The probability map for Nov/Dec/Jan predicting above average temperatures - the Met Office have insisted this a) isn't a forecast and b) requires more knowledge to interpret than what is displayed on it. It is clearly a forecast but what additional knowledge is required? ie Is that map the end product for temperature or an intermediary step? - mixing in projections for other variables such as wind, precipitation and whatever else reducing the temperature projection to the one now being published here.
To illustrate the problem look at the winter of 08/09. Going by Met Office press releases they told us in September 08 winter would be mild (Trend for mild winter continues). In November 08 they were still predicting about or above average temperatures. The Met were at least honest about the outcome - they got it wrong.
For 09/10 they again forecast an average or above average winter and they were wrong again.
I can sort of see a problem and it is one not of science but psychology. The chosen basis period for these forecasts tends to be 1971-2000. As I understand it that is due to it being when various high quality (especially satellite) data started being collected. But that period isn't necessarily indicative of the accumulated weather records going back say, 100 years. That period in time is when a good deal of the measured 20th century warming happened. Are they putting unreasonable faith in that basis period - not in terms of accuracy but how it has been interpreted to suggest that warming will continue?
To my untrained mind it looks like the Met Office and their computers are stuck on a trend line that begins with the warming from 1970 or thereabouts and simply continues into the future. Is their default setting average or warmer? Being at the tail end of a warming period you would stand a good chance of being correct whether your method was right or not. Then if the rate of warming slows (for whatever reason) your projections would then begin to overshoot reality.
However, all that is like arguing over how many angels can fit on the head of a pin. What keeps happening is that climate forecasts are being used to shape policy decisions despite the plain evidence all around us again and again that what matters is weather. Even if winters become mild over a long enough period we will still run the risk of cold snaps and widespread snow and should be reasonably prepared for it.