Jeffro, your interpretation of the chart is correct if Rehov V was destroyed in 900 BCE. But if it was destroyed ca. 925 BCE (as the article 14C dates from Tel Rehov: Iron-Age chronology, pharaohs, and Hebrew kings proposes) then the geomagnetic variation curve would be shifted 25 years to the left. The probability distribution of ages is dependent on the historically dated chronological anchors, and when there is no such anchor and the date is uncertain (as in the case of Beth-shean, Rehov V and Tevet VII) the curve is unreliable.
Further, there are different ways of calculating the geomagnetic variation curve. As the article says:
According to one approach, a probability distribution of ages is calculated using a curve constructed independently from the object being dated. According to another approach, adopted here following Livermore et al., the datum to be dated is included in the Bayesian procedure used to produce the curve. (my italics)
You are trying to use this new method to prove or falsify certain dates. It doesn't do that.