Ok, I was talking about internal balance.
I see two ways of dealing with it.
1. Just accept that the same order is going to be followed, but don't deceive the player letting him think that he has options.
In this case, it would be better to reorder the policies so they are offered in the same order we are going to take them. For example, the Progress happiness policy is always taken last, no matter the other policies, then it should be available only when all the other policies have already been taken.
2. Make the options meaningful. This is harder to achieve since we need to know why some policies are always taken last. In this case, I think both the free worker and the production per city are on equal terms, but science from connection is just leagues better than the other tier 2 Progress policies. Why? Because it is giving a very good flat bonus, while the others are mostly scalable (they are better later). Culture from building buildings is nice, but culture comes from researching technologies mostly, so that's a bit secondary.
Rearranging the policies order might work. Sorry I don't remember the policy names.
Left side: Production --> Culture from buildings.
Right side: Free worker AND Happiness --> Science from connections
This way we are forced to pick the happiness policy first if we want science from connections, effectively making the science from connections a tier 3 policy. This makes getting culture from buildings more appealing since it is still a tier 2 policy. We can get culture from both policies, but happening at tier 2 instead of tier 3 makes it much better in comparison.
The problem with this approach is that it is a nerf to Progress if we keep current values, so some values should be raised as a tradeoff.
Option 3....ain't broke don't fix it
