Here's the way the VSWR shook out when the smoke cleared after 3 hours of snipping, measuring, soldering, measuring, snipping, cussing, soldering, measuring...

The immediate reaction is to look at the curve of the beam when used with the director and think, "That's awful-- look at that, it's resonating too high."
Well, yeah-- that's exactly what I'd think.
BUT- every adjustment of the element lengths interacted with everything else, and what optimized the curve for the reflector totally jacked up the curve for the director, and vice-versa. It's possible a different element spacing could have helped that, but I had decided on a 10' spacing since the configuration I plan to use most often is the DE/Reflector, and figured if I could find a decent compromise for the reverse direction, that would be OK.
To be honest, I did not expect the curves to be that different, but there you have the unvarnished truth.
What I was after was a usable antenna across the 20m phone band, and all things considered, this is acceptable.