Hi Potterman,
I'm going to take a stab at this....
The behavior of this screen changes depending if you are in live or historic mode.
In live mode, the metric values in the table are the current, last measurement values. The graphs are an average across all instances across the 8 minute live metric window.
In historic mode, the graph will re-scale depending on the size of your historic period. The table values are now weighted averages, or smart averages. So there is an algorithm that will average the metrics across the resolution period (168 minutes in your screen shot) and then weight the values so the high values are not averaged out. Depending on the metric, it might not be an average looking value, but the smart averaging will keep the high values making it look more like an addition value.
Now on to your screen shot....
The is sorted on R/T and there are more entries in the table since there is a scroll bar on the right. So if you sort by the stall column from high to low, then you should see the stalls values greater than 1.
Hope this helps, might not be a 100% answer since I can't scroll through your metrics.
Billy