So are you saying that's why you think their NET is high?
I don't think that's right. From what I can tell the NET uses adjusted efficiency which accounts for SOS and should be similar to kenpom. I think it's the other, not very well defined, component of the NET that focuses on "wins and losses" that is inflating their numbers.
Also, Colgate has so few games that the Kenpom #s probably still include some weight on preseason expectations which helps keep their number so reasonable.
My computer rankings, whatever you may think of them, definitely take into account SOS and mine have Colgate #10 as well. It's just a garbage in, garbage out thing, how are you supposed to make any sense of a team that has played against 3 (!) distinct opponents who have also mostly only played each other; only Army and Navy played any OOC games from that conference this year.
For all my complaints about human rankings and bias, etc, this is where humans > computers. I think the computer output is perfectly fine if you look at it from the standpoint of an alien who comes down and can only see results from this year, doesn't know what a Colgate or a Patriot League is, etc.