How do I measure better? Primarily by the number of people on the forums (not wholly scientific).
Heh. I'm not certain but I'm pretty sure Stencyl's "who's online" list is set to show people in the past hour, or past day, or some long period of time, presumably to make their forum look busier. Our forum is still set to the standard default for forums which is 5 or 10 minutes (@Tom will know). If we increased our timeout we could make the forum look like it had a lot more people on it, but we prefer to make it useful (if you see someone in the online list, they're probably online right now). Anyway, you certainly can't directly compare the two without taking in to account the time difference.
Also, last I checked Stencyl it was lacking some really basic features that nearly every C2 game uses: the layering system is something like 2 layers only, they're only now working on an equivalent for Construct 2's triggered conditions for Stencyl 2.0, etc. I think Construct 2 has way more features, not even including third party plugins and behaviors (which we have over 50 of now). And Construct 2 is cheaper to buy, especially in the long run (no subscription).
As for "Flash is better understood than HTML5" or "ActionScript 3 is better than Javascript", why do you need to care? The users of the tools don't need to worry about the underlying tech, just how well it works. And I understand HTML5 way better than Flash, so no need to worry there :) Construct 2 was faster than Stencyl last we measured as well, so I'm not sure there's any credibility left in the "ActionScript is better" argument.
Stuff like "better UI" or "friendlier sounding" or "less intimidating" is opinion. In my opinion Construct 2 has a better UI, is friendlier and less intimidating (obviously). A lot of our users have similar thoughts. There's not much to argue there since basically different tools are just different, and you can choose whichever suits your taste best.
I know the mobile wrappers kind of suck, but we're being very forwards looking with them. In future, mobile wrappers won't suck, they'll be at least as fast as Flash ever was. We could write iOS and Android native ports, but that's 3x as much work (the HTML5 runtime was started a year ago and is still being tweaked), then we're a tiny team weighed down with 3 codebases which also have to be kept compatible, then there's a huge headache with "will your game work *identically* on all platforms?" (small frustrating differences are likely) and "will all third-party plugins be available on all platforms?" (highly unlikely). And by the time we're finished HTML5 might have become fast anyway. So we're sticking 100% HTML5. We're early to the party, but things will get going soon.
FYI we are currently prioritising mobile support. We're going to be adding directCanvas in future so appMobi gets hardware accelerated rendering, and there's more to come after that.