> Using the same elements and a cable, a 18MHz dipole is moved slightly in its position.
Moved again for a couple of feet or so.
gnuplot> load "gnuplot1.txt"; Freq [MHz]=18.1 V1=2.511 V2=2.751 Cursor 1=-5.6e-009 Cursor 2=-1.5e-008 vratio=1.09557945041816 phase1 [deg]=-36.4896 phase2 [deg]=-97.74 abs(gamma)=0.66232897357976 swr=4.92292451384606 cz={16.7351989530969, -38.8485688397987}
The impedance was previously cz={13.4238815578089, -32.7726324246538}, when the antenna was in its old position.
The new measured impedance of cz={16.7351989530969, -38.8485688397987} gives, considering the coax cable lenght of 20m, Zant={11.73, +19.24}, which happens to be almost on the g=1 circle.
Therefore, an open stub of length 1881mm at the antenna feed point shall give you Z=43.28 ohm and SWR=1.16.
With a stub of length 2000mm, the following results are obtained.
gnuplot> load "gnuplot1.txt" Freq [MHz]=18.1 V1=2.463 V2=1.809 Cursor 1=3e-009 Cursor 2=-5.6e-009 vratio=0.734470158343484 phase1 [deg]=19.548 phase2 [deg]=-36.4896 abs(gamma)=0.393920800623182 swr=2.29989876249909 cz={23.8530134489064, 13.8771809482584}
The impedance of cz={23.8530134489064, 13.8771809482584} gives, Zant={37.947, -35.339}.
With a stub of length 1900mm, the following results are obtained.
gnuplot> load "gnuplot1.txt" Freq [MHz]=18.1 V1=2.401 V2=1.491 Cursor 1=8e-010 Cursor 2=-1.4e-009 vratio=0.620991253644315 phase1 [deg]=5.2128 phase2 [deg]=-9.1224 abs(gamma)=0.385725696138712 swr=2.25587443171256 cz={22.2605406131682, 2.9509421122626}
The impedance of cz={22.2605406131682, 2.9509421122626} gives, Zant={67.328, +45.316}.
At 18.1MHz, CMX-200 tells the forward power of 6W and the reflected power of 0.7W, which means abs(gamma)=sqrt(0.7/6)=0.341 and swr=2.04.