Interesting, I always kinda felt that measurements were only valid when comparing those of the same tester/environment/protocol (and even speaker configuration) and your comments seems to support this.
Imagine two individuals that had never heard the BMR, one only reads the NRC graph and the other only reads your graph. They'd both argue endlessly as to how the speakers sound in real life (neither having ever heard them). This is why I like to look at graphs but always trust my ears first and foremost.
Quick question, if the hump in JAs measurements is due to the splicing of the near field measurement and flat speaker will always measure with a hump in the lower end, how can he get a "flat" line with the LSiM 703? Something doesn't jibe. Especially after knowing that they are heavy on the low end.