This is correct. MEME-ChIP automatically runs FIMO using the --parse-genomic-coord option. That option tries to parse genome coordinates from the sequence names, assuming they are using the UCSC sequence coordinates convention. If the --parse-genomic-coord option is not given, or if the sequence name doesn't seem to match the UCSC conventions, the first letter in the sequence is assumed to be at position 1.
There was a slight change to the FIMO plain text output in the 4.11.4 release. We added the 'alternate-id' column. The 4.11.4 versions of MEME and DREME identified motifs by their consensus sequence, as well as the order in which they were discovered.