Steve-
Note on tensor strain readings.
I have noted tensor strain DECREASES on the tensor strain readings about 12 to 72 hours before at least 6 earthquakes on the Bay Area/ San Juan Bautista/ and Parkfield USGS Tensor Strain sites.
I noted two seperate descreases on the Bay Area(Oakland) which were followed by earthquakes of about 4.0 in Orinda, CA. Tensor strain decreases in San Juan Bautista have been followed somestimes by earthquakes, sometimes by creep, and sometimes neither earthquakes or creep.

On to Parkfield, about 2 months prior to the 6.0 in 2004 I put a prediction on Parkfield based on the tensor strain decrease. There wasn't an earthquake in Parkfield during the prediction window. But, there was a sizeable aftershock sequence to the San Simeon earthquake, which at the time Petra pointed out might have been related to the decrease in strain.
I noted that a day or two before the 6.0 Parkfield quake there was a decrease on the Parkfield tensor strain readings. But the decrease was not more sizeable
than decreases that had occured prior to earthquakes in the 4.0 range in the Parkfield area.
Maybe, the tensor strain changes noted on the Parkfield meters now could be related to redistribution of strain from the Parkfield quake? Or maybe the tensor strain meters are not working properly?