At first blush, one might be very impressed by Mark Nolan's low average errors compared to the other Cleveland area forecasters in this experiment. However, Mark's results are skewed by the fact that no prediction for Friday, March 15 was included. For our other forecasters, this contributed quite a bit to their errors, as the high reached a very unexpected 67 degrees that day. No data was included for this day for Mark Nolan, however, because as of Monday the 11th, the wkyc.com Web site had still not been updated with Mark's Sunday forecast!

This oversight spoiled what might otherwise have been the crowning moment in Mark's career: having the undisputed lowest average errors in this experiment. Oh well. Advice: Trust Mark's forecast, but you better be watching it on TV, because you never know when the Web site will be updated.