When you have done the second step, you should have a new figure for capacity and time to market. The following diagram shows the result plotted in a graph:

The game's result put in a graph. As you can see, the lead time shrinks more than the throughput when introducing limits.

Theoretically, there should be a big difference in time to market (ΔL) and a small difference in throughput (ΔT). Practically, this is not always the result due to the big variation in personal capacity, 1-6. If you have four teams or more doing this exercise, your average result will most likely get close to the theoretical outcome.

The time to run this exercise is about 30 minutes.

You can make a tweak to the exercise to get a more obvious result. What you can do is generate a bottleneck by reducing the capacity for the second user. Divide the second user's dice result by 2, rounded up, so if the dice shows 4, he or she moves only 2 sticks, and if the dice shows 5, then he or she moves...