First ... this is an excellent find, you have properly diagnosed a switch matrix grid problem and isolated the issue to a single diode, you just don't know it yet. As I explained in a previous post the triggering of a subway switch is supposed to kick a ball back into play. However, you are seeing this false triggering when you press the left target buttons .... why?
To explain the why, you have to understand the theory of the switch matrix, the switch matrix works because only one column can be active at any given time, the CPU drives a single column low then looks for these low pulses on all eight rows; any closed switch pulls the entire row low and any row that reports low means the corresponding switch for that row and column is closed.
Each switch has a blocking diode to prevent its activation from pulling the rows low when its own column is inactive, if this diode is shorted or installed backwards, the switch will operate normally; however, given the right condition this bad diode will allow the switch its installed in to pull the row low, causing whatever switch is in the same row, and on the active column to be read as closed.
Your video is pretty consistently pointing to bad drop target diode. I speculate that if you were to remove all the balls, press and hold your BUY IN button, then press your Left Top Bank target, you will see the Right Bottom lane switch activate. If this doesn't happen, my theory is incorrect.
A second way to verify this would be to go into coil test and raise your drop target switch, with this switch open, you should no longer see any of the left bank switches activate the Under Borg Hole switch.