Seeing elevated SWR on my mobile dual-band setup (Nagoya NMO-72 on magnetic mount) compared to the same antenna on a through-hole NMO mount. With magnetic NMO mounts the base of the antenna is above the ground plane surface so the resonant length may be off, and the antenna lacks a DC ground connected to the vehicle body because paint and finishes are in the way. The magnetic mount is grounded through capacitive coupling between the magnet and metal underneath, but this seems less effective than direct metal contact. Should I cut the antenna shorter to compensate for the additional height, and do magnetic mounts really perform better at UHF versus VHF as some sources claim?

Your antenna mount needs to be properly grounded for correct function - without a solid ground, you're bound to get high SWR and extremely poor performance. Clean the roof contact area and check continuity with a multimeter from the coax shield to the vehicle chassis. Sometimes a grounding strap helps.

Yes, mag mounts definitely work better on UHF - the shorter wavelength makes the capacitive coupling more effective. On 2M I always see higher SWR with magnets versus through-hole mounts. A 5/8 wave at 150 MHz needs a 42-inch ground plane diameter, but at 800 MHz only 8 inches is adequate.

I'm dealing with the same issue on my truck. The paint really does interfere with grounding. If you find resistance between the mount and chassis, run a grounding wire to a grounded part of the vehicle and keep it as short as possible. Made a huge difference on my setup.