Dude we're in strange territory. The electron latency is irrelevant, the DAC/ADC process means 56k has a minimum of 60ms one direction, that is not a plant decision that is a physical limitation. What I'm saying is that because of DSL's typical implementation it is often inferior to cable in latency to the gateway. You have to consult with the provider to get it to even compete with cable, routing differences granted.
And yet again, I will state that the decision to use DSL interleaving, esp with higher queue depths which is the tech that adds that latency, is plant decision. You can run the line bare or in fast path at the risk packet corruption issues.
Right now I am clocking several DSL lines, and this exact second Chicago to Milwaukee is 12.5ms. It is obviously not running interleaving which adds 5 - 60ms.
The couple of cable connections I just looked at are floating at 15-20ms. These are Chicago to Chicago so I would have hoped that it would have been lower than Chicago to Milwaukee but they are not at the moment. Could be load or other plant design like an overloaded node. At this point they are effectively the same. Cable at least could go faster if we wanted but that minimally effects latency.
Connecting via modem to the back maintenance side... 38ms, 56 byte pings. Certainly doesn't look like the 120+ms pings you are telling I must have.