Does 2,416 sound right? The NES is little-endian, so larger values are stored with the byte order reversed from our interpretation.
2,416 sounds about right.
I've spent this day trying to create the Colosseum battle event and here is what I found out:
For the events in the first event bank, there is a set of pointers. These pointers are what tells the game which events occur on which map. They go in order, with the building insides at the end. Even though there isn't an event on every map, there is a pointer for every map. I'll call them Pointer Set A
Now at the end of that event bank, there are a set of pointers that point to the actual event in the event bank. I'll call them Pointer Set B
Pointer Set A, points to Pointer Set B, which then points to the actual event. That's how the game knows which events occur on which map.
And for maps that don't have an event, Pointer Set A points them to somewhere at the beginning of the bank. And many of them leads to the number 60. and others lead to a group of bytes that look something like this "A9 XX 20 7D 9F". "XX seems to be the only byte that changes. Why, I don't know.
Then when I tried to set the pointer for the battle square to lead to an empty space in the ROM, the ROM interpreted that as code, not data.
So I tried changing it so that it would lead to one of the pointers in Pointer Set B, but it still interpreted that as code and not data.
So it won't work, something is telling those pointers to behave differently than the other pointers in Pointer Set A. And I just can't figure it out.
So for now, the Battle Square mini game will not be doable.
So, I'm going to go back to working on the rest of the maps for the Gold Saucer.