Thanks Gavin
Looking at some historic analogues these tie in quite well to your Autumn forecast. The only exception is temperatures for November which the analogues I have picked out suggest could be much warmer than average.
The 5 years I have selected were all running into an El Nino with an ENSO index of >+1, had a positive QBO phase and similar sun spot activity to what we have at the moment.
This shows that September and October temperatures were very close to average but November was warm with 1987 being an exception.
Rainfall was either close to or well below average in September. In October there is a bit of divergence with a couple of years continuing the fairly dry theme and others going very wet. Overall slightly wetter than average. What seems to have happened is that the point at which the pattern flips from dry to wet varied slightly from year to year. For example in 2002 October was very wet but the first 11 days were very dry continuing the theme from September. So the change to wetter weather might happen in early October or it may not happen until much later in the month.
November was very wet in all years apart from 1987 which clearly had a very different pattern with cooler conditions. The data do suggest a high probability of a mild, wet and stormy November.
Another year that I am tracking closely is 2009. The QBO was negative so it does not appear in the list above but many other factors including ENSO and CET temperature profiles track 2015 very closely. That year also saw a dry and warmish September and a very warm and very wet November.
So there does seem to be a reasonably clear signal for this Autumn. Most likely a fairly dry first half of Autumn with the second half potentially very wet indeed. Temperatures potentially very close to average for the first 2 months but likely to be well above average in November.