Accuweather's 15-day Forecast A Waste of Time

Awhile back we did an experiment with those long-term weather forecasts--up to 10 days out--concluding that they weren't very useful at all (indeed, worse in some instances than just flipping a coin or using the average weather for the day as a guide). (See "Long Range Weather Forecast II--Still Gloomy")

Recently we thought we'd briefly revisit the issue, just for fun. Accuweather.com has a feature that purports to give you a 15-day weather forecast. Now wouldn't it be great if we could get an accurate forecast of a particular day's temperatures and precipitation chances two weeks in advance?

Just think of the opportunities for planning: should I schedule a golf outing (substitute any outdoor activity for golf if you want) two weekends from now? Or, what's the weather going to be like on my trip to _____ in two weeks?

Well, we thought we'd give it a try. This weekend we were the host team for a soccer tournament in Arlington, misnomered as the "Spring" invitational (by my calendar, we have three more weeks of winter, notwithstanding those sunny weatherpersons who are proclaiming it "meteorological Spring.")

Knowing the weather in advance for the tournament would be a big help in planning things such as our concessions: early March being a particularly dicey time, weatherwise--it could easily snow (in which case we'd need lots of hot drinks) or just as easily be in the 70's (in which case cold sodas would sell well). So we consulted with Accuweather beginning two weeks out to see what we should plan for.

[By the way, if you're interested in both advances and limitation of long-term weather modelling --not to be confused with climate modelling--check out this post from our friends at the Capital Weather Gang, who say we're not ever likely to be able to predict the weather more than two weeks out.]

Ideally, the forecast at 15 days would stay fairly consistent as the target day draws nearer--it doesn't help if the 15-day forecast turns out to be right, but in between the forecast varied all over the place, causing one to change their plans. (It would also help if Accuweather--and other long-range forecasters--would post the average highs and lows for a particular day so readers could guage the forecast relative to the average.)

So how'd it go with our soccer tournament? Not too well, we're afraid.

On February 17, fifteen days before our weekend event, Accuweather said we were going to get incredibly lucky: 60 degrees on Saturday and a balmy 62 degrees on Sunday, albeit with a chance of rain on Sunday. That was okay--since all our games were to be played on artificial turf fields, we weren't too concerned about a little rain. The relatively warm temperatures would be perfect for Spring soccer--maybe out tournament wasn't misnomered after all!

But over the next three days, that optimistic forecast steadily declined. By Feb. 20, we were facing a gloomy prospect: a chilly high of 39 degrees on Saturday, with snow/rain, and a rainy high of 46 degrees on Sunday. Ugh. Maybe we should sell ski gear?

Fortunately, the forecast continued to change. The only real consistency was that it was usually forecast to be warmer on Sunday than on Saturday. By Feb. 25--less than a week before the tournament, we were looking at a high of 49 degrees both days, with sun on Saturday and clouds on Sunday.

Then, on Feb. 29--now THE day before the tourney--we were told it would be a chilly 44 degrees with partly sunny skies on Saturday, then turning mostly sunny, with a warmer 52 degree high on Sunday.

What was the actual weather? Turned out it was 52 degrees on Saturday (yesterday), but very windy and mostly cloudy--not particularly pleasant, but certainly playable. On Sunday (today), it got up to 53 degrees under bright sunshine and a light breeze, feeling much more comfortable.

And what would the average high temperature for this weekend have been? It would have been 51 degrees--pretty much what we got.

So, was the Accuweather long-term forecast helpful? NOT AT ALL. Instead, it took us on a roller coaster of emotions, starting with a terrific forecast for the weekend, then deteriorating into an awful one, then rebounding to something more or less average. Just one day before the first games the forecast was off by 8 degrees on the low side.

In the end, we would've been better off simply looking up the averages (at least for temperatures) for our weekend and planning based on them.

Oh, if you're part of the second half of our tournament (for girls teams) NEXT weekend, take heart. Accuweather says it will be unseasonably chilly next Saturday, with a high of only 45 degrees, and windy/cloudy conditions--really not that much fun. But since that's the forecast now, it probably WON'T happen!