1. A random guess model would do the following. First, it would determine which bucket acquire or not acquire has more customers in it. In this case 338 customers adopted versus 162 who did not. Then it would predict that all customers would adopt and it would be accurate 67.6% of the time (338/500).