The last M6+ earthquake in parkfield was in 1966. I believe it broke to the surface (I've been there once or twice). If it released all of the accumulated strain, and the fault is slipping at, say, 33 mm/yr...hmm, lets get my trusty 30 year old radio shack calculator....that would be about a meter of accumulated strain. I believe that back when anyone cared about Parkfield there was a discussion about whether a Parkfield earthquake could also break the next segment to the south, the Cholame segment-maybe that was a published paper. I think the 2 segments together would make a M7. The establishment would consider a repeat of the M~~8 1857 earthquake highly unlikely because studies have shown that the last few quakes on the Carrizo segment have slipped 10 m,and only 4.4 m have accumulated. So, they have assigned an extremely low probability of a break in that part of the fault in the next 30 years (starting a few or 10 years ago). But, they also put a probability of 90 something percent for Parkfield. I don't think they understand this well enough to put such low (or high) probabilities. I know this stuff is important, but I spend enough time on this site so don't feel I can take the time to dig out the precise numbers: I'll try and qualify what I say. I have a vague recollection that there were some 2s or 3s a couple of weeks ago. So, does a 2.8 mean something larger is imminent? I think they have had larger earthquakes there in the last decade-I think one or 2 well over M4, without setting off a M6+. I recall the USGS even issued an alert of some kind after one of these. But, with each passing year the shear stress should be larger. I guess I'll defer to Lowell here. Oh, I guess he has to be careful what he says, also. But, he can put his finger on the size and dates of past Parkfield earthquakes (some early in the 20th century may not have been at Parkfield (not on San Andreas)). Chris