Very very interesting indeed so important that tiny piece of energy

Actually, to be blunt, you look at clusters rather than the means. If you were to teleport inside the Met Office you'd find they use a product called Decider, amongst other things, which does a grand job of producing and analysing clusters. The pros only give mean charts a glancing look, we should be doing the same.

Fixating on a mean chart which shows a continuation of the current pattern is like sticking your fingers in your ears and saying "I can't hear you!" when someone's explaining something to you. The scientific way is to look at all the output (and yes, that means being creative with free trials, or simply paying the 25 bucks a month in order to see more of ECM) and then work out what's likely to happen based on percentages.

Just looking at a mean chart and going "oh, that's going to happen because it's what's happened for the last couple of weeks" is daft IMO.

I would have much more respect for the mean-chart users if they would quote percentages, but they never do!

That's a fair point Darren.

Having got that off my chest, it is true to say that at the moment a SW'ly flow on the 1st January is the favoured option, but that's by use of data from the Netherlands - close enough down here to be used as a proxy, but not such a good match for Scotland!

http://www.weerplaza.nl/15daagseverwachting/?type=eps_pluim&r=zuidwest

There are two things I look out for on there, other than the obvious one. First is the "Dauwpunt", or dewpoint. Subzero runs on there imply either a continental feed or an especially cold maritime feed; bearing in mind the location is coastal the North Sea will have a big moderating impact on dewpoints when the wind is from the west, NW or north. Subzero dewpoints from a SW'ly aren't exactly common.

The other is the "Windrichting", or wind direction. Around 10% show an easterly flow on the 1st, 15% or so show between west and north and the remaining 75% go for a SW'ly flow.

So, based purely on this one run, for SE England at least it looks like a 75% chance of a SW'ly flow on New Year's Day.

That, in my view, is the way to analyse the ensembles.

Some way out, but you've got to hope this feature doesn't turn up:

Hopefully it won't!

We're seeing all this as the models are playing around with the warm advection to the east; it creates a marked boundary with a strong jet that's ideal for spinning up lows. The models especially love to over-egg them at range, from what I've seen, and even 2 days out you won't know the exact track or intensity of those lows.

Nonetheless it's a good heads-up on what may occur - given the circumstances I wouldn't be at all surprised to see a "bomb" or two spin up in the Atlantic, my only hope is that they'll stay well out to the west.

I'd like to see it track W-E, ideally around 500 miles to the south of where that is showing. Help smash the puke-inducing Euro High, whilst aiding the ridging of the Scandinavian High W'wards/NW'wards.

It's pep-up the Channel, too! Give a bit of excitement further south...

Yep, although it's interesting that the Scandy HP isn't swept way to the south on this run and instead maintains its existence- and to a large extent its latitude - albeit a little further east. 

What also caught my eye was that some of the jet is digging further south into Europe which may help explain the positioning of the HP.

Atlantic wins this battle, but the result is possibly a step in the right direction for those with a little patience.

Time will tell...

PS Thanks Retron for that Weerplaza link and guide on how to use it - very useful. 

Still a cameo but some interest for colder weather fans:

Looks to me we have taken a step backwards this morning. No end in sight to this awful pattern

Yes it looks likes Atlantic wins the battle However GEFS persists with HP build to east with around half ENS anticyclonic-Scandi or European HP over new year period.However most of ENS hve HP too far east/south to push back the zonal atlantic flow.

However step by step the next marker is around 28th to see if and where HP builds and  how strong it is.

At present METO is showing no sign of any HP influence in its forecast up to the New Year.

Back to the verification stats showing ECM on top at 240h... That may be the case for the past few weeks, but I think most of us could have guessed the output for that period!  What will be interesting is how the models handle the possibility of warmth heading to the pole.  I think we should look at verification again in a week or so and see who is top (and more importantly how accurate they have been).

Monster storm on the 29th on GFS 06z. Sustained winds in excess of 150kmh/90mph. Gusts completely off the scale. There's been a theme lately for some big storms around the New Year so worth keeping an eye on.

Blues to the south, reds to the north... great picture.  Though these are anomalie, i suspect this still equates to HP over europe.  But at least theres no slug.  Seems the Slug is going on its Jols to Scandinavia :)

Looks like remaining very wet with with up to 150-200mm+ forecast in the N and W between now and the New Year. More Christmases ruined by flooding looks likely..

i was thinking the same beast. Some good banter in here today but honestly, if your pinning all your hopes on the dodgy evolution of one shortwave for cold then your bound to be on the whole mostly disappointed. There can be little hope of any notable cold spell for the foreseeable future and January on the whole could well start on the same note that December has endured. And note, I use the word endure, for even though I like milder weather this persistent lack of variation and windy grey leaden skies crap is enough to make Santa swear. 

Cold has generally been in the cameo role since November. 

ECM is nearly always above UKMO and GFS at all times of the year and having been posting these daily on my website for over a year now I can declare that the times that ECM has been superseded by either of the other two models over that time can be counted on 1 hand but there again what do I know.

It looks like you're missing the point a bit when it comes to statistics. Being able to guess the outcome of something with a 45% success rate is obviosuly better than guessing at a 40% rate - but neither are anything to write home about and in fact would be pretty useless for basing an important decision on. Now if the difference was between 92 and 97% it'd be a different matter entirely...

Those stats you posted in fact show the models are all awful at day 10 - the ECM is just slightly less awful than the rest of them.

The chart also only shows the operational runs (which are known to flail around wildly each day, especially when dealing with a potential pattern change). The ensemble charts would be interesting, but I reckon they wouldn't display much greater accuracy even then.

For those who want to dig more into model verifcation stats, there's a good paper here from the Royal Met Society:

http://onlinelibrary.wiley.com/doi/10.1256/003590002321042135/pdf

Yeah, it's old, but there's a nice graph on page 652 showing how correlation has slowly improved - very slowly, it seems, since 2001!

PS Gibby - if you've been watching these things for 20 years, rather than just a year, you'll know that day 10 charts are very definitely just for fun. I've explained why above, but to give you an idea there have been multiple times in winters past where the entire ECM ensemble suite has flipped... and it invariably happens at less than a 10-day range.

No sign of any cold weather there.

[quote=Retron;745170]

Those stats you posted in fact show the models are all awful at day 10 - the ECM is just slightly less awful than the rest of them.

That was my point... unless I'm reading these incorrectly?  In any case it looks as though Black Line GFS is better over past few weeks than Red Triangle ECM?

I think when you are in rock solid W/SW flow as we are now, then verification surely goes up in the 8-10 day range compared to when the pattern is more uncertain.

In ten days we son't know if it will be heavy rain and 13C or a day between systems and a bit cooler - but it WILL be an Atlantic flow with PV to the NW and big ole Euro High.