How to Create Randomized Teams for a Golf Scramble Based on Player Ranking.

Always use "1". That's as close as they can get to one another in rankings.

I am curious as to why "1" will yield the best results. For example, when I run a sample with "1" the best team averages a ranking of 2.4 and the worst team a 3.2. When i change the MaxDifferential to "0.5" the worst team average rating is 3.0 and the best is 2.6. The second method yields teams with closer average ratings and thus is a more equal team. What am I missing?