Stumpokapow
listen to the mad man
Amazon is making a console?
Why? That makes no sense to me.
Ouya and all these one of Android consoles aren't really doing that well as far as I know and they already have an entry in the table space and push their own appstore in the Android market space.
They're making a console for the same reason Valve is doing Steam Machines as an initiative -- when you have an extremely lucrative software and services stack to distribute content to users, you want as many hardware options as possible to leverage that.
Ouya is a console that manufactured probably well under one million units made by a startup with extremely limited VC and $10 million from Kickstarter. Selling it relied on retail partners who had no margin and thus no vested interest to sell it. Developing for it required additional development effort for no userbase reward. Amazon is able to self-launch hardware products as they did with the Kindle and Kindle Fire, they are extremely experienced with manufacturing, and they already have the #1 electronic retailer in the world on their side (it's them).
Furthermore, I don't think Double Helix has ever made anything better than a mediocre game (as far as I know, discounting the new KI since I haven't played it and most likely never will).
You're thinking the wrong scope of games, for one, but also I think you underestimate how hard it is to build studios from scratch. It's way easier to hire ten key excellent performers from teams across the country (Amazon has been aggressively hiring for more than a year now) and attach them to a team that, while they haven't produced any particularly good games, is able to meet deadlines and get stuff released than it is to build a studio from scratch with all star performers.
As a platformer owner you also need an internal team for developer support, and that's basically a problem that just requires throwing bodies at it. NST is Nintendo's developer support team in the US; they aren't a particularly great developer, but they do handle an awful lot of developer liason stuff, so they offer value.
The reason why big companies do things, whether they end up working or not, is normally pretty simple if you just stop and think about it for a bit.