Hi Gordon,
I think there is a problem with attempting to compare the relative efficiencies of stonerowing and dragging, and that is the fact that in order to do so you must postulate the same small number of draggers as stonerowers. In truth though, there's a technical limit on the number of stonerowers you can use and no limit on the number of draggers. Ergo, given enough people, dragging would be far more effective than rowing and the fact the latter might be more efficient becomes of little consequence. I think your experiments tended to show that to be true.

Personally I think the dominant method was most likely to be quite an army, dragging, but using levering and stonerowing in stoney or marshy or awkward places. Unless of course they didn't have many people - in which case stonerowing might be the main choice.