But does that mean the preference order will necessarily always stay the same?
When comparing the scores in the
Rega vs KEF vs Quad test, the scores of the lowest-ranking speaker (the narrow-pattern, dipolar Quad) improved dramatically in stereo, while the scores of the others did not move much. The score of the Quad did not move up enough for its ranking versus the Rega and KEF to change, and this is what people have focused on as evidence that mono listening preference can reliably be extrapolated to stereo listening preference. However IF the Quads had been compared to other speakers which did not score as high as the Rega and KEF in mono, THEN the ranking easily could have changed.
To me, the more interesting aspect of comparing mono vs stereo scores of the Rega, KEF and Quad is NOT that the ranking stayed the same, but that the Quad's score improved so dramatically. I'm curious about the technical and/or psychoacoustic principle(s) behind this.