The metoffice base their forecast among other things on the atlantic SSTs, it gives an idea of what the NAO is likely to be come the winter. Ofc if the NAO is negative then cold and snow is almost axiomatic (if the AO follows suit).
Current SSTs
http://www.ospo.noaa.gov/data/sst/anomaly/2014/anomnight.8.11.2014.gif
This time last year
http://www.ospo.noaa.gov/data/sst/anomaly/2013/anomnight.9.12.2013.gif
What we are looking for is a tripole pattern, i.e Warm in the south, Cold in the mids, Warm in the north. Last year there was no real sign of this pattern at all. This year there is a tiny patch of cold in the mids, and there is warmth in the north. But no warmth in the south. From this alone, an average or mild winter is implied.
Originally Posted by: Quantum