My attention was mostly on how he argued the effectiveness of math. In fact, I didn't entirely agree with him.
Regarding what you say here, I see your point. I think the notion 經濟人 can hardly be mathematicalized precisely because it isn't a 物貭(色) since it involves the human Psyche. However, an economy as a whole heavily involves物貭 , so mathematics still has some minor success in economics. Yet in disciplines like psychology math is often a foreigner. This was how my friend, a psychiatrist, asked me every once a while: Is a mathematicalization of psychology possible? , and my answer was always a No.