Jeff, Here is one more web page for verifying CPC's seasonal forecasts. It shows the official forecasts and observed seasonal anomalies categorized on a 2x2 grid, and gives the Heidke skill score for that forecast: http://www.cpc.ncep.noaa.gov/products/predictions/long_range/tools/briefing/seas_veri.grid.php Take a look and see what was predicted for the winter and what ended up happening. A couple things of note are that there was a fairly sizeable area of below normal over the eastern U.S. in the forecast, and the score was fairly high, about 50% for the areas where above or below was forecast. When you include the areas where no forecast was made other than climatological probabilities (which gets a zero skill scors) and average the U.S. as a whole, the score was about 30%. That is 30% of the way between getting the same score as random chance and getting the forecast perfect. It is true that the forecast had a small area of above normal over northern New England and this area turned out to be below normal for the winter (December, January and February anyway). I would certainly NOT say that above is "always" forecast over the east. If you use the web tool here, http://www.vwt.ncep.noaa.gov/ , you can select states east of the Mississippi and get the skill scores. The score was in the end small but positive in the east, +11.5 Heidke skill score. Still better than random chance. After all, as this blog has pointed out, you don't expect to always get the forecast right everywhere for seasonal climate forecasts.