N1ql indexing sequence

the value of equalKeys that match predicate in the leading index will be the continuous in the same place ,that will speed up the indexScan.so equalKeys before IN keys is better.
Am I right?