My Telegraph column today is a rememberance of Andrew Rothovius of Milford, who was a volunteer weather watcher for the National Weather Service for 49+ years. The column talks about a year-long project he and I did in the early 1990s, comparing the three-day advance weather predictions in the Old Farmer’s Almanac against what actually happened. Being generous in counting hits, he found that it was correct almost exactly half the time.

The annoying thing is that I don’t have a copy of my column in which I presented his year-long results, so I don’t have his actual percentage, only subsequent reference to it being just a tiny bit less than 50 percent. The column ran before we were digital - Google can’t help me, egad! I’d have to hunt through the microfiche (a headache-producing task) or rummage through a trunk in my attic where I keep old clips to find it.  Maybe one of these days.