For the number of people on the forum, I guess it depends if the number counts/displays the webbots or not.
Also as you said it's not really a scientific measurment anyway, more to get an idea.
There are several reasons I guess. C2/CC are far less represented in the medias then stencyl is. I guess a reason for this is in part the statement in your point 8.
For point 2 : C2 already exports to iOS and Android (as HTML5 sites, and also has "experimental" options but still already working too). Check for the keywords "appmobi" and "phonegap" in a forum search you'll find more than one topic on the subject. Exporter for C2 is not far away either.
AS3 has probably more libraries and scripts available to help that process compared to the younger HTML5/JS.
Also consider Flash being a "dying technology". The author of stencyl knows it and knows that is best "invest" in the long term to make incomes he is to have a support that will earn him money.
Hence making mobile phones a priority. When flash will be "gone" and HTML5 being a "young technology in developpment" will have as much deployment/developmment/etc as flash has now and will be considered the "main" technology if he hadn't this "exit" he'd probably couldn't make a living anymore.
For point 3 : HTML5 is still yound indeed. In a couple of years this point won't matter though.
Another matter of "here and now" I guess.
For point 4 : It maybe sounds friendlier, but as far as I'm concerned when I tried it a few months earlier, the tutorial happened to be quite tedious, and a "developping tool" that requires you to be connected to the internet before doing anything just doesn't cut it for me.
Also communism of assets is cool up to a certain point imo.
+ the fact that I have a slow connexion, uploading assets to a server before incorporating them in my project just doesn't cut it.
On that points, it doesn't make stencyl better than C2 in any way in my eyes.
For point 5 : More or less like above. But on this I won't argue too much as I'm really biased/used to C2. it reminded me of Visual Basic so I found my marks pretty quick. For a non-coder user I understand how intimidating it can be.
On the other end, the progressive tutorials available on scirra's site and the manual makes it easier imo to "enter into C2" than Stencyl's "Follow my lead but don't go experimenting on your own" introduction.
For point 6 : Two man job, it's on the todo list, list of priority, etc...
Flash older than HTML5, more libraries, more scripts, more ppl involved for more time, etc...
For point 7 : Scirra had to find a way to monetize C2. As they are not enforcing you into being connected to their website/server they fell back on including certain limitations.
Also to publish to iOS you need to pay 149$ a year for iStencyl pro. C2 is a one-time fee.
Their "free version" is also incomplete in that sense I guess.
For point 8 : Ashley's last blog article is an interesting read.
I don't know if stencyl is doing better than C2, but it has some success indeed, and the apparently single dev can make a living out of it. Good for him.
I guess it's just that Flash is "here and now" whereas HTML5 still needs to mature. As far as C2 is concerned, it's doing the job though.
And the general spirit of the "coding community" about HTML5 needs also to switch from "bah useless/toy" to "OK, I see the potential here".
This is mostly a matter of time.
As I said, most of the points evoked here are subjective (yours as well as mine).
In the end, I believe C2 and stencyl are different products, aimed at different users possibly.
They can and do coexist though.
As scirra often states, they believe their product is better. I believe it too.
I hardly tested stencyl, true, but not coming from the flash developpment at first, nothing really pushed me into getting deep into it, whereas I'm already used to C2.
I have no reason to do the jump and I really don't know if I could achieve as much in Stencyl as I know I can in C2 and I don't know if I could have as much fun with it either.