I still use the rule of thirds constantly, and I still tell people learning to shoot to lean on it early. It's a genuinely good default. What I've stopped doing is treating it as a rule rather than a starting assumption, because some of the strongest frames I've shot break it deliberately and would be worse if I'd forced a subject onto a grid intersection.
When centering actually works better
Symmetrical subjects, a face shot straight on, an architectural detail with genuine bilateral symmetry, tend to look intentional and strong when centered, and slightly off when nudged to a third. The rule of thirds assumes asymmetry is generally more dynamic, which is true often enough to be useful advice, but it stops being true the moment your subject itself is symmetrical. Fighting the subject's own geometry to satisfy a compositional guideline usually looks worse, not better.
Negative space as the actual subject
Some of my favorite event shots put the subject in a small corner of the frame with a huge amount of empty space around them, well outside anything a thirds grid would suggest. That works when the emptiness itself is doing narrative work, isolation, scale, a quiet moment inside a chaotic event. It fails when it's just an accident of not committing to a frame. The test I use: if I can't articulate why the empty space matters, it's probably a mistake, not a choice.
- Center symmetrical subjects, don't force asymmetry onto them
- Use negative space when it's doing narrative work, not by accident
- If you can't explain the choice, it's probably not actually a choice yet
The honest test for any broken rule
Anytime I catch myself breaking a compositional convention, I ask whether I could defend the choice to someone looking at the frame cold. If the answer is a shrug, I'm probably just being lazy and dressing it up as intentional. If I can point to exactly what the break is doing for the image, it usually holds up. That test has saved me from a lot of frames I initially liked for the wrong reasons.