Leo Colovini's Cartagena is one of the most elegant race-games ever designed. Let's see how it fares under the vivisectionist's knife.
The rules to Cartagena can be summed up in a few sentences. Play a card to move one of your pawns to the next unoccupied space of that type. Slide one of your pawns backwards to an occupied space to draw more cards. If you get all your pawns to the boat first, you win.
Here are the most "complex" rules in the game:
The game would be even simpler if these rules were eliminated. For instance, you could have a hand of four cards, and on your turn you play a card and then draw a new one. However, I don't feel that the game needs to be simplified, and these changes would probably reduce the game's deliciousness to an unacceptable degree.
The only somewhat inconsistent rule in Cartagena is the "population-limit" rule. Spaces which contain three pieces don't function exactly like other spaces. I'm not crazy about this rule, but it's not really the inconsistency that bothers me. It's the coerciveness. See the section on emergence for more thoughts about this issue.
Cartagena has almost perfect clarity. It's extremely easy to read the board and see what your different options are, and to see how each option will set up the following players.
The only mildly opaque element of the game is the discard pile. However, this opacity almost never matters in practice. You could reduce it (but not eliminate it) by making the discard pile "open", but, as a player, I doubt that I would take advantage of this fact. Even if I plow through the discard pile and determine that there's only one "skull" card remaining, I won't know whether it's still in the deck or if it's in one of the player's hands, so the information really doesn't help me much.
Like most other games that use cards, Cartagena contains a good amount of both chance and hidden information. The fact that players have hidden hands of cards means that you can't be sure what moves they're going to be able to make on upcoming turns, and the fact that you draw cards from the deck means that you can't be sure what moves you're going to be able to make on upcoming turns.
Cartagena incorporates one of my favorite design elements: a modular board which generates a different setup in each game. The variable path will contain "pinch-points" (where two identical symbols are very close together), and "long jumps" (where two identical symbols are very far apart). This creates an interesting texture for each race. Add to this the positions of player's pieces along the path, and the particular combinations of cards that you might be holding in your hand, and many different tactical situations can arise.
Despite this variety, there does seem to be a kind of higher-level uniformity to the game. From a distance, the tactical situations that arise all feel a bit similar. Furthermore, there aren't really multiple overarching strategies to try. Racing straight for the goal is almost always a losing proposition, so winning usually requires a tactical balance between surging forward, hanging back, and picking up double-card draws.
Cartagena exhibits no serious stagnation problems or end-game problems. Occasionally it will be obvious that one player can win on his or her next turn. However, in such cases, there's usually at least one other player who might win first, and in any case, the game will be over in less than a minute at that point.
I can see one technical flaw. It seems possible in principle that you could wind up with no cards in your hand, and no legal way to move a piece backwards to draw more cards. This scenario is simply not covered by the rules, so technically, the game is broken here. In practice, I've never seen the scenario even come close to actually occurring.
From the player's point of view, Cartagena exhibits an excellent level of emergence. The most obvious emergent effects are the "long jumps" which are created when strings of identical symbols are occupied. Players spend most of their time focusing on such emergent effects.
From the rules point of view, the "population-limit" rule is a bit "coercive". I understand the rationale behind this rule. Without it, pieces would tend to pile up onto particular spaces, and players would keep dropping back to those overly-populated spaces in order to draw more and more cards. Nevertheless, I would prefer to solve this kind of problem in a more emergent and less coercive fashion.
For example, consider this alternative rule: each space may only contain one piece of each color. Therefore, when you slide a piece backwards, you must put it onto the first occupied space that doesn't contain a matching piece. This creates a more emergent (and perhaps more interesting) "population-limit", rather than the fixed and somewhat arbitrary limit of three pieces. A potential problem with this suggestion is that it seems to scale badly for different numbers of players. In a five-player game, it's possible to draw four cards at once, whereas in a two-player game, you'll never draw more than one card card per action. Perhaps in the two-player game, each player could control two different colors.
Another possible avenue to explore would be to have no population-limit, but to provide some reason why players wouldn't want to hold a large hand of cards. Unfortunately, such disincentives may also end up feeling coercive.
The "three actions per turn" rule may also be considered slightly arbitrary or coercive, but for some reason it doesn't bother me as much as the population-limit rule. It would be nice if there was some way to allow a player to take as many actions as he or she wants per turn, but under the current rule-set, this would allow a player to play a card, move forward, drop backward, draw a card, and play again, indefinitely.
I've never seen anyone accidentally break the rules of Cartagena, or even come close to it. The administrative tasks of playing cards, moving pieces, and drawing cards work very smoothly, and are almost impossible to mess up.
The only potential lack of gracefulness lies in the fact that players have to keep track of how many actions they've used in a turn. This issue isn't serious enough to cause me to lower my rating here.
As always, deliciousness is the most subjective rating. I enjoy very tactical games, and I enjoy the decisions that Cartagena presents. However, most of the people I've played it with have found it vaguely underwhelming. I'm not sure what might be done to spice things up. My intuition is that the problem has something to do with the issues discussed in the variety and emergence sections. It may be that a more emergent "population-limit" rule, allowing players to sometimes draw three or four cards in a single action, would create more tactical variety, more excitement, and more "big moves".
I've always found Cartagena to be a particularly beautiful design. I will never pass up the opportunity to play it, although I do feel that the gameplay lacks a bit in the deliciousness department. I suspect that, somewhere out there in design-space, there's a version of the game that's not only "pleasant" to play, but truly delicious.