Do digital barometers not also use mercury?
My thought would be, rather than one measurement being less accurate than the other, is the sparsity of data points available. I would imagine that the pressure readings on those storms probably come in on the low side, given they didn't have a good way to tell where the exact center of circulation was. Not like they had satellites, and weather stations in every teeny tiny town that could easily report their findings. Telegraph was still relatively new technology then, remember. And I imagine with a storm that size with that kind of cold air, communication pretty much ground to a halt in the still rural south.