Reposted from blog: http://nickbentley.posterous.com/new-game-byg

I had a Eureka moment today, courtesy Christian Freeling. Christian has a new game called Symple, the core of which is an inventive rule specifying how a player may add stones to the board. The rule is: you may either place one stone, not adjacent to any other like-colored stones, or grow any/all of your preexisting groups by one stone each. A conundrum results: do you add one measly stone so that you have more groups to grow later, or grow your groups now without increasing the number of groups?

Meanwhile, I’ve been trying to invent games where each player’s goal is to have the largest group, but where draws are impossible. I’m attracted to this biggest-group goal because it’s profoundly intuitive; humans seem to have an inborn proclivity to try to make things bigger. It feels natural to us.

However, it’s difficult to design such games because it’s hard to prevent players from clumping their stones together, which tempts a designer to gouge his eyes out and eat them.

I’m proud of the one success I’ve had so far, a game called Ketchup, but Ketchup isn’t perfect: there’s a restriction that you can’t place your stones such that your largest group is the same size as your opponent’s largest group at the end of your turn. It’s clunky and requires that you keep track of the stone counts as you go along. It’s no problem when playing online where the computer does the counting for you, but I want a game that plays just as well in meatspace as in cyberspace.

When I read about Christian’s drop mechanism in Symple, I saw that it was ideal for my purpose (Although I did simplify it slightly for my game). If played on a board with an odd number of cells, it yields a drawless game without clunky restrictions. It feels beautiful to me, perhaps because I’ve been pondering the problem for so long.

I haven’t decided if it’s better to include a 12* drop rule, as in Connect 6, or a one-stone-per-turn-with-pie rule, so I’m including both versions here. Let me know which you prefer and why. First I’ll cover the rules, and then I’ll explain why it’s drawless and why Christian’s rule is critical for making it drawless in a fair way.

Rules:

Byg is a game for two players, played with black and white stones on a hexhex board.

Version #1

1. The board begins empty.
2. To begin, White puts a single stone on the board.
3. From then on, starting with Black, the players take turns. On your turn you may either:

* Place one or two stones on vacant cells, not adjacent to like-colored groups or to each other; or
* Grow any number of your existing groups by one stone each (you must grow at least one group). A stone connecting two or three different groups together is considered to have grown all of them (at the end of your turn, no group may have more than one new stone in it).

4. The game ends when the board is full.
5. The player with the largest group wins. If the players' largest groups are the same size, compare their second-largest groups, and so on, until you come to a pair of groups that aren’t the same size. Whoever owns the largest group wins.


Version #2

1. The board begins empty.
2. To begin, one player puts a single white stone on the board, and the other player decides whether to take ownership of the white or black stones.
3. From then on, starting with Black, the players take turns. On your turn you may either:

* Place one stone on a vacant cell, not adjacent to like-colored groups; or
* Grow any number of your existing groups by one stone each (you must grow at least one group). A stone connecting two or three different groups together is considered to have grown all of them (at the end of your turn, no group may have more than one new stone in it).

4. The game ends when the board is full.
5. The player with the largest group wins. If the players' largest groups are the same size, compare their second-largest groups, and so on, until you come to a pair of groups that aren’t the same size. Whoever owns the largest group wins.


Drawlessness and the Freeling Rule
Because Hexhex boards have an odd number of cells, if you go down the line comparing the players' largest groups, second-largest groups, and so on, there must be at least one pair of groups where one group is larger than the other. This is true whether or not we use Christian’s rule. However, if the number of stones that each player places onto the board does not vary as a function of the players' own choices, one player will be ensured what can be thought of as a tie-break, merely in virtue of turn order, which is unfair. Christian’s rule allows the players to choose how many stones to put on the board and when, and as result, getting the tie-break becomes a matter of skill, not of turn order. Beautiful. Thank you Mr. Freeling.

I designed Ketchup some months back, and it’s been tweaked a few times. I just tweaked it again, to fix a first mover advantage that had become apparent, and now I’m confident that the form is final. I hadn’t mentioned the game here, so I thought I would now that I’m more confident about it.

Printable rules and PnP boards here:
http://nickbentley.posterous.com/ketchup-revised

It comes from a discussion I had with Bill Taylor about the possibility of incorporating negative feedback into games. By negative feedback, I mean: the closer you get to winning, the weaker you get. Yinsh is an example: the closer you get to winning, the fewer rings you have to play with. Beyond Yinsh, there are few negative feedback games. I set out to invent one, and Ketchup is the result.

The goal of the game is to end the game with the largest group on the board, but the player with the largest group at any given time can’t place as many stones on his turn as his opponent can. But you can’t just dick around make small groups until the endgame either – if your opponent gets too connected, extra moves won’t help you much. There’s the rub.

It’s simple and quick and tension-heavy. It seems to play best on small boards (start with hexhex5).

Reposted from blog: http://nickbentley.posterous.com/byg-analysis-1

Abstract game designers publish many designs but not much analysis.

This is odd, if not sinful. Isn’t our goal to design games harboring catacombs of secrets, waiting to be discovered? Unless we enter the catacombs and report our findings, we neglect to partake of the very pleasures we set out to create. It seems a waste.

And so I’ve resolved to write about my effort to understand Byg, my new game:

http://nickbentley.posterous.com/new-game-byg

(Is this kind of post appropriate here? If not, this will be the last of this kind I post. Let me know)

I’ve been evaluating the two versions, and today my friend Scadeau and I played some test games. Insights were had.

First, we played on a Hexhex5 board, and it wasn’t too small. Good: small-board play is less likely to intimidate new players.

Second, Scadeau, a (WBC champion) Eurogamer who’s played many of my games, declared that Byg is my best. Since I dream of designing abstracts with appeal beyond our weensy group of abstract players, his sentiment is encouraging. He liked it enough that he’s going to design a Eurogame around it (now that I’ve said it in public, you’re locked-in my friend! How do you like that gambit?)

Our most important insight: we prefer the version where a player can create only one new group on his turn, rather than 2. Here’s why:

A key moment in each contest is the turn when one player switches from group-creation to group-growth for the first time. The fun wratchets up at this point, so it should come early. The less powerful group-creation is relative to group-growth, the sooner it will come. Group-creation is weaker when you can only create one group rather than two, so the design choice is clear. One caveat: the key moment shouldn’t come so early as to create an advantage to one player based on turn order. We have yet to know how early that is. We may have to revise our conclusions later to adjust for this.

It occurs to me that there is another group-creation rule intermediate between the one- and two-group-creation rule: you can create one 2-stone group. Since I don’t yet know where the sweet spot lies, it’ll be important to test this version.

Another way to make group-growth happen earlier is to make it more powerful. One way to do this might be to remove the rule which says that a stone connecting 2 or three groups is considered to have grown them all. But that might be too powerful. I’ll test it and report later.

Strategy note: the center is often a poor place to have a group, at least on the hexhex5 board. The reason:

If your groups are close to each other too early, you may be forced either to a) connect too soon; or b) not grow one or more of your groups when you choose group-growth, to avoid connection. The problem with having a central group is that it can make all your groups too connected, too soon.

This last observation points to a general truth: Byg is not completely hot. Rather it’s both hot and cold, depending on where in the game’s arc you are and on the the geometry of your groups. The mixture lends variety and subtlety, another reason to rejoice. The presence of some cold moments also makes first-mover advantage less likely to be a problem.

So far so good. Stay thirsty, my friends.

Reposted from blog: http://nickbentley.posterous.com/byg-analysis-3-in-which-the-rules-are-utterly

My attempt to design Byg has clarified my thinking about the design challenge inherent in Byg’s win condition, which is:

The game ends when the board is full and the player with the largest group wins. If the players' largest groups are the same size, compare their second-largest groups, and so on, until you come to a pair which aren’t the same size. Whoever owns the larger, wins.

I like this goal because it’s simple, drawless, intuitive, easy to visualize in-game and easy to know who won at game-end (it can often be seen at a glance).

The difficulty of turning it into a game can be seen if we consider what happens in the most basic game we can design: the players take turns dropping stones without restriction until the board is full.

What’s the problem? there are two fundamental actions in this game: create a new group (by placing a stone non-adjacent to any like-colored group), or grow a group (by placing a stone adjacent to a like-colored group). The problem with this simple game is that it’s always better to grow a group than to create one, and this leads to a trivial game with obvious choices.

Ideally, I wish for a game in which group-growth and group-creation are both important throughout the game, so that it’s hard to decide between them.

How to do it?

The present rules of Byg are one solution. The Byg turn rules make group-creation valuable, at least during the opening, by making it a prerequisite for maximum group-growth at the end. I’ve grown disenchanted though. The game has two clear phases, where group-creation is the obvious choice in the first phase and group-growth the is the obvious choice in the second. There are delicious decisions in the transition between phases, when it’s not clear whether to grow or create groups, but the phases themselves leave something to be desired.

The first solution I considered was a hard limit on the number of groups you can grow.

That lead me to the following rules for Byg:

1. The board begins empty.
2. To begin, White puts a single stone on the board.
3. From then on, starting with Black, the players take turns. On your turn you may either:
* Place one or two stones on vacant cells, not adjacent to like-colored groups or to each other; or
* Grow up to 3 of your existing groups by one stone each (you must grow at least one group). A stone connecting two or three different groups together is considered to have grown all of them (at the end of your turn, no group may have more than one new stone in it).
4. The game ends when the board is full.
5. The player with the largest group wins. If the players? largest groups are the same size, compare their second-largest groups, and so on, until you come to a pair of groups that aren?t the same size. Whoever owns the larger group wins.

This works somewhat – it makes group creation less important in the early phase, but the late phase remains group-growth only. I decided to keep searching.

Guiding my thinking was this key observation: group-growth becomes increasingly important as the game goes on as long as the win condition is to have the largest group at the end. As a result, it’s important to give strong incentives to create groups late in the game, but without making them the only viable choice early . Tricky.

Grappling with this problem, I came up with the following game:

1. The board begins empty.
2. To begin, White puts a single stone on any vacant space, and then Black does the same.
3. From then on, starting with Black, the players take turns. On your turn you may place as many stones as you have groups on the board, and you may place them onto any vacant spaces. Except:

* You must place at least one stone.
* If you have more than 3 groups on the board, the maximum number of stones you may place is 3.
* You may not grow any group by more than one stone.
* A stone connecting two or three different groups together is considered to have grown them all (at the end of your turn, no group may have more than one new stone in it).

4. The game ends when the board is full and the player with the largest group wins. If the players’ largest groups are the same size, compare their second-largest groups, and so on, until you come to a pair which aren’t the same size. Whoever owns the larger group wins.


Note that with these rules, you can both grow your groups and create new groups on the same turn if you wish.

Now there’s no cost to creating new groups late in the game (since there’s no stone-penalty for doing so). This allows a natural group-creation incentive to operate: groups will merge late in the game and you’ll have to create new ones to keep your group-count up.

In addition, there’s no great cost to merge groups early in the game,since you can create new ones on the same turn to keep your group-number up to the maximum. This allows the natural incentive to merge groups shine through (up to a point) in the early game.

The dynamic which prevails is clear: each player’s goal is to create at least 3 groups and stay at or above that number either by growing them without connecting them, or connecting them and creating new groups) until the last possible minute. If 3 groups seems too low, you can change it by changing the limit on the number of stones a player may add. Testing will be needed to determine the optimal number. I suspect somewhere between 3-5.

I also don’t yet know if there are balance issues. I’ll report more enlightenment later.

Reposted from blog: http://nickbentley.posterous.com/new-game-population-pressure

Two tribes are competing for territory by breeding. Couples make babies. My most thematic game :)

Population Pressure is a game for 2 players, played on a hexhex board. I recommend Hexhex5 to start.

Rules:

1. The board begins empty.

2. To begin, one player places one white and one black stone on any vacant spaces on the board, and the other player decides whether to play as White or Black.

3. Starting with Black, the players take turns. You may place one stone on any vacant space on each turn, until you have a group on the board of size 2 or larger. After that, on your turn you may place as many stones on vacant spaces as you have groups of size 2 or larger on the board.

4. The game ends when the board is full, and the player with the largest group wins. If the players’ largest groups are the same size, the player with the fewest stones on the board wins.

On larger boards, you might experiment with higher minimum group sizes (Polyamory?)