At Agile 2008, Don McGreal and Michael McCullough ran a session that showed how to use games and exercises to help improve our understanding of Agile principles and practices. After the conference they created the Tasty Cupcakes as a repository for all Agile games.
Among the games that they document are:
- We're Having a Party! : Demonstrate the benefits in Single Piece Flow over Batch & Queue
- Football SCRUM : Enforce the rules of the daily stand-up meeting in a fun way
- You are in Control (Not) : Build paper planes using only specialists
- The Story of our Sprints : A interesting way to run your retrospectives
Since then a number of Scrum Masters and Coaches have been discussing other useful games and variations. One participant suggested Boris Gloger's Ball Point game (documented by Kane Mar).
The objective of the Ball Point game is to get as many balls through the team as possible within two minutes.
Rules:
- Provide an overview of the game and the rules.
- Everyone is part of one big team.
- Each ball must have air-time.
- Each ball must be touched at least once by every team member.
- Balls cannot be passed to your direct neighbour to your immediate left or right.
- Each ball must return to the same person who introduced it into the system.
- There are a total of five iterations.
- Allow the team two minutes of preparation time to determine how they will organize themselves.
- Get an estimate from the team of how many balls they can pass through the system.
- Run a two-minute iteration.
- Allow the team one minute to discuss how to improve the process.
- Repeat for five iterations. Make the fifth iteration a challenge. If you need to, make up some ridiculous statistic such as “The world record is 150 points. Can you beat that?”
Kane has a small variation. After four iterations teams will have reached its maximum velocity, at this stage the only way to improve is to change the system. Kane uses the fifth (bonus) round as a way of forcing the team to change the system - perhaps changing one of the rules. After the round is played they examine whether this change had a positive or negative effect.
Finally Angela Druckman offers a game she uses when teams are uncomfortable estimating in points (usually because they say hours are more reliable):
I display a list of 10 things I ask people to estimate and indicate their +/- range. These things are basically obscure trivia obtained via the Internet, such as the average number of worldwide deaths from snakebite per year or the number of known species of shark. I tell them the object is to get as many of the estimates correct as possible. They estimate and we look at their estimates and the correct answers together.
Invariably, they are way off on a few (often most) and their +/- ranges are way too small.
So my question for them is why did they not make their variance ranges bigger? And we talk about the "implied precision" that an estimate in hours gives and how, even without being told to do so, they felt pressure to choose small variance ranges.
Then we estimate another group of items but this time we use t-shirt sizes (xs, s, m, l, xl). It might be country populations, prices of luxury items, etc. Invariablly, the group interacts more when estimating in this fashion and they estimate quicker. We talk about the benefits of this kind of estimation.
Mike Sutton, Agile Coach with Wizewerx Consulting, just ran a workshop that featured games at its core: "There is nothing as effective to accelerate learning as a physical immersive game. The simpler the better, better still with near to no props. As low tech as possible. You get to see the penny actually drop with some folks too - and that is a great moment"
Don and Michael promise to update Tasty Cupcakes with more games as they have time.