Why should you pay attention to physical states of all reactants and products when writing equilibrium constant expressions?

Answer 1

Because they matter. The usual way to write an equilibrium constant expression is to use the concentrations, but that only really applies to solutions.

For gasses, the partial pressure is a pretty good equivalent (technically, the fugacity should be used instead, but at the level I'm guessing you're at based on the question that's a pretty fine distinction). For solids and pure liquids, it's more complicated. In practice, there's going to be a dependence on the surface area, since the reaction will only take place at the surface (unless the liquid is also the solvent, in which case it takes place everywhere). However, for simplicity's sake, they're usually just assumed to have a concentration of unity (equivalent to leaving them out of the equilibrium constant expression entirely). In general, this works pretty well in most cases, with possibly an added factor to represent surface area if that changes dramatically from one run of the experiment to another (for example, if you finely powder the solid instead of just leaving it in a single compact chunk).