Right, and "successful" is not worst case. Your logic solid until you base worse case numbers for sales in a projected number of stores (4.5K stores) based upon a non-worst-case success. Additionally, there are non-U.S. 7-Elevens for further distribution, which ups the store number by some further number we don't know.
You're splitting hairs with your general assumptions, and then suggesting the numbers given by the comapny as "worst case" should be used to compute an estimate of a "successful" launch, which if successful would not be worst case scenario..