentic battle recreations. Every great theorist of war from Sun Tzu through Napoleon and Clausewitz to Fuller and Hughes has pointed out how friction, fog and the fickle finger of fate must always be accounted for. I think arguments can be made for both the "dice basket" approach and the CRT on both simulation and playability grounds. Picking the right combat mechanism is an important design judgment right up there with turn sequence, zones of control, scale and other fundamentals. A design mistake at this level will result in a game that's either broken or an obvious failure in realism.
As an aside, I think the deterministic battle resolution in Bonaparte at Marengo is interesting and works well within that particular game and that specific historical situation. Its wider applicability is questionable.
The more intriguing question, I think, is what role luck should play in games that are NOT attempting to be historical battle simulations. Here the designer confronts a fundamental philosophical question between determinism and fate. Is the player the master of his own future and actions or is he subject to the capricious whims of fortune? Other posts have noted the classic scope of this question. The 'right' answer will depend on the culture and the personality of the player.
In some cultures games of pure chance or that are heavily influenced by chance are popular. On the one hand, people are supposed to learn to accept their fates. On the other hand, these games are egalitarian because luck doesn't favor one person over the other. Some people will rage at their ill-treatment by the god of fortune (dice, cards, etc.) but others will find themselves comforted that they lost because the other guy was simply 'luckier' not 'better.'
Other games of pure skill will be popular in cultures that believe in the power of human action. I don't think it's accidental that the "scientific' Communists in Soviet Russia were such big fans of chess. Likewise, many players will absolutely despise losing due to luck, preferring to believe that they lost fair-and-square to the 'better' player. Some games seem to strike a good balance between the two points of view, such as poker and backgammon, providing a mixture of skill and luck that seems to appeal to both tastes.
My personal preference, in non-wargames, depends on the 'frivolity' of the game. I like quick luck-dependent games, while quick, light abstracts seem too dry and predictable. On the other hand, if I invest a lot of time playing a non-wargame, the last thing I want is to have the game decided by luck.
Rick Young: Well, I think we all agree on far more than we disagree on, and we are homing in on personal preferences now which is fair enough, as long as the language and rationale we are basing our perferences on are clearly understood (which in my experience, less disciplined discussions have not always taken the trouble to do). However, I still have a little difficulty when hearing people argue that dice baskets present better statistical averaging (less chaotic) yet at the same time allow for more radical results (more chaotic). If that can be true, no wonder I don't care for them. I don't see the "luck management" possibilities there.
Dave, if one prefers to think of a hidden player input as friction rather than fog, that's fine...the effect is similar enough not to get hung up on. Personally, I categorize imperfect intelligence as being a part of the "fog" of conflict which extends to the intentions or actions of your opponent. A "gambit" in Chess is a one of the few devices that players of that particular game have to disguise their intentions but does not change the deterministic nature of the resolution; but, that degree of nuance is starting to be like debating the number of angels that can fit on the head of a pin. The point is that there is both perfect and imperfect intelligence and random vs. non-random means of resolution. The way games handle these elements is what we have been discussing and can be why we prefer this game over that, along with many other factors of course.
And still addressing Dave K's latest observations, I'm not sure if you don't mean the 2-1 column which traditionally had a one in six chance of going against you (A Elim). In my experience, the worst that can happen in a 3-1 battle is an Exchange meaning you both lose an amount equal to the smaller of the forces. It's not the result that you hope for, but is still a "win" in gross terms. That was the whole point of the CRT's design - to mirror commonly held military doctrine.
However, CRTs themselves underwent a great deal of modification over the years - some introduced a 3-2 column to provide an additional level of granularity between 2-1 and 3-1. Others went to a 2D6 matrix which allowed for even finer tuning. I still maintain that all had the principle characteristic of luck management inherent in them.
There are many decent games out there in which the dice-basket approach works just fine. Axis and Allies suits that mechanic, being a lighter game in the first place; but, as an aside, in that genre I like the approach to combat employed by Eagle Game's Attack! (expansion version) as offering something a little newer and fresher in concept. On the other hand, I don't think that a more serious treatment of the same period such as Panzer Group Guderian would fare as well with the dice basket system. As others have said, the intent of the game should dictate the method of incorporating chance.
It seems in recent years that there has been a tendency to "throw the baby out with the bath water" by replacing the venerable CRT with dice baskets in games where the application just doesn't seem to fit well. An example of that may be found in the Games USA's "Eagles of the Empire" series of Napoleonic battle simulations, which when taken over by Avalanche Press, was re-tooled to incorporate a dice basket battle resolution system vice the original CRT approach. A designer's note explaining the rationale in a game meant to be more of a simulation would have been extremely welcome. As it stands, the series is still worthy of play but the mechanic rings somewhat hollow for me absent of a rationale other than, "that's the way it's done nowadays..."
I totally agree with the earlier sentiments that shorter, lighter games can get away with a lot of randomization, whereas longer more serious games need to be more careful. Shin's observations regarding some randomization element to facilitate replayability is important here as well. The better game designers take this aspect very seriously while others seem to throw in random elements as just more chrome.
I readily admit to being fairly dogmatic about the appropriateness of design mechanisms used in "meatier" war games, but I can also enjoy a good game of Risk: 2210 just as much, now that the more hideous aspects of original Risk have been fixed...
Pierce Ostrander: I'm going to address this issue from a "wargamer vs. eurogamer" perspective. Eurogamers seem to have a MUCH lower tolerance for luck in games than wargamers do. Most wargames are trying to provide a theme-rich simulation experience. Most Euros emphasize the game experience - with Euros "the games the thing" with wargames "the themes the thing". A person who values both has the preferential flexibility to change his judgment criteria as he sits down to games of each type.
One of the main differences between war games and eurogames from a game-play perspective is: the amount of randomness. Wargames tend to have a lot of it, and rely on it to provide "realism". (reference Alfred's opening article).
Eurogames tend to add just a bit of luck to make the game somewhat more variable: random chit draws in Puerto Rico and Tigris and Euphrates, random goods cube draws in Age of Steam... or to have no non-player created variability at all like Maharaja and Lord of the Rings: the Confrontation (except for the order of play at the start of the game). Even small amounts of luck are not preferred by a "pure" eurogamer (reference: Morgan's comments about T&E, and also the comments that appeared with his recent "3" rating for Parthenon).
Some eurogamers will say that even a game like Maharaja has way too much randomness… player-generated though it is (often referred to as "chaos" rather than "luck"). Sometimes, you simply can't predict what another player will do, especially if they are inexperienced in a particular game, and for some this is just too far out of their control to make the game "fun".
Jay, Hammer of The Scotts is outrageously random. It is one of the most random wargames that I have every played. There are others that are less so, but 99% of all wargames published have significant random factors built into the game.
I used to be exclusively a wargamer. I have become a eurogamer. My values changed. I now prefer games with much less luck and much more control than I used to. My tolerance for luck has diminished, but I don't think it will ever reach "zero" – I've never been a fan of 2-player abstracts. I played HoTS about 3 times, and quite frankly, it was WAY too random for me. However, I didn't let that influence my rating. I guess you could say I put on a different "expectations hat" when I sit down to a wargame.
I brought up Parthenon because in a way, it is a "crossover" game from a theme / luck perspective. Manilla, for example, was a Euro that you EXPECTED to be random when you sat down. Parthenon on the other hand, did not have a warning label attached. It is quite random. A eurogamer sitting down and expecting a "T&E" type experience will be sorely disappointed. However, I quite enjoyed my one play so far… primarily because the game does such a good job thematically! In this way, it is akin to the wargame experience rather than the eurogame experience. Theme and luck trumps controllability and "pure" game play. I guess in my view, it all comes down to the expectations and preferences you have when you sit down at the table. Knowing your own will help you choose games to play and games to avoid… or at least help you predict how you are likely to respond to the experience.
Larry Levy: I'm not a wargamer and I wouldn't know a dice basket from a bread basket. But your mention of Risk brings up a pet peeve that I rarely see mentioned. (Then again, maybe it doesn't bother anyone else.)
The combat resolution in Risk is quite clever, as it uses simple rules and a small number of dice to give appropriate odds during battles. But I find they have one great flaw. Many of the results cause one side to lose units, while the other side loses none. This can lead to very extreme results. I have witnessed (and I'm sure all of you have as well) a defending force of one or two units decimate an attacking force of ten or more, while losing no units of its own. This adds considerably to the variance of attacks, which makes the game even more luck-driven than expected. As a result, I gave up playing Risk while still in my teens, and I have no real desire to go back.
I think the game would be greatly improved if, instead of combat results of 2-0, 1-1, and 0-2, the only results were 2-1, 1-1, and 1-2. That is, either one side loses two units and the other side loses one, or they both lose one. This would guarantee that, no matter how
poorly one rolled, you could predict the maximum loss you could absorb and still take over a territory. This would assist with planning and make the game far more tactical.
I realize that you are all saying that extreme results should be part of a combat resolution system, but I still think those are possible with what I propose. Risk is a game of attrition. Losing more units than anticipated can lead to the inability to capture one or two additional territories, which can be crucial. A steady stream of 1-2 results sounds pretty extreme to me. With the given rules, you can see the equivalent of AE results on attacks of 5-1 or more. There's such a thing as being TOO extreme.
Anyway, that's my Risk rant. I came up with a variant that meets these requirements of mine a while ago, but have never had the chance to try it out. Somehow, I doubt I ever will.
Dave Kohr: When it comes to "luck management", one of the big benefits of the "dice basket" resolution mechanic compared to a CRT is that, over the course of a game, you make a lot more die rolls. Each die roll tends to be much less important to the outcome of its battle than the few dice (or single die) you roll when using a CRT. This means that, while you may be hosed by the dice in a particular battle, it's much less likely that you'll be hosed by the dice over and over again in different battles. In this respect, "dice baskets" are less chaotic, as Rick said.
When it comes to the range of different possible outcomes, CRTs are usually pretty limited. The classic CRT that uses a d6 is an extreme case: there are only 6 possible outcomes. (I'll grant that it's possible each outcome could be applied in different ways, such as choosing one unit versus another to receive a step loss--although in my experience rules such as that usually give one side or the other an unrealistic amount of control over the outcome of the battle.) Since real battles typically have many more possible outcomes, this is unrealistic.
Meanwhile, most "dice basket" systems allow for a wide range of possible results, in terms of how many hits a unit takes (and in some variants, "hits" may have other effects besides causing losses, such as forcing retreats). "Dice baskets" can also allow for extreme results, and automatically assign an appropriately low probability to their occurrence, while such extremes can't even appear on a CRT due to its limited range of random inputs. So in this respect, "dice basket" systems are more realistic than CRTs.
Rick says: "I don't think that a more serious treatment of the same period such as Panzer Group Guderian would fare as well with the dice basket system". This game is actually an excellent example of one serious problem with CRTs. It heavily rewards "factor counting", in which the attacker tries to get just enough points of attack strength to achieve a certain odds column (such as 3-1). And in terms of simulation value or rules weight, there is really nothing more "serious" about PGG, with its CRTs, than the newer WWII operational-level game "Tigers in the Mist", which uses the "dice basket" mechanic. And I would say PGG is somewhat less serious than Columbia's "East Front", which also uses the "dice basket".
Rick Young: I'm not sure where it was that I would have said that. I think I've been pretty consistent in stating that I think them more chaotic than I would like (a lot more, actually). And unrealistically random to boot (but I admitted to be "hung up" on a modicum of realism when it comes to war games). Pierce Ostrander seems also to believe them to be pretty chaotic: "Hammer of the Scots is ridiculously random," and Shin Yoo stated that, "Dice bucket is the wildest beast among the bunch."
Dave, I think I can show where you too have stated that dice baskets are more chaotic (but that's why you like them); however, in a couple of recent posts you find they are less chaotic than CRTs (but that's why you still like them apparently). Seth stated that while "fluke" results can happen with dice baskets which is apparently a good thing, CRTs are more prone to "luck problems" (which is something different than allowing flukes I guess) and that's a bad thing. I have also said that if dice baskets can be both less chaotic and more chaotic at the same time (or would it be at different times?) then I would have a problem with the mechanic for that reason alone. So, for that reason, and the fact that I don't accept the accuracy of the results you get on a statistically demonstrable basis (which CRTs can, dry and boring as it may be), I can't take their results seriously in a game that is supposed to be more serious.
Flukes I grant you can surely get with dice-baskets but would argue that "fluke" battle results occurred less often in history than has been implied. If these fluke results occurred only once in a blue moon (which is historical), I could get past it, but they seem to happen just a little too frequently for my taste.
The "luck" complaint with the CRT in Africa Korps usually refers to the low odds battle that inevitably takes place at Tobruk, but what was it I said about odds battles? How would the dice basket mechanic change this without departing from the near-run aspect of the campaign in general, and that battle in particular? Or does the feeling about AK result more because of the way the availability of supplies was modeled in that game (which accounts for most of the "luck" in my opinion but can't be blamed on a CRT)? If you are referring to the lower unit counter density, resulting in fewer but more decisive individual battles, then by all means you can criticize AH for using the generic "one size fits all" CRT in their early classics. Later games would improve on this while not disposing of the inherent applicability of an "Odds-based" battle resolution technique. It goes back to the intention of the game design as I think we've all agreed.
David Kohr: "Dice basket" systems certainly can be analyzed statistically to determine the average outcome and probability of extreme outcomes. It simply takes more effort than CRTs because the random input is more complex. Actually, I regard this as yet another disadvantage of CRT-based systems: they tend to give the player too much knowledge about the likely and possible outcomes of a battle. {/color]
Seth Owen: From a simulation standpoint both approaches have their merits. As pointed out, "dice baskets" allow for fluke results (which is historical) while mitigating the overall influence of luck. On the other hand, not all die rolls are of equal importance, and it's still possible for a key roll to go against you, even in a dice basket game. CRT games are more prone to luck problems, especially if they have relatively few rolls. (Afrika Korps is a famous example of this).
On the other hand, dice basket games are essentially "firepower" based, with the amount of damage based on the number and effectiveness of the firing units and with a limited ability to cope with situations where there is a big differential in protective strength or historical situations where massing force on the tactical battlefield is difficult.
Dave Kohr: Absolutely not true. "Dice basket" games often use special rules for such situations that are very similar to those of CRT-based games. In the case of protective terrain, the attacker's to-hit number may vary based on the defender's terrain. As for the difficulty of massing forces, "dice basket" games often use stacking limits just like CRT games. Many games go even further. For example, many Columbia games permit only a limited number of units to cross a hexside, or border between areas, to enter a battle.
Seth Owen: I didn't say inability, merely a limited ability. There are work-arounds. Rommel in the Desert, for example, forces units to attack certain kinds of units if present. In War at Sea ships have to match up against opposite numbers and then the excess can concentrate fire. Hex-side limits and the such also help restrict the attacker's options somewhat, but within each individual engagement the dice basket mechanic is still firepower based. Damage is based on the number and quality of the firing units, with only a limited ability to account for things such as armor and forts. Some of the block games, such as Hammer of the Scots, EastFront and Rommel, resort to 'half-hits" and it helps, but CRT-based games have more flexibility in accounting for defensive strengths.
Dave Kohr: My point is that CRT-based games and "dice basket" games often use basically the same mechanisms, so "dice basket" games are not any more limited in their ability to represent these situations. The only major exception I can think of is that CRTs often use a "column shift", which IMHO is a terrible kludge that does not make them better able to represent the effects of terrain, it just makes resolution a little faster.
~FIN
And that's it for this time, folks! Hope you enjoyed it!
Musings On...
Edited by Tom and Laura Vasel
October 17, 2005
www.thedicetower.com
| Back to the
main page! |