RFE: Armchair Accuracy - Ways to reduce errors

[rgleason]

Bob Bossert wrote in an email 2/14/2025 in considering rgleason question/suggestion:

Armchair accuracy. - Excellent idea.

How about identifying the +/- error of the sight positional data based on statistical analysis. This gets at the question everyone should have when deciding to use Cel Nav plugin. And I think this analysis can be periodically done with synchronized with improvements (undetermined frequency).

1. I'd start with a good size statistical sample of sights for each body (stars as a group, aries, venus, mars, jupiter, saturn, moon, sun, polaris).
2. Compare plug in derived GHA and Declination to the Nautical Almanac (calculate mean difference and std deviation).
3. Some altitude adjustments (like dip, refraction, IE, barometer/temp don't need statistical analysis because OpenCPN uses the same simple math formulas used in Nautical Almanac...so no variance).
4. But there might be a couple "body specific" altitude adjustments should to be analyzed (Moon Semi-Diameter / Parallax was an example...and Venus and Mars has some additional small adjustments I haven't yet looked at).
5. Once the statistical analysis is completed, then the precision (avg error and std deviation) could be output in the Calculation form next to the GHA and Dec or other adjustment.
6. Maybe a maintainable table which someone can update periodically after re validating the "next year's" data.

Ways Navigators reduce errors

1. Take a run of sights (3 to 5) and selecting the best one to reduce. What I do is to input my run of sights into the plug-in, multi select the sights to see the lines of positions. Then, if one is way off, I won't use that rogue sight.
2. Take the avg of a run of sights and use the average for sight reduction.
3. Navigator loads all of his/her sights over time (at least 100), and determine their avg personal bias error.
   A navigator colleague of mine (ex US Navy), knows his avg error (Hs too high, or too low by decimal minutes). And then adjusts his sight to take into account his personal bias. You can load about 30 sights in the plug in. After 30, it gets a little dicey when you load the next sight. The plug in used to error out when you exceeded some limit. Now, it just hangs until the excess sights are deleted.

The best way to minimize position errors is to do a 3 different body sight within 20 minutes (I think a few nights ago, Sun, Moon, and Venus made a good 3 body sight...it was too cold where I was). The 3 bodies should be spread apart. Enter the 3 body sights and time. When the plug in draws the line of positions, it will be a triangle. Your fix is in the center of the triangle. If your boat was moving while doing the 3 body fix (each fix has a DR location),

You should be able to use the plug-in to advance the Line of position of the 1st and 2nd sight based on the distance traveled (60D = ST) to the 3rd sight's position. It is a fix if the 3 sights are made within 20 minutes, if more than 20 minutes, it is a running fix.