It was worth a punt, but with testing a 20 mic subset there was no change - approx 15 degree tilt regardless of what knot settings I used (tried x=2 - 6, y=2-16).
…
Had time to quickly check two micrographs with CTFFIND5, where output is:
Estimated defocus values : 12829.99 , 12681.25 Angstroms
Estimated azimuth of astigmatism: -37.07 degrees
Tilt_axis, tilt angle : 187.43 , 33.00 degrees
Estimated defocus values : 18104.75 , 17911.55 Angstroms
Estimated azimuth of astigmatism: -44.30 degrees
Tilt_axis, tilt angle : 178.70 , 36.70 degrees
So maybe CTFFIND slightly overestimates.