Roger-I looked at your files, and day zero averages out to being 12 hours, so you have to double the results from day zero, or use a combined number, day, hour, minute. The reason it is 12 hours is that your program only looks forwards-so the average time of a quake is 12 hours after the day starts. For example, your third earthquake is at 11:14...it only looks at 12 hours 46 minutes for a distance match. I could be wrong, but what I say makes sense. If you end up thinking that there is a significant increase in earthquakes in day zero and day 1, then you may be interested enough to break it down into 6 hour bins.....
Chris