We know that the climate is changing. It has been changing every since the earth began. The question is whether we can predict the change from our models. Whether our models add any value to the prediction. Let’s suppose we take the default hypothesis that next year’s weather will be the same as this year’s. The AGW model should be able to generate a prediction based on its own calculated simulation. Then let’s compare the AGW models to this simple-minded “no change” prediction. When the next year comes we can compare the difference between the “no change” forecast and the AGW forecast against the actually observed values. Which prediction is better? If the AGW model can do no better then it doesn’t have any obvious value. If it is consistently better than the simple model or any other model then it has some predictive value. (Hat tip: ED for this approach to thinking about it).