OK, for the sake of transparency, this is not a work issue... I am a clinician-scientist and just cannot find the solution for this anywhere:

We have twelve continuous variables (HCI) - These are twelve golfers handicap indices
We now want to create two groups (ie. two teams) with the small difference in (HCI) between the two groups - therefore the groups should have the smallest difference possible in terms of the mean HCI per group.

Any help will be appreciated!!