So, I recently got the urge to do something, and picked up a discarded idea from years ago. I discarded the idea because the rules didn't work right, so I poked at the rules and came up with something I thought might work.
How to test it, though? The rules are simple, but the strategies that should evolve from them are harder to predict. So I created the board game as a computer game, and played it repeatedly against myself. At this point, it seems to work, but playing it against myself, I find I've fallen into a pattern, and I wonder if I've found the "optimum" strategy. A game that HAS an optimum strategy is a game that's broken, IMO.
Take turns placing pieces on the board. You may place them on any empty intersection. The object is to either wipe out your opponent or (more likely) claim a larger segment of the board.
The interaction of pieces is based on a comparison of threat and strength. Strength is based on how many friendly pieces a given piece is touching, including itself. Threat is based on how many opposing pieces are touching it. If threat is greater than strength, the piece will change color to the opposing side.
Example:
x x x o o o
The center x has a strength of 3, because it is touching two other pieces. The flanking x's have a strength of two. All three o's have a strength of 3. The upper two o's have a threat of 2, as each of them are touching two x's. Each outside x has a threat of 1, and the center x has a threat of 2. None will flip. If an x is place to either side of the upper o's, threat will increase to 3, which will still not be sufficient to flip them.
Strategies: Build in clumps to keep strength up. Never approach the opposing color without strength. Attempt to envelop and trap your opponent's pieces.
Program functions:
Arrow keys or the mouse will move the cursor (the flashing colored dot) around the board.
Enter or left-click will place a piece according to the color of the cursor. Pieces will automatically flip when appropriate.
Spacebar will highlight all empty spaces.
Esc will close the program.
The game will end if one color is wiped out, or when the board is full. (126 turns)
I downloaded and played a couple of times, but I don't know if playing against myself is proving particularly helpful! I am not much of an abstract gamer, though. Can you detail the "broken" strategy you came up with?
Here's what seems to work best. Once you meet your opponent, one or the other will make a move to envelop the other. The other is basically forced to counter-move, which ends with them pushing toward one of the walls. When you have reached the wall, the one to secure themselves first goes to find an open space, often behind their opponent, that's large enough to try to build another strong area. Leave open spaces that you've already dominated, because it frees you up to move elsewhere, and it may eventually force them to move there, which will give you a piece for free.
With this strategy, Misplaced stones or Taking risks is the only thing that will affect who wins.
Now, this is playing against myself. I don't have much call to ask someone to sit down at a single computer with me to try it out against someone else.
I'm convinced Trian needs three players (and, ideally, a triangular board). I really like the three-theme. The strategies are going to be incredibly different.
In the original game, we know that this:
Q Q
X
will turn the X into an Q, clear as day. And we can only put down Qs and Xs.
BUT in a three way game, this:
@ Q
X
won't affect any of our playing pieces.
---
In the original game, this pattern is obvious (and is the quickest win):
[Q]
[X ]Q
X Q
[Q]
Q Q
Q
Assume the brackets are "placing" phases. We get three Qs. Q wins!
----
Now for three players:
[X]
X [Q]
X Q
[@]
X Q
@
and now I fail at my imagination at fitting triangles into these grids. so I suck at figuring out additional moves that might happen after this point. BUT:
I think there should be some method for the X and Q players to create "@" on their own, without the influence of any @ on the board. I'm not sure how this should happen, but I honestly and truly feel it in my gut that this should happen. I just need to play with a 3-player board to figure it out.
Hmm, maybe an AI player that detects when an X Q appears and automatically places an @ to complete the triangle? It wouldn't have to trigger on every single adjacent X and Q, just when they were adjacent and had no other neighbors.