With the multiple reversals occurring I wondered if the NWP models would be able to quickly assimilate the impacts of a SWW. They have overall been hugely volatile in the past but I thought the extra verticle levels might reduce that confusion. 2018 remains a bit of an outlier in terms of how quickly the models built the likely impacts into their outcomes and that seems to have been largely due to the rapid speed of the reversal downwards. These events seem to be showing that the models still have a lot to learn. Not surprising I guess due to the lack of historical data available to backtest the algorithms and the fact that each event is rather unique. I'm assuming that the spread of outcomes between runs, in the ensembles, and between models makes the averaging process of the anomaly charts rather meaningless right now. I guess the best way is either to hunker down until a more consistent outcome is determined or focus on the teleconnections. Allround the final location of high pressure looks very undetermined.