Doesn't matter which day it is. The inverse Lincoln equation is still probably the most practical:
S = (668 - Sqrt(668^2 - 820*(463 + P)))/410
Or, if all you want is a rough estimate, multiply Plato (Brix) by 4, divide by 1000 and add 1 e.g. 7 Plato ~ 1.028 SG
Neither of those solutions would work with a refractometer, where the refractive index is affected by the presence of alcohol, resulting in a Brix reading that is very inaccurate.
Using the example of a beer that starts with 10 Brix, and finishes at 5 Brix (and uses a wort correction factor of 1.04), Sean Terril's calculator gives
OG = 1.0384, FG = 1.01, and ABV as 3.6%
Promash gives
OG = 1.03844, FG = 1.00862, and ABV = 3.91%
Beersmith gives
OG = 1.039, FG = 1.008, and ABV = 4.02%
These compare to your readings of
OG = ?, FG = 1.0197, and ABV = 2.65%
I must admit, I've never found any of these calculators to be particularly accurate, but your FG calculation is clearly wrong. I agree it would be correct if you were using a saccharometer, but not with a refractometer which requires a correction because of the presence of alcohol.
Because of this using the inverse Lincoln equation is not possible with a refractometer once fermentation has started.
-a.