Playtesting Swords & Strongholds
During the development process of Swords & Strongholds, David and I reached out to our respective communities to ask them to playtest the game. We did five rounds (not including our own in-house playtests). For each round, we'd send the current print and play draft and asked questions like these:
- How many games did you play?
- How long did each game take?
- Who won each game?
- Was there strategy involved?
- Were the rules clear? If not, what was unclear?
Answers to these questions helped paint a picture for what was going on in the games. If we had a win/loss ratio of 9 to 1, we knew there could be something wrong. Or if players reported that the Diplomacy card was the key to every strategy, the game could be broken.
Early in the play test process, I submitted the game to my friend Wilhelm in Seattle. He grabbed his friend Dylan and they set about trying to break the game. After a week or so, we got a detailed report back that ended with these sweet words: "…no degenerate strategies immediately lept to the fore for either of us." Those are the most beautiful words a designer can hear, I think. A "degenerate strategy" is a term of art in game design. It's a play that will either break the game and prevent either player from winning or one that will guarantee a player win if undertaken. "No degenerate strategies" meant that our game had a solid foundation.
Wilhelm's assessment made us very confident as we proceeded into further playtests, but I knew it wasn't enough. We needed to have Wilhelm's analysis confirmed. So I submitted the game, on trembling knee, to my friends Rym and Scott of Geeknights. I've seen their talks, I knew that if there was a problem with the game, they'd find it and point it out to me (or possibly in public, in front of 400 PAX attendees).
It took them about a month to make time in their busy schedules to dig into the game. I was on tenterhooks! But eventually, Rym sent me a very detailed play report and analysis. I've included the bulk of it below. I'll let Rym speak for himself (in italics).
There are no obvious degenerate strategies. There was also no consistent way to indefinitely prolong the game (a problem with the superficially similar Oshii). The latter was a worry just from a read of the rules (since most games avoid this by having a fiddly bit about not moving in the same pattern more than X turns in a row).
Our win ratio was very close to 50/50. We were unable to discern any material disadvantage to going first, but it would take a lot more playing to suss that out. It's definitely minor if it's even present.
Attempts to rush a stronghold victory with a single mouse are extremely unlikely to succeed. The power of the Diplomacy swap action is heavily mitigated by the fact that the sword can easily kill anything close to such a victory.
The board has some subtle tricky spots that present a "positional heuristic" problem similar to Othello, in that they're much weaker than they appear. Take for example this configuration:
The white mouse in the lower left cannot stronghold to win in one turn. He must move first in order to have a card played on him, and can not move in such a way that would allow him a stronghold victory. Interestingly, this position is usually obtained by killing a defender with a sword, providing that player with a chance to recover. Similarly, the black mouse strongholding in the upper right cannot achieve victory in one turn either, due to the inability to play a card on it after exiting a stronghold. Furthermore, exiting the stronghold in this manner would prove lethal if the white player had a sword.
Consider this layout:
The mouse in black at which the card edge is pointed, despite having a sword to expedite movement, cannot move into a position on the upper right where he could then achieve a stronghold victory the following turn. It would require two turns of movement, followed by a third with stronghold played, to achieve it.
These aren't problems, but they show that the board layout is, due to the nature of the movement available, not as straightforward as it appears. It could likely be graphed in terms of stability/momentum in the same way that an Othello board could.
The best strategy was found was to be very aggressive and center-focused, attempting to slowly capture the other player's mice. Stronghold victory is usually an afterthought. The best counter to this is to be compactly defensive. The former requires the luck of drawing sufficient sword cards. The latter is fairly universally applied despite cards.
Older players will likely find that card-counting is moderately rewarding. I only attempted to keep a soft count, and it gave a good sense of when to expose my stronghold-moving mouse due to the unliklihood of an opposing sword.
I was blown away: They confirmed that there were no obvious degenerate strategies and went a step further and dug into the game itself and found things that even we hadn't seen or intended! Their comment about Oshii confirmed what I suspected, but the positional strength of certain squares based on the cards in your hand was not something that had occurred to me.
So it was very good news indeed. Once we heard this, we did another round of playtesting with our fans to make sure the rules were clear and prepared our Kickstarter. If Rym and Scott had found problems with the game, we wouldn't be where we are today. We would have gone back to the drawing board and pushed off the launch of the game indefinitely.
So three cheers for playtesters!
Thanks for all your support. We'll have more news soon!
-Luke & David