I had heard that removing a street can speed up traffic, but I hadn't really internalized it until I read:

When a network is not congested, adding a new street will indeed make things better. But in the case of congested networks, adding a new street probably makes things worse at least half the time, mathematicians say.

Half the time? Then there's a very cheap way to reduce congestion in our cities: test which roads speed up traffic when closed (or heavily tolled) for rush hour. If we can use simulations to get a good idea which ones to test, all the better.

That quote is from 1990, based on a 1983 paper. Why aren't we testing closures yet?