Two properties of light that people conflate

No - a brighter photo is not a better submission, only a differently exposed one. Light has two separate properties, and brightness is the less important. Direction decides what the image shows in relief; exposure decides how much of the frame is legible, and it only needs to stay the same each time.

Fix exposure, do not maximise it

Exposure needs to land in a workable middle range and then stay there every time, not climb toward "brighter is safer." Overexposure blows out highlight detail the same way underexposure buries shadow detail - both destroy information the rubric would otherwise read, and neither is recoverable after the fact. The limit is physical: Debevec and Malik (1997) note that cameras "are not designed to capture even close to" the range between dark and bright found in real scenes, so pushing exposure up simply spends that narrow range differently. Image models notice the shift too - Hendrycks and Dietterich (2019) included brightness and contrast among the 15 everyday corruptions in their benchmark of how image classifiers hold up when a photo is degraded. A consistent, moderate exposure that repeats across sessions matters more than a bright one that varies, which is the same logic behind holding every non-subject variable constant: exposure is one more variable to fix, not one to optimise.

Direction is what actually moves the result

Where light comes from decides what the surface shows in relief - side light reveals texture and contour that flat, on-axis light flattens out entirely. Two photos at identical brightness but different light direction can produce meaningfully different reads on presentation-adjacent axes, because the model is responding to what the light lets it see, not to how many lumens hit the sensor. Direction deserves its own protocol line; brightness mostly just needs to stop moving around, the same way a fixed camera position removes drift elsewhere in the setup.

Where this shows up elsewhere

The same distinction exists in how a model reads an image generally: what is visible in the frame matters more to the read than how much total light was used to produce it. It also has nothing to do with a physical reading - measurement work cares about the tape and the reference object, not the lighting rig - so this is a scoring-specific concern rather than a general photography one. And it is a different conversation from what a person reviewing a photo might notice about lighting quality, where mood and flattery are legitimately part of the judgement in a way they are not for a rubric.

Set exposure once, keep it, and put the actual attention on where the light is coming from. That is the variable a service like Rate Cock is reading when it scores presentation, and it is the one worth getting right on purpose.