Improving graphics while leaving gameplay stagnant is a really easy trap for a game team to fall into, I think, because it's easer in a sense to work on graphics. It's generally pretty clear what you need to do to improve graphics in a game: use more triangles, higher-res textures, more sophisticated lighting/shading models, more elaborate particle effects, etc. That doesn't mean it's not a lot of work, but you basically know what you need to do to get the graphics to a certain level, and you can make steady progress.
With gameplay, though, it's more of a guess-and-test business - you can't really predict what's going to be fun or not, so it's easy to invest lots of time in prototyping and testing different gameplay ideas, and end up not actually making much progress. I'm not a game designer so maybe it's not quite as bad as that if you're more experienced in that area, but that's the impression I get, at least.
It really takes a lot of discipline for a team to keep focused on the hard problem of improving the core gameplay, and not get sidetracked too much into the easier problems of improving the graphics, presentation, and so forth. Not that those things are irrelevant, but it's tempting to spend a lot of time on them, and then you end up not spending enough time on the core gameplay.