So my Directional Coupler performed fairly well. Here is the output from my spectrum analyzer. Basically one trace is for the forward port when the antenna is matched, the other is when the antenna is an open circuit. Then the other two traces are the same condition on the reflected port.
As you can see all the powers show a nice flat -20db coupling. Only exception is the one cyan trace that shows the reflected port when the antenna is matched. Since there are no reflections as expected the response is much less. It is also not linear which is to be expected.
Pretty similar results to my last attempt. Though going with fewer windings this time (8 instead of 12) seems to have results in a slight stronger coupling (-20 dBM rather than the -23 dBM I got before).
All in all I'd say its a success.