Most of the discussion about dispersion is pointless because we don't have a direct measure of ice dispersion. The area vs extent figures include the effects of melt ponds. We know from published research that early melt ponds are important to extent drops later in the summer. Thus the conflation of the effects of dispersion and melt ponds is a major problem in our discussions of dispersion, as Neven gently noted. As someone trained in science, not diplomacy, I do not tend to be so gentle.
Available data sets, such as JAXA extent, show that this year is running neck and neck with 2012. However, the distribution of early melt ponds, open water and thick ice is quite different. The weather has been different.
One factor that is worse this year is stored heat in the Pacific water layers, in particular, the summer water layer. Large amounts of heat have built up in the Chukchi and Beaufort seas at depths of 50m-100m. Moreover, heat in the Atlantic water layer has increased over the past several years. I'm uncertain about how the Atlantic layer heat now compares to 2012, however. What is clear to me is that late summer cyclones on the Siberian or Alaskan side of the Arctic ocean could stir up this heat like the GAC did in 2012. On the other hand, cyclones over the Canadian side could bring a cool close to the melt season and no new record this year. Late summer cyclones between the CAA and the pole have been good for the ice in recent recovery years because they stir up icy cold fresh water that cools the atmosphere.