Pseudo-Mono?

Any empirical testing done to confirm that, or is it anecdotal?

Should be easy with a DAW. Record a short piano piece in the sequencer, then record the audio in the DAW in stereo. Record same SMF again in mono (from the mono out of the arranger)... Convert the stereo recording in the DAW to mono... Compare.

Normalize BOTH tracks to ensure identical levels. That ought to make it apples to apples. It'll be interesting to hear the results (hint, hint!)...