Well, we're back in the weekly statistics business again. What looked like a minor issue instead meant recoding several modules in the system. Silly me for not unit testing every module before releasing THA to the unsuspecting public.
It turns out that the daily Top Horse picks combined with the Best* flags were correct. That's the good news. The bad news is that the weekly statistics were wrong for all of the TH/Best* conditions, not just the TH/BestS condition that was originally reported by markcol. All of the remaining Best* statistics checked out OK.
Please speak up if you spot any more anomalies. This system is huge with thousands of lines of code and lots of opportunities for coding errors. The more accurate the statistics are, the more useful they can be to you (and me). Thanks!
Note: The average field size for OP is zero but will be fixed by next week's statistics.
|