the average precession for the major-axis of the Earth-Moon
barycenter is in the region of:
problem being that the observational data could not satisfy
the requirements for the effect on the Earth-Moon barycenter
of Uranus. We would actually need a sample 924
years for Uranus, but is more closely aggregated to 913
years to properly account for Jupiter and Saturn. And that
is outside any observational time-frame. But notice that in
the earlier example that started at 1900,
the 59 year average is very close to the
237 year sample that began at 1773:
per year (as/Ey).
Scenario  is in 1500
second quanta. This causes a computational error-margin of
about 0.26 as/Ey over the 237
orbits of Earth. Scenario  and scenario
 are 10 and 100
times more accurate respectively. The latter would take a
40-day process to evolve full-time on my
Windows 10 glorified pocket calculator, so
you could get better results than me when using my software.
If you do extract better Perihelion Precession data from OGS15,
please let me know by contacting me here: cosmology.africamotion.net.
Though you may want to first see how to operate the algorithm
I prefer to openly express the uncertainty in the quantities,
because an open methodology always offers transparent room
for improvement. This is a process of increasing accuracy,
never a final answer. It is dialectic methodology, not dogma.
The Earth's orbit is becoming more circular. This is true
also of Jupiter, Saturn, Uranus and Neptune. Mercury and Mars,
the least massive bodies, and the most eccentric orbits, are
however becoming more eccentric by contrast. Mars' orbit is
analyzed in detail in the next section: