Considering the current SST situation, I expect there to be an unusually long lasting polynya in the ice in the ESS until a record late date, perhaps mid-November or so.
Not so much polynya as just simply the seas not refreezing.
With so much insolation taken up this summer, combined with influxes of advected heat through intrusion of warmer southerly water on both sides of the Arctic, along with continued massive plumes of sensible and latent atmospheric heat (warm air, high moisture content), I'm expecting a refreeze along the lines of 2016/17, and a possible new low maximum.
Over the coming weeks, Siberia and eastern Europe will chill dramatically while the Arctic seas will remain quite warm. There will be coastal freezing, but I expect the present heat and incoming storms through the fall will keep the ice at bay for a very long time, particularly in the Barents and Laptev.
On the Pacific side, I'm anticipating a late refreeze of the Chukchi, Okhotsk and Bering for much the same reasons, continuing and building on the trend we've seen the last few years in those regions.
I suspect the Beaufort, coastal Laptev and much of the ESS will freeze up fairly fast, again starting at existing ice edges or the coast, and leaving a warm gap between for some time.
The Kara I think is a wild card. While it is shallow and close to the "cold continent", I think we're going to see a lot of flow into it from continuing tropical storms that will then sweep around counter-clockwise from it into the Arctic basin proper. That may keep it open and the Atlantic side warmer for longer and generally reduce the desperately needed FDD's required for a strong refreeze and good volume production.
Not willing to speculate quantitatively on "Max" numbers yet even remotely, but both interested and concerned about what lies ahead.