[TorpX]

My tactics are similer to aanker's.

If you are plotting in conditions that make for large error margins, then you must make more observations and "fair the plot", to obtain good data. In effect you will obtain average values of speed, course, and range.

If, for example, you are operating in difficult conditions, and you estimate the observed speed error to be, after 3 observations, to be 20%, then if you instead make 12 observations, you could expect the error to be reduced to 10%. Real-life crews usually had to do this anyway, as the stadimeter was limited in it's accuracy, positive identification of ships was problematic, and ships were often zigging.