You can't really talk about "downgrades" when deep cold was only ever shown on the shores of la-la land at 240hrs+. It's going to cool down significantly from the beginning of next week. After next weekend, nobody really knows. The Met think it will continue cool, the Beeb think it will warm up. So expect models to waver between both options for a while yet.

ECM 12z depicting a 3 day cool snap with less chilly air edging in at 144 as then low anchors off Biscay/ SW approaches~:

Thought I'd stick the ens up - Op a warm outlier, with a majority going with a continuation of the chilly to cold weather. Based purely on the GFS ens set the most likely outcome is for colder weather to win, in a mostly blocked scenario. There's a lower chance of less cold weather as shown by the scatter from circa 24-26th November and thereafter and a small chance of it turning milder than average. Down south at least there's not a strong signal for rain although it won't be universally dry.

If the stragglers start turning into a solid cluster then confidence in a milder solution will grow, but at the moment it's not the favoured outcome. 

Whether the ECM Op picks a colder or milder option in its FI run later on is not massively relevant - whether the Ens support the Op is more relevant and if the ens are split, what the clusters are like.

I find life a little more balanced if I look at the ens before the Op - less highs/lows that way

Should just mention I'm far from convinced we're going to get a decent cold spell out of this set up (I/e one with near ice days and snow) but it remains possible whilst the models support a set up that can bring decent cold to our shores

good, I could do without cranking the heating up for too long.

broad scale signals suggest otherwise however:

http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/ao.sprd2.gif

AO an important factor as well as NAO when it comes to the bigger picture & what might affect us longer term. 

ECM looks very good at 168h for a NE attack lets see