Yes. In fact, I would argue it should not be possible to have this "the best team" - if teams in different parts of the org are directly comparable, then they must be doing the same exact thing, and why do you have multiple teams scattered around doing identical things? That is, a well structured org will have teams working on Thing A and Thing B. If a brand new Thing C comes along, it isn't either Thing A or Thing B, and you have no real information on how good either team will be at Thing C, all you can know is how good they are at Thing A and Thing B respectively.

At one time IBM had the resources to assign the same major challenge to two internal organizations.

Fair point, and that sounds pretty awesome for getting data on things, but a luxury most don't have. I suspect nearly every organization where this is the case is doing it accidentally rather than conscientiously.