"I find both sean's and the bet smith adjustment tool to work fairly well. I'd be more concerned I suppose if I cared if I was 1 or 2 points off."
+1.
A while ago, I compared the hydrometer reading to the refractometer reading, and found that the hydro was consistently 4-5% lower. In checking with a 10% sugar solution (20 grams table sugar brought to a volume of 200 ml with distilled water) the hydro read 1.034, and the refr read 1.038. Since then, I use only the refractometer. I am looking for consistency and replicability, rather than absolute accuracy