That's a fair point. Here are the left outer TRS-1007 traces at 192 Hz, followed by the right inner, which should be worse.
Pre-peak
View attachment 515932View attachment 515933View attachment 515934
Peak
View attachment 515935View attachment 515936View attachment 515937
Right Inner
Pre-peak
View attachment 515945View attachment 515946View attachment 515947
Peak
View attachment 515949View attachment 515950View attachment 515951
Actually, where I did find something on cross-talk was at the release phenomenon at the dip post-peak :-
View attachment 515958View attachment 515959View attachment 515960
I amplified by 30 dB
View attachment 515961View attachment 515962
This is a part of the sweep where the cross-talk is still down 25 dB to the fundamental - on the the post-peak red trace in the sweep below ...
View attachment 515963
... so at 55 dB below the fundamental there are more jagged peaks and troughs.
Where I then become confused is what actually is "mistracking". What the term suggests is that inevitably the stylus not following the groove accurately, possibly causing damage to the stylus or the grooves. Stylus/cantilever misalignment can be one cause, best treated by improving alignment. One sign of good alignment is low cross-talk.
However, cross-talk also results from electrical interference between channels, in the coils, tonearm wiring etc.
In the MP-700 case, it may also be a resonance phenomenon affecting the moving magnet interaction with the coils, at the cantilever-tie-wire connection. As the stylus-cantilever/magnet-tie-wire assembly frequency approaches the tie-wire's resonant frequency range, tie-wire resonance transfers energy back into the cantilever/magnet and generates spurious signal in both the fundamental and the unmodulated channel. This energy transfer amplifies the coil output and causes a peak in frequency response and cross-talk in the "captured" frequency range. The stylus then drives the cantilever out of the resonant frequency range and there is a "catch-up" where the cantilever/magnet-tie-wire resonance hunts from the tie-wire resonance frequency to the stylus-generated frequency, with a sudden drop in output at the fundamental frequency, until the cantilever is reentrained by the stylus. Cross-talk peaks and dips actually slightly follow the output peaks and dips, during the above transitions. Maybe this is spurious signal during hunting between frequencies, at a guess. It perhaps also makes it more likely to be an electromagnetic phenomenon, as a stylus interaction with the grooves (non-grooves in the unmodulated channel) might be anticipated to generate peaks and falls in cross-talk in lock-step with output.
Alignment, resonance and electromagnetic phenomena may all play a role in crosstalk, which cannot simply be reduced only to inaccurate tracking of vinyl grooves. I am not convinced that a breakdown in crosstalk at the fundamental frequency always represents physical mistracking, especially since it's a break up of a signal that shouldn't even be there in the first place. You have to consider the possibilities why it is there and what might affect it. Then think of experiments.