According to Kenpom himself, Luck, is the deviation in winning percentage between a teamβs actual record and their expected record using the correlated gaussian method. The luck factor has nothing to do with the rating calculation, but a team that is very lucky (positive numbers) will tend to be rated lower by my system than their record would suggest.
If a team's efficiency suggests they should have won 10 games but they actually won 12 (especially if those extra wins came in close games), they're considered lucky.
If their efficiency suggests 10 wins, but they only won 8 (losing close ones), they're unlucky.