Why bother with a luminance histogram at all? If you haven’t blown out R,G, or B, then you’re fine. Contrariwise, you can blow out one of the colors, and still have a luminance histogram that looks ok.

The real thing you’re looking for is whether any photo receptors are blown out. Consider measuring liquids. If your measuring container can hold only a cup, and more than that gets poured into it, how do you know how much was poured? (answer: you can’t — that information is gone). The same thing happens with photons coming into a photo site.

On the other end of the scale, when a site doesn’t get enough photons, the few that arrive there are overwhelmed by noise (of different types, but some of it is from photon randomness — so there’s no way to engineer it away completely). So you want the darkest parts of your photo that aren’t going to be featureless black to be above your ‘noise floor’.

The easiest way to satisfy both of these restraints is to try to get the RGB histograms as close to the right-side border as you can, without crawling up that side (meaning you’ve lost some information to overflow) — and take your pictures in raw format so you can make use of all that information in post.

One of the reasons I encourage beginners to start with mirrorless cameras is that those cameras typically have live histograms — and DSLRs don’t, other than in live view (which is, essentially mirrorless mode.

BTW/FWIW, if you’re using a Nikon DSLR, you can look at RGB histograms as you zoom in in playback/review mode. The histogram will change to reflect just the part that you’re zoomed in to. Handy for shots of things with bright colors (e.g. flowers), to make sure you haven’t blown out a color. Blown colors are essentially unfixable in post