Of course, you are correct. I was a bit distracted for a moment... You don't open up from an incident shadow reading to "place" a shadow; that's what you do with a reflected-light meter. Thanks for the correction; I wouldn't want to mislead anyone.
What I wanted to point out is, if you just take an incident reading from the shadows, you will certainly get enough shadow detail. The question is: do you really want it? Many times you want darker areas in the shadows to be completely featureless, and the incident reading wants to place them higher. For me, this is "overexposure," which with LF B&W may not make much of a difference, but may be a consideration with smaller formats.
I can envision a scene with deep shadows that contain gray objects, maybe a darker brown or something, that would fall on or above Zone III with the incident shadow reading, but which I would really want to be rendered black in the final print, since they are the darkest areas in the scene.
I guess that's the real drawback of incident metering in my opinion, the inability to use the meter reading as a visualization tool. I use a spot meter primarily, and, despite all the inaccuracies inherent in the whole system, can fairly well determine where print values will be. Often I depart from "correct" exposure for expressive reasons. I just can't do that easily with an incident meter.
Also, one poster suggested taking an incident reading from the shaded and lit portions of the scene and averaging these. This would also be less than optimum, and, depending on the scene, could result in either over- or underexposure.
If I were using an incident meter, I would be using BTZS meter techniques for sure.