Thank you very much for your reply. I followed your instructions to evaluate the directional deformation and got a similar overview like in your first screenshot. Unfortunately, the export via right click "Export Text File" just provides the data of the "by" timestep I chose when setting up this result, this is the text file:
The manual copy-paste process of the table (right in your first screenshot) will kill some time... Do you know of a scripting way to do this?
In regards to your last question: I would like to compare certain results of nodes at the path with each other in the time domain. Of course I would have to split the path in defined paths (or choose single nodes from it) and work with means unless I visualize the data in a 3D way, what also is a possibility at the moment. The export via a path geometry seems easier to me than twenty times a single node, at least without scripting.