Romancandle: If you feel like scrolling through the entire rom, look for a pointer table that is only 25 or 26 pointers long. Don't know if it will have a nature at 0 so that would make the table one pointer longer. The table you encountered might be related if it is indeed the proper length. You would just follow the pointer, though I fear it is much more complicated and probobly involves an asm routine at the other end of the pointer.
Is there any way to search for this stuff more quickly?
Also, the table AFTER the nature names just point back to it. Figures.
Edit: Eh, it's not worth the amount of effort. Changing the neutral natures into +HP just isn't worth it.
It's a great idea, and it can be done. We just don't know much about nature yet. However, in case you ever feel like you want to give it another try, I have noticed that simillar things tend to be stored near each other. So, the table your looking for could only be a little bit away. Don't know.
There is a better way and it involves debugging. If you can see what part of the ram is being called upon when a nature is loaded, you can seach for those parts in the rom and find it. But its tough because it's not as obvious when natures are called upon compared to things like attack. I'm sure someday, someone will figure it out.
Bad pointer? Is it for all of them, or just that one move. If it is just the one, there was probobly just a problem with your offset. Just try recompiling it and then put in the new offset. If it's all them, it could be anything from a table error or a pointer to table error. Experiment with defferent things and see if something you do causes a change.