Chapter 5: Cures for Pandemics and Alpha Players

[Kris comment: This chapter on cooperative games and the idea of a party contract and the role of “alpha players” echoes the trade-offs and design of the cooperative game of human governance. It recalls the timeless tension of equality/social cohesion with efficiency. It reminds us that incentives and design are inseparable. Balancing our instincts with our higher-order intellectual abstractions such as justice, freedom and safety is a delicate problem.
Education is also a metaphor here. The party contract of how much to pull up the bottom vs accelerate the top. Are laws and tax codes the practical expressions of platonic ideals of what the party contracts should be or are they just path-dependent layers of political wins and losses reflecting the desires (or illusions) prevailing at their inception?
This is just the beginning of the questions.
Where do party contracts come from? Philosophy? Rules about family leave also come to mind. What kind of society do we want? How do family leave laws affect women, half our population, and our conceptions of freedom?
What does meritocracy mean in theory? In practice? Many people smart in autistic ways have legible kinds of intelligence that society finds easy to reward today because the input/output function to efficiency and therefore value creation is easy to see. But is this because of our short-term accounting? A preference for [fake] meritocracy is usually self-serving because they are the best at finding the optimum if you give them the rules. But the design of the rules requires bringing in values that were never as legible as the autist likes to think. A real meritocracy would likely be far more progressive if it took the long (in time) and wide (in geography) view. Is the value of such a galactic view so widely distributed that it could never gather a critical mass of advocates?
In a similar vein, we wonder, to what extent are local optimums even avoidable? When does a local optimum become species-level quicksand? Is centralized top-down authority the only coordination system capable of avoiding paper-clip maximization? Are we forced to choose between the tyranny-violence cycle or the trap of local optimums?]

A common problem for organizations is incompatible goals.

For example:
A team of lawyers has been tasked with a complex legal problem. The team is composed of several senior partners and a few juniors. At first, everyone is happy and united. As time passes, pressure builds, and the file becomes more complicated, and the seniors get frustrated by the slow pace and high error rate of the juniors. Eventually, they start doing all the work themselves, relegating the juniors to mundane clerical roles. The job gets done. No one yells at anyone, or even says a harsh word. It does not seem like there is any real conflict here. Yet everybody ends up frustrated, especially the juniors, who are left wondering why they were brought on to the project in the first place.
Scenarios like this, which probably arise more often in most companies than episodes of outright hostility, are hard to manage because they are not covered by standard mechanisms for resolving disputes, like rules of workplace conduct. The only way to address them is through some kind of unspoken company-specific meta-rules about the trade-offs. [A familiar trade-off is] short-term goals such as task efficiency and long-term goals such as mentorship, morale, and staff retention.
Believe it or not, this is a common problem faced by board game designers, and some have come up with novel solutions for addressing it…Certain kinds of board games can help your organization solve or at least identify conflicts that arise when different people within the organization have different assumptions about the best way to balance short-term task execution with long-term capacity development.

The “play contract”

Whenever you sit down to play a game, whether realize it you or not, you are entering into an unspoken agreement with your fel- low players. There is no universally agreed-upon text for the play contract. But if there were, it might include these basic precepts, which flow from our discussion of the magic circle in the book's first chapter:
  1. I agree to abide by the rules of the game as I understand them; no cheating.
  1. I agree to take the game seriously enough to make a sincere effort to win; no throwing the game.
  1. I agree to not take the game so seriously that it will affect my real-life relationships with my fellow players; no behaving like a jackass.
[Kris: reminds me of the spirit behind the social club I helped start]
The problem is that the second and third points sometimes come into conflict. How hard do you have to play to satisfy the need for a sincere effort? How easy do you have to take it on your fellow players to keep things from getting too heavy? Those stories you have heard about wounded feelings and damaged friendships: in many cases, they are caused by players coming to the tabletop with conflicting answers to these questions.

The Alpha-Player Problem is one way cooperative games can mirror the conflicts in real-life cooperative groups

For a lot of people, the prospect of playing a game with someone who is much better than them is attractive. But in Pandemic, the other players are not your opponents, they are your teammates. In theory, playing with someone who is much better at it than you should be a good thing because she is on your side. If
But is it always better? Suppose one player can see more clearly than the others which moves are more likely to lead to victory, and which ones are more likely to cause problems. If she is following the play contract, trying to play as well as she can, she should be doing her best to help the group win by telling her teammates what she sees. If you are about to make a sub-optimal move, she should point out why it is a problem and perhaps suggest a different move you could make, one which would achieve a better result. No harm, right?
Now suppose our expert player does this for all of her teammates. Suppose she does it every single turn. The group may have a better chance of winning but now only one of them is actually playing. The others are reduced from teammates to spectators. They might as well not even be there. Perhaps this reminds you of the junior and senior partners at the imaginary law firm I described a few pages back.
This happens fairly often in cooperative games-often enough to need a name. And so players and designers have dubbed it the alpha-player problem. You might also hear it referred to as quarterbacking. In a traditional competitive-style game, there is no alpha-player problem because the play contract ensures a natural kind of sustaining tension-everyone is out for his own gain. But in a co-op game, the play contract doesn't always work.
Solutions to the alpha-player problem:
  1. Follow the third precept of the play contract at the expense of the second. That means the alpha just accepts that she is not going to play her absolute best. She allows the others to play the game at their own level, and to win or lose on their own terms. A less paternalistic way of going about this would be for the non-alphas to impose this solution by shushing or otherwise declining advice from the alpha.
  1. Reframing Redefine "optimal" is according to a long-term framework that takes larger goals into account such as a firm's need to develop future talent and leadership. In the same way, an alpha player might see it as "optimal" to allow newer players to make mistakes because it helps them get better through trial and error, thereby helping to keep the gaming group together and lay the groundwork for more fun in future games. (A gaming purist might say that this step serves to break the magic circle because the alpha player must compromise in-game objectives to satisfy the real-world social and psychological needs of other gamers. But as with all forms of doctrinal absolutism, this sort of fundamentalist attitude to gaming serves to make the enemy of the good.)
  1. Design nudges Good game design can get us around this problem….In Antoine Bauza's award-winning Hanabi (2010), for instance, players work together to accomplish a common goal but each has access to information teammates cannot see, and all are severely restricted in their ability to communicate. Without access to the entire picture, no alpha player can know for certain which moves will or will not be beneficial to the team. So everyone has to try his or her best with the limited information they possess. Another way of dealing with the alpha-player problem in cooperative games is to put players in a time crunch. In Eric M. Lang's alien-invasion resistance epic XCOM: The Board Game (2015), Vlaada Chvatil's space-borne comedy of errors Space Alert (2008), and Kane Klenko's nail-biter of a bomb disposal game Fuse (2015), all information can be shared among all players, in theory. In practice, it is impossible to share more than a few snippets of it because players must make choices within a severely limited amount of time. Potential alphas do not have time to micromanage everyone else. They and all the rest must decide when a piece of information is important enough to be worth expending the group's precious time and attention to share it. [Kris: Both are interesting points in broader real-life contexts of how constraints can actually unlock ingenuity] Can this trick of tackling the alpha-player problem by providing players with limited information or limited time to act on that information be extended to real-world organizational behavior patterns? Definitely. One of the main reasons some bosses micromanage is that they do not have a lot of work on their own desks. Give a boss something to do, and she will tend to give more autonomy to her minions. Likewise, minions who are tired of being told how to tie their shoelaces may rebel against corporate higher-ups by hoarding data within their fiefdoms and throttling the flow of information. The boss cannot micromanage a department she cannot fully survey or understand.

The right teammates

Ultimately, the best solution to the alpha-player problem isn't based on picking the right rules. It is based on picking the right teammates. In a game of alpha players, nobody is an alpha player. Similarly, if everyone is new to a game, players can stumble happily along as a team, making mistakes together and learning lessons about how to win on the next attempt.

Final word:

This is not an airport-rack business book to show you how to get the most out of your employees, improve staff morale, or reduce friction within your organization. But playing games really can provide important lessons for people running companies. In particular, cooperative games such as Pandemic teach us that group dynamics can get more complicated, not less when people are trying to cooperate rather than compete. This is important because most businesses, NGOs, government agencies, social clubs and even families can be thought of in some way as cooperative projects, even if real life tends to lack the well-defined rules and victory conditions you would find in a cooperative board game.
Like players in a cooperative game, close colleagues may end up struggling with the balance between the ruthless pursuit of short-term performance and the long-term effort to enhance their colleagues' skills and sense of engagement. Too much of the former, and the company becomes a haven for alphas but alienates everybody else. Too much of the latter, and the company might have trouble satisfying its clients even as it develops a great reputation for in-house vocational mentorship.
In most organizations, there are unwritten rules that govern the desired mix, much like the play contract at the tabletop. These rules go a long way toward defining an organization's culture. And if you are going to join up, you will probably want to make sure that your own values are congruent-whether it is a multinational corporation making widgets or just a few friends gathered around a game board, trying to save the human race from extinction.