If you think about local field strength and how an electromagnetic wave interacts with a dipole, it doesn’t matter if the dipole is a flat plate and which angle that plate is at.
Well … within reason.
If you’d get to more than quarter wavelength you would probably start getting issues as when a zero transition arrives you get counteracting induced currents.
But this width is chose for how the wave resonates once coupled into the antenna.
As it’s signifantly shorter than a quarter wavelength, you won’t get much effect from any self cancellation at the zero transition and the power coupled into the antenna is the same regardless of the angle.
As to why it’s not more effective when orthogonal to the wave: It doesn’t really interact with more of the wave front because it’s wider, at least not significantly.
This is because there is such a thing as an effective cross section (i think that’s what it’s called anyhow).
The electromagnetic waves have a certain “width” to them, so the antenna will interact with a significant horizontal width of the wavefront which again is much larger than the width of the antenna.
So any additional width of wavefront being coupled into the antenna is negligible.
This is all very unscientific and based on my very poor understanding of electromagnetic waves.
But that’s how it works in my head.
Maybe @prog wants to try a better explanation.
This video seems relevant: https://www.youtube.com/watch?v=H3VVY9vBGoI
Seems … effective aperture is the better keyword than effective cross-section.
Yeah not sure if that’s what really applies … as i said my understanding is poor but my guess about the topic would be:
Dimensions a lot shorter than the wavelength: The electromagnetic wave doesn’t care 