I found
this quick tech write up about how it would need to be worked around which is even more complicated than I thought it would be due to hardware design. The real SNES allowed the games to directly speak to the device in the controller port. If the game found a response from the device that let it know it was a multitap then it would use a loop to read the button presses on each of the controllers connected through it.
Side note: On games that support less than 5 players (all but an NHL game and homebrew) controllers 3 and 5 act like a single controller. On games that support less than 4 players (Secret of Mana and homebrew) controllers 2 and 4 act like a single controller. This is because controllers 2 and 4 are multiplexed and 3 and 5 are multiplexed. I have to take back my statement that all the controllers connected to a multitap behaving as a single controller on 1-2 player games since I thought the circuitry was a bit different. I have worked on a project with the 3DO where it multiplexes all the controllers on a single data line - allowing up to 255 (if I remember right) controllers to be used. Does anyone have a multitap and original SNES hardware to test what happens with extra controllers when you have a 1-2 player game?