The colours in the images supplied are unreliable from the viewpoint of assessing the colour in real life.
Just look at the shade of blue of the sky, it’s five different colours in the five images. Which one is real, if any?
So the colouring of the bird in the images cannot be relied upon for an accurate assessment.
Sony, or any other manufacturer cannot deliver 100% accuracy of colour when shooting a moving bird at range against a blue sky with the object lit from above.
Of course the dark tipped bunched primaries are also supportive of the identification as Kestrel.
Often (usually) it is best to put down the camera, and just watch the bird, when the identification will soon be apparent.
Retrospectively trying to identify a bird on one (with respect, and taking lighting, distance etc. into account) poor quality image, without any clues on size, jizz, accurate colouration, is usually a pointless exercise.
The irony of this exercise is that “I never saw the bird!” with the naked eye!
Only when editing the SD card some 3 hours later, that I spotted the image in the bottom left hand corner of the frames.
Having used the camera for several years now I’ve become accustomed to its quirks and merits, particularly when shooting at distant birds in the sky.
Of course I’ve “lightened” the image to extract as much resolution as possible, this you can see with the different shades of blue in the sky.
That withstanding you might note, that the subject’s body colour to include contrasting white head is relatively consistent throughout.
Dependant on the strength of the light source relative to the subject, it has consistently given the correct body colour even at distance, indeed it’s surprised me time and time again.
However on the other hand, it has failed (not infrequently) to register the “subject” in the viewfinder to the point of refusing to shutter release!
This prompts me to keep hitting the shutter till I see an image in the VF.
My initial assessment was of it looking more like female RFF than Kestrel, then I began to wonder why the pixcellated odds might be…favouring RFF over Kestrel with an almost gingery tint to the underbody against such a pale/white head.
Had it been a Kestrel, it would have looked like one…imo it just doesn’t.