I know; but the chart's source data almost certainly doesn't use figures that are calculated in the same way either. At best, they'll have been manually correct

I know; but the chart's source data almost certainly doesn't use figures that are calculated in the same way either. At best, they'll have been manually corrected for estimated biases. (Will they have correctly accounted for how many people get shifted off into disability payments and other health-related benefits that are used to massage unemployment figures? I have my doubts.) My point was more that a suspiciously precise number of 5.8% was chosen as a magical barometer of goodness, justifying a switch in hue, not just brightness or saturation. The magic number of 5.8% doesn't seem to have support elsewhere.