Lies. Show me one example where VGChartz has proven to be off by more than 15% total units.
While that's not dead point accurate, the fact that it is the only public metric out there makes it relevant. Being off by 500K units when total sales are over 5M is sample variance, not inaccuracy.
You have to consider the changing industry and the way DA:I sold.VGC counts only physical units but in the case of DA:I, most sales happened online.Why?Because of the following reasons.
1.)Overall, as time moves on more and more people buy games online, and less in stores.The people who buy games in stores, are generally(not all) those who don't spend that much time online and base their buying decisions on the basis of popular title names such as CoD or AC which they hear of via local word of mouth.
2.)Dragon Age Inquisition is highly advertised on Origin and other online platforms and is flooding people with advertisments that say "Buy DA:I on Origin, it is GOTY".People who look at these advertisments and decide to buy the game, generally buy the game on the platform it is advertised on.
3.)Digital download bonuses and online preorders.
4.)Online features such as the keep, encourage the "do and buy **** on the internet" mentallity.It might sound far fetched, but it is proven that games which incorporate online features get substantialy more digital sales .
Overwall, it is pretty safe to asume that DA:I has a HUGE sale number that is ignored by VGCcharts due to many reasons.It is safe to assume that with the digital sales DA:I outsold it's predecessors by a mile!
Keep on rockin' GOTY!
EDIT:
Also to your arguments that games usually sell less after the first few months.That is true most of the time.But NOT with GOTYs.Statistically, GOTYs get TONS of sales long after the game releases since people want to experience these "games that define a year" that are presumably, extremely good.