Photos
Two light sources, one repeatable
Direct flash flattens and varies with distance; window light on an overcast day is the same every time you use it.
Guides on Photos: A complete protocol for comparable submissions, Everything about submitting more than one image, How to find out how much of your score is you
For a rating submission, window light beats direct flash - not because it looks nicer, but because you can get back to it a second time. Both light a subject well enough for a tool to work with; only one repeats reliably.
Why flash is hard to repeat
A phone's flash is a point source sitting right next to the lens, and its effect changes sharply with distance - a few centimetres closer or further and the exposure and the shadow pattern both shift, because the light is falling off fast at close range. It also flattens surface detail, since a light fired from almost the same point as the lens casts almost no shadow at all. Two flash photos taken at slightly different distances can look meaningfully different from each other even when nothing about the subject changed, which makes flash a poor choice whenever the goal is a submission worth comparing against another one.
Why window light holds still
Light through a window on an overcast day is diffuse and roughly constant for as long as the cloud cover lasts, and it does not depend on exactly how far the subject is from it the way a point source does. It is also directional in a way that is stable and repeatable: pick a spot relative to the window, and the shadow pattern it produces will be close to identical the next time, and the time after that. Its colour is fairly stable too: the standard illuminant D65 in ISO/CIE 11664-2:2022 represents average daylight at a correlated colour temperature of about 6,500 K, against about 2,856 K for tungsten lighting, which is why mixing a window with a bedside lamp tints the frame unevenly. The general case for fixing light direction applies to any light source; window light just happens to be the one most people can hold constant without buying anything.
The rule, stated plainly
Neither source is more accurate; a tool has no concept of "correct" lighting, only whatever the frame shows it. What matters is that the same source, from the same direction, produces close to the same image each time - and of the two, window light is the one that survives being repeated by someone who is not paying close attention to distance.
Where this sits
This is a narrow point inside a larger one: what a model actually does with whatever detail the light exposed is the mechanism, and the gear question underneath it is a single lamp or a window rather than equipment. The same preference holds if a photo is headed to a human judge rather than a model - a reproducible light source is worth keeping either way. Rate Cock's photo guide recommends window light for the same reason argued here: not because it flatters, but because it repeats.