I haven't used RethinkDB in ages, but when I did, I used it exclusively for a year.
It took a few weeks for ReQL to really sink in, but once it did, I much preferred it to SQL. As my sibling comment says, its very linear: do this then this then this then this. It also feels very much like sequence processing in functional languages (and I was using Clojure at the time, so it really meshed very well).
I also found it much easier (than SQL, at least in my limited experience) to reason about performance because I knew what it was doing. Is this looked up by index? Yes, because that's the command I used.
So, yeah, ReQL actually ended up being one of the things I liked most about RethinkDB.