can someone please explain to me why it makes a difference to alignment whether I chose this setup -
basically connecting three “Transform Geo” nodes in a row…

…okay now the results are identically with both methods… but why?

Does this have to do with the Euler Rotationaxis Order ?
By defining a pivot I use rotational constraints… but wahy is the “TG-Node” evaluating any differently than the “AEtV - Node” ?

Because of your input values… try having X,Y and Z components for all input values and the results will be different again. That should be the clue. (And sorry for the “red herring”, but the purpose of this red herring was to make you think about the input values specifically.)

Yes.

Because they are fulfilling different functions. The TG-Node chain will give each operation equal precedence (and each operation is effectively in global space), but the AEtV-Node chain will only give the last operation precedence (and each operation is effectively in local space).

Spend some more time experimenting on this and it should become clear.