Not sure this will help you as I'm quit new to astronomy and just getting my feet wet with Skyfield. My approach for testing for specific satellites that could pass in my sky were as follows. I use geocentric = iss.at(times), subpoints = geocentric.subpoint() to determine the geographic position elements (latitude, longitude, elevation) assuming these are ground track figures. I then test if the figures fall within the bounding box I defined loosely based upon the horizon distance formula. My bounding box is about half of what the distance formula suggests thus I'm assuming the great circle distance will always be less than the horizontal distance and within my bounding box. I retain only those geographic positions that meet the above criteria with also the addition of being within evening hours. I have yet to validate my findings as they are still off from info I'm finding on other sites for satellites of interest to me. Haven't figure out yet the discrepancies but I believe the approach is sound in narrowing down the potential set of satellites over my area.