Rowridge Vp was Planned by UK Spectrum planners to protect from co-channel Continental transmitters using Hp. They expected new aerial install/replacement in the areas served to use Vp rather than Hp (and eventually not need Hp transmit). Then Arqiva did the COM7/8 thing.
How was the aerial changed to Vp?
A) Taken off the cradle and refitted? so the cradle and pole was in-between the directors?
B) Taken off the pole and the cradle to pole fixing adjusted to Vp, so there's a sideways load on the pole and fixings?
Either way the pole and other metalwork could well have frequency dependent interactions. More so than in Hp orientation.
Non line-of-sight signals can include diffractions and reflections off things and may need the aerial to be moved in a cube of space with roughly 1 metre sides up/down crab left/right and fore/aft to find the sweet spot where all wanted signals are roughly equal. A pro installer uses a spectrum analyser display to aid this process. Even tilting the aerial up to the sky a degree or two can make a difference.
DIY installers have to use sight of receiver signal meters and check all frequencies after every aerial position move.
It's all to do with wavelength, path lengths of direct vs reflected, selective absorption (tree leaf size) etc,. giving cancellation and/or reinforcement.
https://www.aerialsandtv.com/knowledge/aerials/aerial-polar-response-diagrams has some plots of a sample real receiving aerial at ch 21 Vp and Hp. Not very signficant differences to the front. Rear/side lobes however are different.
i