After a very early minimum (7 September) Arctic sea ice started very strongly to get itself back above the 2 x standard deviation [2xSD] region (the very light grey area in the graph) during the remainder of September. That acceleration ran out of steam (pardon the pun) and the sea ice extent has now fallen back well below the 2xSD and is currently the lowest it's ever been for the 19th of October since theses records started in 1978. It's a shade lower than it was in 2007 and only 69.9 of the average for this day so less than 30% below normal.
Looking at the Arctic ocean there seems to be a dearth in sea ice along both the Russian Siberian coast in the east and the Alaskan Canadian coast in the west. It looks pretty dire for so early in the season, and must have a knock on effect in the coming the northern hemisphere winter in some way.
I would like to report that in the Antarctic things are looking up but I'm afraid not! The gains in recent years are a thing of the past , and the early maximum this year on the 28th of August is a sign that things aren't any better down under. The extent has just slipped out of the 2xSD area of the graph and currently for the 19th of October is the lowest it's ever been on that day of the year. The good news is that it's less that 5% lower than average, and not 30% lower as in the Arctic. The eagle-eyed amongst you will be wondering why 1986 isn't top of the lowest table, the reason is that reports back then were every other day and not daily.
It's hardly surprising in what will be the warmest year on record globally, so you'll have to get used to more Winter's devoid of snow as I have in Devon during recent years, running outside in a morning to grab a picture of the hoar frost before it melts.