Meteorologists are drowning in data. We tap scores of weather models from around the world, looking for consistency and agreement before we issue a forecast. When models agree, our confidence levels go up.

It doesn't make me happy to admit this, but we have a problem. NOAA has taken an entrepreneurial approach, with scores of models. The European Union has one model, the ECMWF. Once again, it outperformed NOAA's models with Joaquin, predicting the hurricane would pass harmlessly out to sea six days ago. Much like with Sandy in 2012, the "Euro" was superior to models from the U.S., Britain and Canada.

It's not just raw computational horsepower; it's the physics within the model — and how new data are "initialized" that's different — and superior. We have some catching up to do.

Fine weather lingers into much of next week as winds aloft blow from the Pacific, a zonal, west-to-east pattern favoring milder- and drier-than-average weather. The next chance of showers? Thursday. I see 60s, even a few 70s, into the third week of October. That's from the ECMWF model. Sorry.