The intent is to decrease cost by decreasing the total number of
designs and increasing the number of launches per design.

What reduces cost is competition. So I'd definitely try to keep at
least 2 rockets in each payload category. This is the premise behind
the EELV program (with Delta IV and Atlas 5 being the two
competitors).

Having said that, there is a glut of rockets relative to the number of
customers. This is mainly driven by each nation's desire to have
their own rather than buy abroad. Is this changing with the Starsem
deal to launch Soyuz from French Guyana, Sea Launch, Atlas 5's
licensing/purchase of a Russian engine, etc? Time will tell whether
globalized launch companies become the norm, or remain somewhat at the
margins of the industry.

I chose the Space Shuttle over the Soyuz simply because the
Space Shuttle is more flexible.

What's your goal? Cost or flexibility? I'm not sure even its
defenders would say that shuttle is likely to win on the former.