The Week Ahead, And Why I Am Probably Wrong About Thursday
Seven days from the Sky Desk, with the confidence I actually have rather than the confidence a forecast box implies.
Background written by the news staff. Contains no argument or recommendation.
ITHACA, N.Y. — A ridge builds over the eastern Great Lakes on Monday and holds until it doesn't. Here is the week, with the confidence I actually have.
| Day | High | Low | Conditions | My confidence |
|---|---|---|---|---|
| Monday | 81°F | 62°F | Clear, light SW | High |
| Tuesday | 84°F | 65°F | Clear, hazy by afternoon | High |
| Wednesday | 86°F | 68°F | Building cumulus, dry | Moderate |
| Thursday | 79°F | 66°F | Thunderstorms, timing uncertain | Low — see below |
| Friday | 77°F | 61°F | Clearing, breezy NW | Moderate |
| Saturday | 75°F | 58°F | Clear | Moderate |
| Sunday | 78°F | 59°F | Clear | Low, it is seven days out |
Thursday
Thursday is where I will be wrong and I would rather say so on Sunday than apologise on Friday.
The ensemble is split. Roughly two-thirds of members break the ridge on Thursday afternoon with a line of storms coming through between 15:00 and 19:00. The remaining third hold the ridge and push the whole thing to Friday morning.
That is not a small disagreement. It is the difference between a washed-out afternoon and a pleasant one, and there is no honest way for me to resolve it four days out.
So the box says Thursday, and this paragraph says: it might be Friday, it is about two to one, and if you are planning something outdoors on Thursday afternoon, plan the alternative on Wednesday night when I will know.
Contrail persistence
Low all week, and for a reason readers of this desk's analysis can now predict as well as I can.
The ridge brings subsidence. Subsiding air warms and dries. Dry air at flight level does not hold ice, and a trail in dry air is gone in seconds.
Expect short trails Monday through Wednesday, a spike on Thursday as the front's moisture arrives at altitude ahead of the surface storms, and short trails again from Friday.
If you want to test the desk on something, test it on that. It is the most specific prediction on this page and it is falsifiable by looking up.
Last week's score
Seven forecasts, four inside the stated bounds, three outside. The three misses were all overnight lows and all in the same direction — I was too cold by an average of 3.4°F, which is a bias rather than noise and which I am now correcting for.
Running total since September 2006: 5,221 forecasts, 3,391 wrong. That is 64.9 per cent, down a tenth of a point on last week.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
A forecast, scored the following day against the observed conditions at a fixed station, using a scoring rule unchanged since 2011. Each day below carries the desk's honest confidence. The running error rate is published every Friday.
Who we spoke to
- National ensemble guidance, Public model output. Read daily August 2026
- The desk's own rooftop observations, Twice daily since 2015. Direct observation 8 August 2026
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.