You are maybe of the opinion that psychology is not a science? MBTI is a personality psychology theory, and can be evaluated by scientific standards, and it doesn't fare very well.
There is "hard science" and there is "soft science".
If you put psychology to the same standards as math and physics, the whole field is likely to be classified as bunk.
But being a "science-fascist" when it comes to fields like human behavior is not a reasonable way to go about it, especially when the scientific community is still in the dark (for the most part) on how the human brain works.
The ancient Greeks had detailed astronomical charts that were used for accurate navigation, but they speculated that the stars were holes in the heavens from which the light escaped. Just because they were wrong about what stars were, doesn't mean that their knowledge of the stars' movements was useless.
Same thing with Myers Briggs. It does not give you an accurate scientific justifications why one type behaves differently from another. But in my experience it has been an
extremely useful tool when dealing with people and explaining their behavior. I will go as far as saying that it is the most useful knowledge that I have obtained in my entire life.
Typology is a science in it's infancy (even though it has been talked about in one way or another for millenia), and there will be plenty of theories and original assessments that will be proven ineffective and will be discarded as a whole or replaced by more effective methods. I do not follow the "cognitive functions" part of the Myers Briggs. I've had people argue with me that since I dismiss the cognitive function part of the theory, which is (according to some) is an "essential part of the Jungian typology", then I discount the whole MBTI as a valid theory. That argument... is idiotic. David Kiersey established a system that avoids dealing with cognitive functions that I find to be a lot more effective than some of the original theories of Myers Briggs. And I'm sure down the road, someone will propose a theory that improves on the original works of Kiersey himself.
But in conclusion, I just want to say that there are two types of people that show up on MBTI boards and grandiosely announce that MBTI is a "scam" or "no different from astrology": scientific fundamentalists who refuse to acknowledge the existence of anything that cannot be quantified by currently known scientific methods, and attention whores, who say things that are most likely to generate a controversy given the present crowd. And maybe it's just me, but I never got the impression that Victor is a science-type-of-guy.