MtG head designer, Mark Rosewater recently offered the following definition of a game:
A game is a thing with a goal (or goals), restrictions, agency, and a lack of real-world relevance.
I think this definition is insightful and captures all the essential features of a game. The part of the definition that I would like to talk about today is "agency," which is the idea that a true game must have meaningful decisions and choices. If it doesn't involve decisions, it may look like a game, but it is really just an activity or event. Rosewater offers movies as an example, as well as the nominal game of Candy Land, in which players move their pieces by drawing cards, but never get to make any real decisions.
Decisions make games interesting and also provide opportunities for higher level thinking, such as planning, strategy, and tactics. But here is a question for you. Is it possible to specify a certain number of choices or options that would be good to have while taking a turn in a game? It's got to be somewhere between 2 and 99, but can we narrow it down? This is not just a question about the psychology of decision making, it is a practical issue for game designers to consider as well.
Choose a card, any card... but choose wisely!
The critical empirical research here can be found in Sheena Iyengar and Mark Lepper's influential 2000 paper on choice overload. In the most well known version of the experiment, shoppers were given the opportunity to taste and also purchase different varieties of jam. At random, they were offered either 6 or 24 flavors to choose from. Contrary to the common sense notion that more choice is better, participants appeared to be overwhelmed by too many choices. They actually bought more jam when they only had six options. Does this research have any relevance for board games? You need to have choices, but too many choices could adversely affect people's ability to learn and play the game.
It's interesting that many card games have a maximum hand size close to Iyengar's optimal number of jams. Magic: The Gathering, for example, has a hand limit of seven cards. This is also a typical starting number of cards in drafting games like 7 Wonders. It would be possible to play with more cards, but it would slow the game down considerably. In the supermarket experiment, people walk away without buying anything. In a board game situation, people fall into analysis paralysis. There are other examples. Puerto Rico, the number one game on the internet for many years, has seven roles to choose from. Is seven the magic number?
Maybe not. Quite a few games have fewer choices. Rondel games like Antike have only three practical choices, unless you are willing to pay the penalty for moving your piece further along the wheel. This keeps the game moving and minimizes downtime for the rest of the players. On the other end of the spectrum, worker placement games like Caylus and Agricola have many more choices, at least by the end of the game when all the options have been revealed. This actually fits with some of Iyengar's later research. It turns out that we can handle choices better if we work up to them slowly. A great many options at the very beginning of the game is probably not the way to go.
This takes us to my old favorite, Chess. There are twenty possible choices for White's first move, 16 pawn moves and 4 Knight moves. The number of options can increase dramatically from there. Well, chess does often lead to analysis paralysis. There was a reason for the invention of the chess clock after all! This observation could lead to a common criticism of chess, that it is not a good game, or a well designed game. Hey, it doesn't even crack the top 400 here at BGG. But as in all games, experience clarifies and reduces the glut of information. Serious chess players would not even consider most of those twenty first move options. Later in the game, as GM Alexander Kotov sagely recommended, the best way to decide on a move is to first narrow the field down to a few, maybe three "candidate moves" for further consideration. Only at this point does chess begin to avoid the dilemma of choice.
Let's hazard a conclusion. Games clearly require choices. A small number of choices will probably lead to a game that is light, fast playing, and easy to learn, whereas a large number of choices will likely lead to a game that is complex, heavy, and difficult to learn and play. I'm going to go out on a limb and speculate that a moderate number of choices, somewhere between 3 and 9, might be the sweet spot for many game designs.
Iyengar, S. S. & Lepper, M. R. (2000). When choice is demotivating: Can one desire too much of a good thing? Journal of Personality and Social Psychology, 79, 995-1006.
Statistics and Speculations on the Behavioral Science of Board Gaming
12 Dec 2018
- [+] Dice rolls