Friday, June 1, 2018

Design Pattern: Repeat Play Within A Single Game

I know, it has been a while since I last talked about "Games of Strategy" (October to be precise: Why Do Players Do It Wrong?).  But I did knock out chapters 6, 7, and 8 over the winter... and then got busy and stopped blogging.

So lets talk about repeat play.  I once heard Richard Garfield say that poker is close to being the perfect game.  One of the biggest reasons being that each iteration is very short and so the game lends itself very well to lots of repeat play.

I do not remember him going into all of the game theory behind why repeat play is so good (I suspect he knows and was just trying to spare his audience the math).  After reading this part of Games of Strategy I think Mr. Garfield was right to avoid the math.

Just kidding.  A little.  I did skip much of the math...

... But the idea of repeat play was my biggest take away from this part of the book.  So lets get into the weeds on why repeat play is good and then talk about my big idea of how to take advantage of repeat play when your game takes 2 hours instead of 5 minutes.

Mixed Strategies

In chapter 6, the book discussed mixed strategies.  The idea of mixed strategies will be familiar to anyone with a passing knowledge of strategies in sports from football to baseball to the book's favorite example of tennis.  A mixed strategy is when the ideal play is to randomize between a sub-set of choices.

For example, no football team will pass the ball every down -- even though it is almost always more effective.  Instead they run the ball some times... even on 3rd and long.  If teams did not mix their strategies, the defense would be able to predict the offense's strategy and pick the best defense every time.

The math I glossed over can show you the ideal mix of strategies -- maybe 60% pass and 40% run.

Repeat Play

But this all really works best if play is repeated.  If you are only playing one hand of poker, you play to win that 1 hand.  You are much more likely to simply use your best strategy. 

If you are playing an entire evening of poker, you are playing to win the night not each single hand.  Now it makes sense to vary your strategies.  Sticking with only one strategy through-out will just not be the most effective as the night progresses.

Games Within Games

Poker is a great example of repeat play creating a wider strategic space.  More options open up because you play the game over and over and various mixed strategies become the ideal options.

This is much more fun than a single game of poker.  But poker is a 5 minute game.  How do you do that if your game is an hour long?  How likely are people to play that hour long game more than once in a night?  Does repeat play have the same affect if the plays are months apart with different opponents?

What if a single long game can contain smaller mini-games that do repeat?  What does that look like?

The quick answer is that it looks a lot like a football game or a night of poker but I think there is a complication inherent in larger board games that you don't have in sport or poker.

I envision that the strategies I chose for the mini-game would not be constrained by the strategy I have for the whole game or by my current standing in that game.

I think this separation between the larger game and the mini-game is important.  If strategic options in the mini-game are affected by the larger game some of the mini-game's strategies can be made in-effective.  If that happens enough, than the mini-game loses its opportunity to have the 2-3 strategic options needed to allow for mixed strategies.

The Zero Sum Euro?

Euro games are largely non-zero sum games -- what you do hardly affects me at all.  This contributes to the lack of interactivity in most euros. 

It also means that strategic mixes are generally worthless.  The whole point of the strategic mix is to keep the opposition guessing.  But in most euros they are hardly paying any attention to you at all.

What if you put a zero sum mini-game inside a non zero-sum euro?

Now, I care what you are doing -- at least inside that mini-game.


I do not hold out this idea as any sort of ideal of game design.  I enjoy a good 3-4 hour game that I will never repeat. 

But the idea of combining repeat play within a larger context is an idea I would love to explore.


Hat Tip to Isaac Shalev at Kind Fortress who introduced me to the idea of game design patterns.  And to the excellent Lodology podcast where I first heard Mr. Shalev discuss design patterns.