Analysis: Carcassonne

The following is an account of a board game play and analysis excersize for TC 455 - 3D Game Design. I and a team of three others were assigned to play a board game together for at least two games and provide an overview and in-depth analysis of its rules as well as an account of how it might be adapted to digital format.

The game we chose to play was Carcassonne by Rio Grande games. It's a strategic land acquisition and resource management game whose name is derived from that of a fortified French town whose settlement dates back as early as the Roman Empire, when the strategic significance of its hilltop location was first identified. For many centuries Carcassonne had a reputation for resilience, consistently proving impossible to invade or besiege except through political means, making it an apt title for the board game as the player who comes to dominate the town and its resources wins without ever having to engage the other players in combat.

The board game itself is set in the medieval period and centers on a battle of territorial acquisition, with players vying for domination of towns, cloisters, fields, and roads. Players begin with a starting tile and each take turns drawing a new land tile from stacks around the board, placing them adjacently to one another in such a way that any road segment that clears the margins of a tile continues across the next and so forth with city segments and fields as well. When a player places a tile he or she may place a follower somewhere on that tile. Depending on where the player chooses to put the follower it becomes something different: a knight if it's in a city segment, a thief if it's on a road segment, a monk if it's placed in a cloister, and a farmer if it is placed laying down on a field. Each player is granted eight followers each, one of which is held in reserve as a piece to move along the scoring board. The remaining seven may be placed during the game and return when a city, road, or cloister is complete but are effectively removed from a player's pool for the duration of the game if they are placed as farmers as farmers are permanent fixtures on the tiles they are placed. All followers are color-coded to their players for ease of differentiation.

It is important to note that the contiguousness of segments of land applies not only to a player's options for placing segments but to his or her options for placing followers as well; players cannot put a follower on a piece of land that is contiguous with one that already has a follower on it regardless of whether it's one of their own followers or someone else's and regardless of distance; the entire contiguous stretch of land belongs to that follower. It's possible to have more than one follower on a stretch of land by clever tile placement; for instance, if a player places a city segment, then another player places another city segment with a gap between it and the other one, and then one of them fills the gap later, they may both have legally placed knights in the same city. This applies to fields and roads as well.

Points are earned during the course of the game as players complete cities, roads, and cloisters. The points for any of these vary but go to whoever has the most followers in a city, or on a road. A cloister can only ever have one follower since cloisters don't continue across multiple tiles. In the case that two players tie for number of followers in any given city, on any particular road, or supplying any particular city, they both are awarded all of the points. The number of points is 2 per tile in a completed city (plus 2 more per each segment that bears an emblem) or 2 per tile on a completed road; cities are completed by closing their boundaries off with walls and roads are completed by ensuring that both ends of a road end at either a crossroads or a city. Cloisters are completed by ensuring that there is a land tile occupying all eight spaces surrounding the cloister regardless of what kind of land it is, and 9 points are awarded for completing the cloister. Again, when a city, road, or cloister is complete, the followers of all players occupying them are returned to them that they may place them elsewhere. At the end of the game, when all tiles have been laid and no more turns can be taken, points are awarded for farms and incomplete cities, roads, and cloisters. A player receives 4 points for each city that their farmers supply regardless of number of farmers supplying a city or the city's size; a city is said to be supplied by a farmer when there is a farmer occupying a bordering field, regardless of the farmer's distance from the city. Farms are separated by roads. Again, the player with the most followers supplying a city scores the points for that city. Each city should be considered individually and with all borders taken into account. Incomplete cities net a single point for each tile in the city, incomplete roads net a single point for each segment, and incomplete cloisters net a single point for each tile surrounding the cloister as well as for the cloister tile itself. Whoever has the most points at the end of the game wins.

The game's rules, although they seem complex and lengthy on paper, are fairly simple in practice once all players are versed in them and provide for several interesting dynamics. The larger the city or road the more points it's worth, and the number of city segments with emblems insures that a very large city would be worth quite a lot of points; one such city was worth 38 points in the game played in our group. Meanwhile farmers get more points for the quantity of cities as opposed to the quality, so it's more advantageous sometimes to complete many two-tile settlements as opposed to a single large settlement. What I found is that if a player places a cloister, the other players will likely avoid completing it as much as possible as it is worth a relatively large body of points without any chance for a fight. It's possible to use predictions on what people will and won't be predisposed to complete in order to control the board and insure players shift attention to one side of the board or another, especially through strategic placement of cloisters. Aspects like this contribute a value beyond point value; a cloister actually requires much more space than other kinds of tiles in order to score points and will score comparatively fewer points than a city with nine tiles, but the strategic value of a cloister can be invaluable.

It is advantageous to a player for him or her to be the only person occupying a city and to try and isolate other players' followers. Yet two players may cooperate in the event that other players purposely try to minimize or avoid completing a city and form a truce that they might both obtain a mutual lead by tying for points or helping complete one another's roads or cloisters in the process or awarding dominance of surrounding farmland to a cooperative player. The aspect of contiguousness insures that players are able to make extremely far-reaching plans, gambling on the idea that a piece ten turns away might seal their dominance on the board. In this respect it's a game about thinking ahead much in the way that chess is except much less obtuse in that it's easier to predict that a city will be completed than it is to predict on the first turn of chess what the other player intends to do with their next ten turns.

The first game we played began relatively smoothly. After about five turns all four players were well-versed in the rules although questions about strategy pervaded throughout the game. I was very involved in the game having been designated its keeper and having been the one to read the rules and explain them ahead of time. The other players' levels of interest were considerably lower, with one member doing something else on his laptop throughout the game and not looking up but for when his turn came around and a consensus among the other team members that Age of Empires seemed like a suitable digital adaptation and improvement on this game. The team member that was on his laptop maintained a negative attitude with respect to Carcassonne's lack of violence and combative situations, completely missing the point as he spent most of his turns early in the game placing farmers despite my advice that farmers were better suited to the final turns when cities had already been blocked out. Nevertheless he was able to construct a strategy around his early farmers that cemented him as having a monopoly in farms throughout the game and put a heavy emphasis on the number of cities being completed. He had the red pieces. The other members had blue and green respectively, and I had black.

The green player was the first to draw a city tile and thus claimed the first city, which he built throughout the game to such a size that he accrued 38 points from it alone. He, like the red player, claimed disinterest in the game, but that didn't stop him from maintaining a lead with his cities from mid-game on. He and the red player found a cooperative strategy; it was in the red player's best interest for him to have as many cities as possible and would often try to help him complete them. The blue player was slightly more optimistic about the game than either of the others and displayed more interest in exploring its rules and possibilities, trying to diversify his strategies and claiming many roads, a handful of cities, and aiming to compete for farmland with the red player. He took the early lead with the first completed cities and roads and maintained it until the green player's large city was finished. Meanwhile I took a misanthropic strategy, knowing that players who were trying to acquire farmland and city space so aggressively would be difficult to compete with, and placed all three of the game's cloisters on the east side of the map to push the other players off to the west, where the red and green players built small cities and roads that went nowhere for most of the rest of the game while the blue player and I completed roads on our side. I purposely complicated the act of completing the early cities, which were by far the largest, by placing tiles such that only extremely specific ones--a corner with a road going past it on specific sides of the tile, for example--could be selected to finish them. I wasn't able to stop these cities from being completed as I eventually had to help finish them myself to gain more points for my cloisters, but the delay and distraction helped me to do what I wanted without anyone attempting to directly compete with me. By the end of the game the red player had caught up with the blue player, the green player had a large lead, and I was between them. When the endgame points were totaled, I seized an enormous lead thanks to my claim over large amounts of unfinished roads and cities and the points I was awarded for my cloisters, ending at over 60 points. The blue player took second, having as many other resources as me and a few farms but lacking the cloisters. The green player hadn't paid much attention to anything other than cities and consequently didn't move much farther than his midgame lead had gotten him. The red player lost by a wide margin as he'd exhausted his followers on farms. The second game went much like the first. We decided against using the included river expansion for the sake of having consistent rules to write about. I didn't bother pointing out strategic flaws to my teammates and merely manipulated the board to the best of my ability, once again claiming dominance over the cloisters early on while much more mindfully keeping the green and red players under control and more directly competing with the blue player. I won the second time as well.

I enjoyed playing Carcassonne and would like a chance to play people who are less ignorant and more open-minded. I enjoyed playing the manipulative villain and turning my opponents' strategies inside out. Despite my teammates' relative disinterest in the game I enjoyed their company and the farming-related jokes. There wasn't a lot that I disliked about it. I feel that the core game has an exceptionally good balance, though it seems that the strategy involved tends to go over a lot of peoples' heads. I think my teammates probably felt like the game was too mechanical and self-playing in the way that the board filled itself in and weren't all that aware that they were actually employing strategy. They certainly weren't aware that I was.

I'm not sure I would change this game given the opportunity. As I said, it seems very well-balanced and very thoughtfully designed. The river expansion's rules offer some interesting implications in that it requires that the river pieces be placed before everything else and therefore dictate the shape of the board much more strictly early on than in the normal game, but in a game this strongly defined it doesn't seem like an improvement so much as a way to mix things up and to add additional challenges to players' strategies. To provide a digital gaming analogy it's like someone playing a game of Ratchet and Clank and restricting himself to using only the lowest-level weapon that he has in his arsenal, not allowing himself to upgrade his armor, or only permitting himself to use Ratchet's wrench. It doesn't necessarily improve the game so much as provide a different experience for experienced players. The key term is "experienced" as players need to understand the basic rules before they can begin to understand what changes new tiles would bring. On that matter, it would be a poor idea to try to introduce new kinds of resources to the game as the balance has already been struck between them. It would not be a bad idea, though, to add tiles and pieces that modify the value of the existing resources; livestock to enhance farms' value would've made the red member's strategy much more viable, for instance.

Many of these expansions already exist, though. In fact, there are over a dozen, and it wouldn't be proper for me to assume credit for their innovations. I did, however, devise a few interesting variations just based on the interactions between players. Two players might share a score, for instance, enabling them to think cooperatively. Another possibility is an entirely cooperative version of Carcassonne wherein all players on a board work to try to maximize their shared point total at the end of the game. One complex variation I thought of involved several separate games of Carcassonne being played at once with the boards "bridging" over onto one another's edges; the games would be treated as separate games until separate territories connected, at which point they would be able to play as if the game were taking place on a single, large Carcassonne board. Even if they didn't bridge, the players' scores would all be compared at the end of the game.

A digital adaptation of Carcassonne exists--on Xbox Live Arcade, to be specific. It maintains the rules for placement of land and followers, its primary additions being that the tiles are represented in 3-D with models of castle walls to help players visualize the board, stronger randomized drawing of tiles (even a computer randomly choosing them from a set pool is superior to the multiple, messy stacks that we used), and AI for single-player play and practice. The AI uses algorithms that help it to make the most of every tile and every follower it places. Unfortunately this digital adaptation--apart from in its multiplayer aspect--can't simulate the ability to cooperate with the player as he or she finds himself unable to communicate or reason with the computer. Even if there were some form of system to help manipulate the AI to cooperate--apart from a very strict "team" system where two players share a score--there would be no way to enforce the player's end of any deals that he or she makes with the AI. Additionally AI can't reason as complexly as a human being and at best can be programmed to place tiles aggressively and use pre-defined algorithms to insure that what it places it can complete. It can't tell that a player is going to avoid completing a cloister and use its placement as a distraction; the concept of "distraction" doesn't mean anything to it. This isn't necessarily a major shortcoming as far as the adaptation goes as board games are by nature about social interaction, though this does beg the question as to whether internet-based interactions can be compared to the face-to-face contact of the physical board game. That's a topic for a different discussion, though.