Before you give Shan too much credit, you may want to take a map and, assuming we are limiting his predictions to what has been posted here, plot all the predictions he has made for south Asia with the predicted magnitudes and dates labelled...for the last year, or 3 years, or whatever. Realize that there are known faults and a (complex) plate boundary so that distribution of large quakes is not random. Given that his closest prediction that you cite is well over 1000 km off...and the length of a M7.6 thrust rupture may be only 100 km +/-, I'm not impressed yet. Doing it this way would be a nice visual way to evaluate what to make of Shan's predictions (I don't think the method is credible so the only way for someone like me to take this seriously is for the predictions themselves to clearly be much better than informed chance).
Chris