Long-range forecasting, as done by the Climate Prediction Center of the National Weather Service, is a probability game.

Here's what I mean: let's say we're forecasting temperature. Assume there are three possible outcomes: above-average, near-average, and below-average, and that each outcome has an equal chance of occurring, about 33.3%. In the Weather Service's long-range outlooks, these percentages are adjusted up or down based on forecasters' level of confidence. And the adjustments usually aren't very large.

As an example, let's consider this summer's temperature forecast for our area: here are the probabilities for each category:

Warmer-than-average: 40%
Near-average: 33%
Colder-than-average: 27%

Clearly the Weather Service is leaning toward a warmer-than-average summer, but the confidence isn't that high. This forecast still has a 1 in 3 chance that temperatures will turn out about average, and a better than 1 in 4 chance of colder-than-average.

Tomorrow night we'll check the precipitation forecast for the summer.