AF or SS problem

I met Mike Ramy at the Conowingo Dam, where there's a fall concentration of Bald Eagles and got one decent sequence.  Except that almost all of the shots are a bit soft.  This one is an example where I don't see anything perfectly sharp except possibly the wing tips.  I did get one or two that are nicely sharp, so it seems like it's something that I'm doing.  I was using autofocus case 6, all 65 points, and the full AF area.  Is the problem here the ss or a problem with how I've set up the autofocus  (Canon 7D2, 400mm? f5.6, 1/1600s, f/8, iso 400, tripod with loosened ball head, 77% ff)

Mark:  This looks like a focus issue, not a shutter speed issue.  For almost all flight photography that I do, I use AF case 2.  That probably isn't the biggest issue here.  The bigger issue is using all 65 AF points.  That is allowing the camera too much freedom to figure out what to focus on.  I'd suggest using the center AF point with 4 surrounding points enabled.  At 77% of the full frame you could leave the AF point dead center in the view finder and crop for composition later.  The center AF points are generally more sensitive than others anyway.  The rest of the settings sound reasonable.  

When we teach flight, one of the first things we do is get down to 1, or 1 + surrounding points for AF.
 

I concur with Keith.  My first thought was that the problem could be tied to the 65 points all being "active."  I'd also default to center point plus expansion.

Max

I also agree, Mark.  The only time I'll use more than that is when I know I'll have nothing but a sky background, and I still won't use all 65.  When I set up a new camera and select which AF modes to cycle through in the selector, I don't even include the maximum number of points as an option since I've never used it.
 

Thanks for the good advice.  I got to do some testing today, with small birds coming to my feeder and it's clear that a smaller "area of interest" works much better.  I also got to see the sensitivity with a single sensor as picking one in the outer array couldn't find the bird in the thicket, when a single sensor in the central array got it every time.