The disconnect is in attributing second order effects as completely dominating the first order effects. In the thought experiment where
poof the H1-Bs never happened and
poof were filled by qualified US citizens, totally different conclusions.
The first order effect difference: I claim effectively all of those positions could and would be filled. The "this is egregiously wrong" position claims that effectively none of those well paid jobs would be filled. I know why people will proactively seek a well paying job. I am not able to follow the position that no one will bother to seek those positions at all. Seems at odds with basic human motives to me.
With regard to the alluded second order effects, multiplier effects.
Consider the Disney situation where "about 250 Disney employees were told in late October that they would be laid off. Many of their jobs were transferred to immigrants on temporary visas for highly skilled technical workers, who were brought in by an outsourcing firm based in India."[1]
Sure the local Indian restaurant will prosper. But that is 250 households whose kids can longer afford to go to college. 250 home purchases vs 250 apartment rentals (folks on 2-3 yr visa rent not buy in general). So on and so forth.
On the surface, to me, 250 middle class families with well paying IT positions spins off far more positive effects than what we have now which is 250 US families with significant if not total household income loss, on unemployment etc. In no way is this compensated by having 250, lower paid, immigrants on temporary H1-B visas (again NYTimes phrasing). I see a net-net huge loss, others seem to see a net-net huge win.
[1] Quoted from NYTimes.