I'm a fan of Floyd Toole, who retired from Harmon as V.P. of acoustical engineering after 26 years with the National Research Center (formerly known as the Canadian Research Center). He invented much of the leading edge measurement techniques used today and has published in the Audio Engineering Society. Amir also used to work for Harmon and invested $100,000 U.S.D. of his own money into his modern testing setup, called Kippel, something that very few manufacturers would even bother learning how to use. Kippel allows for very accurate measurements that closely correlate to what is heard from large panels of trained listeners using double blind testing, again something that almost no manufacturers ever conduct.
Over the years Toole has done both modern testing and extensive double blind testing, which puts him in extremely rarified air. Amir is trying to follow in Toole's footsteps, but doesn't have a panel of trained listeners available. He gets most his speakers from individuals who eagerly loan them to him, which he states. From what I've read of his loudspeaker reviews he has quirks, he uses various Pink Panther figurines to characterize his overall opinions of products he tests which I find very hard to interpret, and likes his music really loud and bass heavy. Which means an inefficient small monitor isn't going to be his cup of tea.
Toole's measurements have correlated well with those panels of trained listeners. By the way, he has also researched the types who make the best listeners and found audio engineers some of the best while reviewers and marketing types to be among the worst. So much for the golden ears most of us love to glow over. So with his background and the extensive research done at Harmon it shouldn't surprise anyone that Revel and JBL loudspeakers score well with Amir. He discloses his former company association with every review of a Harmon product. I see no agenda to sell Harmon on his part. As Harmon follows Toole's methods I'd expect their loudspeakers to test well and helps explain why Amir has tested so many have volunteered their speakers for testing.
As far as listening or measuring first, a little background might be helpful. Again with $100k invested and high correlations with double blind listening panels it makes sense that measuring be done first. Once setup the Kippel automatically completes testing in about 2 hours, while Amir can do other things, like test DACs or write up a review. He apparently takes than two hours to audition afterwards, depending on how worthy of his time the speaker is. Note that his listens to just one speaker at a time, as recommended from Toole's research.
Jim is a very nice guy, but like others on Audio Circle, he runs a small company lacks a Kippel, anechoic chamber, or large well trained listening panel setup in double blind testing that would be ideal. So it should be of little surprise that one of his lesser products doesn't hold up to serious scrutiny.