Because film and digital fail differently at the extremes.

With negative film, shadows tend to have a steeper cutoff, while highlights roll off more gently. So it makes sense to expose enough for shadow detail and manage the highlights during development/printing.

With digital sensors, the opposite problem is more important: highlights have a hard ceiling. Once a highlight clips, that detail is gone. Shadows usually aren’t “gone” in the same way—you can often raise them later, though with more noise.

That’s why ETTR means exposing as brightly as possible while still preserving the brightest tones you care about. In practice, you meter for the brightest important area, not necessarily every specular highlight, and avoid clipping there. Then you can pull the overall exposure down in post.

So the short version is:

• negative film: protect shadow detail; highlights are more forgiving
• digital: protect important highlights; shadows are noisier but often recoverable

ETTR is really about maximizing captured signal without crossing digital’s hard highlight limit.