Wow Roger. Algermissen is all over the internet with all kinds of earthquake studies. Some of the stuff I have looked at so far does make the attempts here seem feeble. Still, all of the information I have seen related to Algermissen in relation to earthquake timing seems to be based on historical seismicity. If in fact earthquakes cannot be predicted, the historical seismicity has limited benefit. Not only that but the time frames are huge in prediction terms.

There must be some way to take the location probability information and improve on the timing.

Is it impossible or extremely difficult? If you have an idea to improve on the timing and hesitate because of the size of the task, then it is not impossible.