Roger-looking at your revised numbers, there are about half the quakes in day zero as in day 2. That makes it look like the quakes average out occurring at non (half before, half after), so you average only getting only 12 hours worth of quakes.

If this is the case, day 0 and day 1 have more quakes than any other day, which means that there could be a highly statistically significant relation, even if the actual increase is very small. If this is the case, you may want to look out by hours-maybe 6 hour increments: combine day, hour, minute into one number and do it that way....if too much computer crunching, you could look out 10 days rather than 30.
Of course, only do this if you enjoy it/are interested.
Chris