Let me teach you to be a classical statistician. Go any climate site and download a time series picture of the satellite-derived temperature, any will do.
Now look for a ruler, perhaps the one you used in a math class. Make sure you really like this ruler.
Place the ruler on the temperature and plot it along the data where it most pleases your eye. Now draw a line on the straight edge.Now erase all the raw data.
Now if asked if it was colder or warmer and since you have erased the raw data, insist on the scientifically of that line.And if asked,say that according to its sophisticated inner-methodology, you projection for temperature is up or down based on your orientation that pleased you when you made the line.
Don’t laugh. This analogy is not far off from the truth. The only difference is that statisticians don’t use a ruler to draw their lines—some use a hockey stick (Now you can laugh). Instead, they use the mathematical equivalent of rulers.
Statisticians are taught that data isn’t data until it is modeled. Those temperatures don’t attain significance until a model can be laid over the top of them. Further, it is their dogma to dismiss the data and talk solely of the model and its properties. They love models!
Something always forgotten: for any set of data, there are always an infinite number of possible models. Which is the correct one?
There are models that will say the temperature has gone down, as others will say that it has gone up. The AP statisticians used models most familiar to them; like “moving averages of about 10 years” (moving average is the most used method of replacing actual data with a model in time series); or “trend” models, which are distinct cousins to rulers.
All of their models are models and should not be trusted until they have proven; using past data to predict present weather. None of the models in the AP study have done so.