(board) Game design

Although I wouldn’t call it the greatest game ever, there is a lot going for Settlers of Catan:

  1. There is something to do even when it isn’t your turn
  2. There are different ways to win (ep with the expansions)
  3. Since the board is random each game is different

If I were to design a game, I’d try to incorporate these 3 elements.
To be sure there are minuses of SoC:

  1. Some people think it is too random
  2. The more you are ahead the faster you get points (i.e. it is hard to catch up if you get behind) (Cites and Knights reduces this with the Wedding and other cards)

What aspects of other games do you like? What elements would you use if you were making your own game? What would you reduce?
Brian

I definitely like games where there is an obvious feedback loop. That is, a game where, if I win, I feel like I can say “a-ha! A key moment in my brilliant victory was back on turn 18 when I clever foobled the smortwangler and purchased the last snozzjuggler left in the bank” as opposed to “well, I won, but I guess I don’t really know why”. Similarly, when I lose I like to feel like I understand why. A big drawback of Puerto Rico is that, even after playing it many times, I often feel like I have no idea why the winner won, or why the losers lost.

Settlers, on the other hand, does well in this category.
I also like a good balance of randomness, which I think settlers falls down on. In particular:
(1) Discarding half your hand when a 7 is rolled is SO swingy. It’s such a crushing blow, particularly when you’ve been building towards something, and it feels so random and arbitrary. Particularly when you’re already losing, and are trying to mount a comeback, you just feel like you’re being punished for no reason at all. It rarely feels like that rule is punishing whoever is already winning. Some other rule could easily discourage card-hoarding without feeling like so much of a kick in the nuts.

(2) The progress cards (both the knight-monopoly-VP-roadbuilding ones from settlers and the more complicated 3-color ones from knights and cities) vary DRASTICALLY in power, with two of them (spy and deserter from K&C) being actively anti-fun. Having things that you’ve worked to get (either knights you built or cards you are waiting to play, and in fact may not have even had a CHANCE to play) being arbitrary taken away and owned by someone else, at the whim of a random draw, really irks me.

(3) And obvious the dice rolling sometimes just completely determines the outcome of the game.
We correct for (3) by playing with a 36-card-deck of random dice outcomes with the proper weighting. We have on occasion corrected for (2) by playing with our own homemade K&C cards.

My favorite games all feel a bit like puzzles to me. You get a certain amount of money, or resources, or information, and the person that manages these factors best does well. This sounds so unexciting, but I love games where efficiency counts.

So games I might try to emulate when designing are The Scepter of Zavandor or Louis XIV . In Scepter, there is some randomness in starting set up, income and the order that cards for auction come out. But what wins the game is how you deal with what you get. Knowing when to really go for an auction, or when it’s okay to sell things to get another something you know you really need is crucial.

In Louis, what is important is adjusting your strategy to the cards you are dealt. You might really really want to win the special number 2 space, but will that tank the rest of the round?

Of games mentioned, I consider my favorite “Settlers” game to be Anno 1503. It’s not an actual Settler’s title, but it is also designed by Klaus Teuber, and shares many mechanisms. It reduces the randomness of the dice rolls by giving each player a game mat with all of the lower level resources assigned to a number 1-6. (you only roll one die for resource production) So you get something every turn, and can easily sell cards you don’t need back to the bank. There is no trading.

I enjoy the Empire Builder games. Every game is random, there is luck involved, but it also takes a lot of skill deciding what runs to make and how and where to build your tracks.

Carcasonne also has a lot of randomness, combined with skill. The Count and The Tower expansions really add a lot of strategic thought to play. (I don’t care for The Count, but I like playing with The Tower.)

What does this mean? The way I’ve parsed it, instead of rolling dice, you draw cards. Cards that have a distribution which is exactly the same as dice. ???

Yes, dice decks have one 2, two 3s and so one to match the different dice probabilities. That way you are guaranteed Gaussian distrubution.
If you remember previous draws you can have a good idea of what is coming up. This can be good or bad depending on your point of view.
There is an official expansion that includes a dice deck but there is a “reshuffle” card that makes it more random. I think it is placed 5th from the bottom in the rules but one could place it up or down depending on if you want the deck more or less random.

Brian

When we’ve played with the ‘dice deck’, we randomly shuffle the reshuffle card in with the deck - so it could be 5 from the bottom, or five from the top, or anywhere. That makes the distribution of numbers pretty random, but not quite as random as actually rolling the dice.

Another approach is to do something like reshuffle on the 5th (second to last) seven. We also lay out the deck face up, sorted, as we go, so that you don’t have to break your brain trying to remember what numbers have and have not been “rolled”.

I see. So you’ve introduced the Gambler’s Fallacy into the game. Interesting…