What John (MidLandOne) said.
I’d be happy to steer to within 5 degrees of my intended compass
heading; whereas Maggie seemed to be able to steer incredibly accurately (±1-2 degrees?).
Given that the typical steering compass
is only graduated in 5 degree increments, and identified in 30 degree increments, anything smaller than a 5 degree difference is a visual estimation.
Generally estimated values, Set & Drift introduce further (unknown) errors into our reckonings.
Any calculation should only be assumed to be as accurate (at most) as it's least accurate component value.
Of course, there’s no reason to deliberately add MORE inaccuracies/unknowns to our plotting & steering computations.