A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Frank Reed
Date: 2020 Jan 30, 17:57 -0800
Jim, you wrote:
"After all that, I graphed the results and two things surprised me. That there was so much variability in my individual sights, despite paying particularly close attention. Getting the "exact altitude" is harder than it looks! But the other thing that I concluded was that I Anti-spoof is a great tool for helping hone shooting skills. The averaged line looks like a good fit overall. Is it? "
In fact, your sights look very good. What you should look at it is the average and standard deviation of your nineteen cells H6 through H24, which show the difference between the app-predicted altitude and your observed altitude. I see you did calculate the average and got 0.05'. This is not significantly different from zero. And that's excellent.
You did not directly calculate the standard deviation (I don't see it at least). But if I enter STDEV(H6:H24), I get 0.31'. And that's really very good. I'm not sure that the line that you have graphed has any real significance. I would say you don't need that.
Just considering the two parameters we find from your sights, an average near 0.0' and s.d. of 0.31', you should see that you should be able to get an excellent fix by averaging just a handful of sights. Random error drops off in inverse proportion to the square root of the number of sights. So imagine you're shooting the latitude at local noon. If you take just nine sights and average, you could expect your latitude to be accurate to the nearest tenth of a minute (the mean error is zero and 0.31/sqrt(9)=0.1').