I hate that somehow all electostats have to take it on the chin for some old ML panel. My guess would be that the ML tested was an old beaming panel. The newer ML recently tested in stereophile and also the Quads have a much better power response. So my guess is that they would perform much better in the Harman listener preference lab.
I don't think the "beaming" would change with the newer panels, they seem mostly cosmetically changed to me. Clear instead of black "spars" between panel segments, different hole pattern (more but smaller holes) in the stators, same 30 degree cylindrical arc section, though, and essentially the same mylar diaphragm material (conductive coating may be improved, maybe it's thinner, or stretched differently).
I think, in the Harman Test, in a single speaker configuration, after being "trained" to listen for "wide smooth dispersion" the cones and domes may have a little advantage.
If two speakers were used, there might be some "filling the gaps" coming from the other speaker.
I don't hear much difference in the two-speaker comb effects here between the ML and JBL. But I don't have everyone else's HF hearing.
If I put on some correlated (mono) pink noise using two speakers, both ML and JBL have about the same "problem" with combing effect when moving my head left and right on and off axis and through the centerline.
I hear sheee shooo shough shoo sheee about equivalently with either while moving back and forth.
When measured with a single mic, the ML may be at a disadvantage - they interfere with themselves because the source is not so much a point or line but a several inches wide panel. JA usually notes this in his measurement comments. The sound from the sides of the panel arrive at different times than from the center of the panel, so there's some comb filtering with itself.
"Real" sound sources are rarely "point" sources, so...??? A firecracker is, a piano is not, a guitar is not, a violin is not, a drum is not, a voice may be, a choir is not, and so on...
Measurements from Sterophile:
20+ years ago - 15 inch wide panel - the generation before my speakers, probably about the same, though. Crossover at 125Hz 9off scale from the measuring range) and 180Hz on mine (would also be off scale):
ML Monti - newest speaker with the modern X-Stat panel with comparable measuring display, slightly narrower panel (11.3"), 340Hz crossover:
I don't see a radical difference, maybe you do...
A JBL cone and horn, with what would
look like more limited horizontal dispersion, 3-way with 750 and 3khz crossovers:
About the same rate of attenuation off-axis, but smoother looking in the measurement, except at the mid-tweeter cross.
The newest ML Renaissance 15 - JA didn't have all his tools, so not the same display, and not normalized to the on-axis measurement, 15" wide panel:
--
When I listen to my little JBL 308 vs ML critically at the sweet spot, it's like switching on some electronic "stereo enhancement" mode with the JBL, but if it
were something I'd switched on, I would turn it back off after a few minutes. It is probably room reflections doing it - spatially vague to my untrained ears. ML makes what initally sounds much narrower, but isn't, more like "farther away", and to me, much more stable and lifelike.
Off axis they're fine, I listen to either set of speakers and acclimate very quickly. Listening to JBLs off-axis right now, they sound fine.