Becuase thats what it takes to get a COD game out everywhere and support Warzone so it keeps printing money. CoD makes them $3.5 billion a year. If anything, they will need more studios to pitch in when they go next gen only with Warzone 2.
No one is going to be the one to suggest cod take an yearly break and they lose $3.5 billion in revenue. Half of it pure profit.
They meant the standalone COD games taking a yearly break or 2-year break between releases. No one said anything about Warzone not staying annual, it wouldn't make any sense to do so either for the reasons you state (guaranteed revenue now being put on ice due to stopping regular yearly updates).
That said, they don't need all of those teams on Warzone; that's just what Activision wanted to do in order to pump out even more revenue from the IP, and they had that specific need due to their status as a 3P publisher up until the acquisition. Now as part of Microsoft Gaming, they won't have the same need, so they can ease up on assigning so many teams to a single franchise.
They'll make a lot more money through diversifying their content output across multiple IP because a company like Microsoft in particular, has multiple big franchises that earn them a lot in revenue annually. Activision-Blizzard only "really" has COD (even if individually, it generates more revenue than any one single other Microsoft franchise (or any other franchise in gaming, for that matter)).
IP are much more valuable than people that. That point is clear. Yes, they need the talent to work on it, but developing new IP is incredibly risky. Microsoft has a huge stable of successful IP that makes new games a safer bet. Bungie has...Destiny. And that is a successful game, but when they tire of Destiny there is a ton of risk involved in a new project with a new IP.
Yep; underestimating the brand power of IPs is foolish. There's a reason companies like Nintendo have persevered and are so well-known across the world, it's due to the strength of their IPs, which they've built up over many years with regular iterations and (usually) keeping the quality at a steady output level.
Talent means a lot but definitely not everything; you can have all the talent in the world but if the IP is just not very appealing, and there's no massive marketing push, then the game will fail commercially. Even with all that going, bad timing can also kill it, as evidenced by games like TitanFall which by all other measures should've been massive hits. Generally speaking, if you've got a well-established brand and the marketing muscle, having "good enough" talent will ensure a generally expected success, and timing is less of an issue.
OTOH if the IP is unknown and marketing muscle is weak, you'll need top-of-the-line talent just to hope the game doesn't get lost in the shuffle. And if you release it at the same time as a solid entry in a well-known popular IP? Forget it, that new IP might as well be dead.