JFC, no wonder they thought they could hit it at $299; those specs sound like a disaster. 4 GB DDR4!? Mostly ARM cores!? (keep in mind, this is early 2010s; ARM cores weren't nearly as good or probable as PowerPC/x86/x86-64 replacements back then as they are today). Not sure about the GPU, can't even tell if they were planning to use AMD or not from the graphic I'm looking at but it seems they'd still be aiming for ~ 1 TF of compute performance which is about in line with what the XBO provided anyway.
Those specs give me shades of SEGA's GigaDrive from back in the early '90s; on paper, interesting for what they could've provided and it seems they wanted to make the Kinect the added value proposition but, like SEGA and the original GigaDrive, perhaps too insular and narrow in reading the market for what it'd eventually want, particularly among hardcore and core enthusiasts of the time. The 720 seems a lot like something that tried straddling the line of the 360 and Wii, but doing the bare minimum to increase over 360 while mainly going hard into Wii's motion controls demo with Kinect V2 (the Kinect that should've launched for the 360 in 2010).
In that context it's similar to GigaDrive's half-in/half-out approach to mainly push 2D sprites forward that MegaDrive has set (mainly attempting to match or surpass other 2D systems at the time like SNES, Neo-Geo and their own System 32 arcade hardware) but being a lot more modest with 3D graphical abilities. Thing is, the hardcore and core enthusiasts (particularly in the West) really wanted 3D games at home with 5th-gen, so SEGA misread the customer market direction with the earlier GigaDrive design, perhaps because they were too embroiled in the 2D console war with Nintendo and also to not cut off the market for their own 3D arcade hardware too soon. Microsoft made a similar mistake with the 720 here, focusing on motion controls when the customer market for console wanted more performant 3D, core-orientated games, but Microsoft were too caught up in the success of Kinect (and previous 360 success prior giving them some goodwill from most hardcore & core gamers even in the Kinect years) to notice.
Ironically they also suffered from SEGA's problem of corporate division conflicts, but for the opposite reason: SEGA didn't want the home console to diminish the value proposition of their 3D arcade machines. Microsoft's other divisions wanted to use Xbox to grow themselves and fully align Xbox with the larger MS corporate strategy even if that wasn't the best move for Xbox (incidentally, Sony had a somewhat similar problem with the PS3).
One good thing about 720's specs tho is that it had HW-based BC with 360; the system would've still been curbstomped by PS4 in tech specs for then-current gen games (even if PS4 only went with 4 GB RAM, because it'd of still been GDDR5-based thus much more bandwidth than 720 regardless), but ironically it could've probably had a much better early reception than the actual XBO received thanks to 360 BC out of the box. If that came with performance enhancements...MS could've probably actually kept toe-to-toe with Sony in US & UK, maybe even lead them slightly, for the first couple of years considering 360 & PS3 were still getting cross-gen releases up to at least 2015 (later in the case of specific franchises).
Though eventually 720's technical weaknesses compared to PS4 would've given Sony a huge boost in sales after that initial point, probably an even bigger boost than they actually saw vs. XBO, it's possible the 720 might've found its own niche and held over more Wii gamers as a result, making things a lot harder for Nintendo and possibly slowing down some of the Switch's initial sales momentum. Meanwhile, given the COVID and lockdown stuff that'd of happened anyway, Sony would've potentially still cut PS4 mass production short, or delayed the PS5 by a year or two, and we might've gotten a PS5 a decent bit different from what it is today (for example, potentially a good deal weaker and less emphasis on fast SSD I/O), and at a cheaper launch price too.
Meanwhile MS wouldn't have likely done the 2-SKU Series S/X approach; they'd of likely released an updated 720 in 2017 equivalent to the One X (but closer to 4 TF than 6 TF), and released 720's successor in 2021 probably pushing some form of HoloLens with it and probably going for 6-8 TF. They'd of prob also skipped the 2-SKU Series S/X strategy altogether. I'm also assuming in this timeline, Sony wouldn't have made a PS4 Pro, and saved those upgrades for a 2021-2022 PS5 probably closer to the initial leak in power of 8 TF compute performance, reduced SSD I/O features, probably about as fast storage as actual PS5 and having more & faster RAM than 720's successor.
Who knows, maybe Sony tries accelerating price-reduction of production costs on PSVR in this timeline so they can have a cheap standard VR by default with PS5 instead of doing that a generation later (something I hope they're going to do with PS6).
Anyway it's fun to speculate about but also interesting to consider how releasing the 720 as leaked here could've influenced MS to go a different direction even earlier than they did (probably also postponing or just not doing their Day 1 PC porting strat altogether, forgoing Game Pass etc.), influenced some of Sony's moves (no PSVR2, prob still a PSVR tho and maybe one with a version that can run without a PS4 at a higher price, influencing some of PS5's design considerations), and made things harder for a Nintendo comeback (grabbing more of the Wii audience with the 720, retaining more users on Xbox so they don't jump ship to Nintendo when Switch releases, potentially influencing the Switch to be less of a hybrid portable and more a continuation of the Wii U but where the tablet functions on its own as a weak portable to continue the DS/3DS/GB line of handhelds etc.).