I think the Near Beer Game models your logically reduced version of the game, ignoring the original game’s irrelevant information and non-positive-value options for the intermediate players.
In this version, the manufacturer sees all the inventories, and all the middle layers pass all stock to the next layer. (The game also has a trivial demand function, so the only challenge is to detect or predict the single step change in demand rate, and then calculate 3 weeks ahead to smooth out the supply chain.)
https://forio.com/app/showcase/near-beer-game/
The game was played for 35 years before you demonstrated that it was a broken over-complication of a trivial game?
Or did you break the game by coordinating with your teammates on strategy (or, equivalently, all players computing the perfect Hofstadteran superrational strategy https://en.m.wikipedia.org/wiki/Superrationality ), when the game was meant to simulate the general human tendency for hyperlocal optimization, and the problem of dealing with chaotic incompetent peers?
I haven't played the new version, but there's a good chance they cut off the information you need (likely because of us, honestly) because each person needs to see the amount of inventory going into the previous step in the supply chain. That way they can keep inventories perfectly drained except at the retailer.
If you use knowledge of the deck, you can obviously pre-solve things, but that was not an assumption here - our thing works without knowledge of the order deck.
The "new beer game" looks totally different, honestly.