NavList:

I personally do not think there is a hard/fast approach to this.  Each LOP is not necessarily going to be equal, despite what the mathematics say.  Environmental conditions can affect sight quality at different azimuths and the brightness of the star can make placing it on the horizon (real or artificial) more or less difficult. So my personal methodology has me evaluate each sight for its quality and then weight it accordingly.  Stars such as Nunki, anything below 15 degrees elevation, and the moon in general; I always have a low confidence in, just based on experience.

When I shoot stars at sea, I try to shoot the stars found in HO 249 Vol 1, along with any planets that may be available. If the horizon is particularly poor in one direction or another, tI will throw out the stars in that area (I make a mark on my sheet). If the horizon is good, I may shoot a body twice, once in early twilight, and a second time in later twilight.

About the only time I'm shooting rapid sights of the same body are for sunlines (three) and lunars (seven).

I don't often manually plot too much, since, for me, it adds additional error I don't want.  I let the computer/calculator crunch the numbers and then it spits out a MPP which I then plot by Lat and Long on the chart. My average positional error for a full round of starts has historically been about 0.8 nm at sea, with horizon quality being the biggest influence. With excellent shooting conditions (calm and a sharp horizon) I have fixed position within a ship length (200 meters) using a full round of stars.

Jeremy