I encourage you to read the short article about Vienna, which explicitly addresses the concern of concentrating poverty. Vienna has avoided this by building good affordable housing in desirable locations which is owned by the city government, and then they only require people to be poor upon entering the housing. As their income rises, which is a predictable effect of affordable housing, they are allowed to stay. This means that the housing is mixed income rather than concentrated poverty. The city owns or controls approximately half of all available housing units in the city, which helps keep private rents reasonable as well through market pressure. https://www.huduser.gov/portal/pdredge/pdr_edge_featd_articl...

I think the research is out there to support these policies, whether the data is an RCT or something else, as I hope the above argument highlights. If you remain neutral while you wait for an RCT, there can be real world consequences for vulnerable people.

I do agree that the US is terrible at building transportation infrastructure, but as advocates like Strong Towns point out, we're actually bankrupting local towns with car-dependent sprawl (due to high tax burdens to maintain the infrastructure) so we really must improve that one regardless. That it will benefit the poor is simply another reason to accelerate these changes.

Sorry my idea isn't to sit out on the sidelines.

It's the difference between "Look at all these policies that reduce crime." to "Here are some good ideas for policies we should test out in as controlled an environment as possible to see if they reduce crime".