What properties make a good catalyst? Specifically, I am thinking about metal catalysts that do not (apparently) take part in a direct chemical reaction. Is there any general method of either finding a catalyst for a given chemical reaction or the reverse, finding a chemical reaction that might be catalysed by a given metal?
 A: I think the only 'general method' is trial and error.
A good catalyst


*

*is not expensive

*is not used up at a high rate (e.g. by poisoning by side reaction)

*enables high rates of chemical production


The physics of how catalysts work is that, of course, they are involved in the chemical reaction. They react with the starting materials and then enable the products to be produced through a lower energy pathway than would be possible without them. Thus...

Now in principle we could calculate with quantum mechanics the pathways with and without the catalyst, but this is a very significant challenge for a quantum mechanical calculation.
Even if we can make this calculation then there may be other factors (e.g. how easily the catalyst gets poisoned) that need to be taken into account to be a good catalyst. 
edit inspired by MSalters comment
High surface area is often a feature of catalysts and it may be sometimes that a high surface area material - like a zeolite - is used. Metals do not normally have very high surface areas, but they can be coated on high surface area supports to increase their surface area.
