With the new/current (antlr4) parser this query no longer fails silently with incorrect results in qserv; it fails explicitly, because the antlr4 parser/IR factory rejects it: we have not added IR-creation support for multiplication in select yet (as in SELECT (a*b)).
I discussed this with Fritz Mueller in our team meeting, and I put a story point estimate on this issue, and changed the title & description to reflect the current work that needs to be done. We agreed this had been a critical bug because it was failing silently, returning incorrect results. Since now it fails explicitly and does not return results I changed the priority from critical to undefined. I changed it to a story: it's a feature we may need to support (needs to be prioritized & scheduled by Fritz), but was not part of the initial scope of work for the antlr4 parser conversion.