GolferGuy - Look closely at the first post and the second where I have bold text and you will see a difference the IBM Growth table. I only duplicated this for clarity/correction, not for emphasis.

What is the data? There are 1300 of them, so just about every type of number you can think of. For example in growth:
growth of revenue yoy, qoq, yoy 12 month trail, qoq sequential, etc.
growth of earnings yoy, qoq, yoy 12 month trail, qoq sequential, etc.
growth of oper. cash flow yoy, qoq, yoy 12 month trail, qoq sequential, etc.
growth of free cash flow yoy, qoq, yoy 12 month trail, qoq sequential, etc.
...and on and on and on...just for growth.

As you can see it is not really important to the question that I have. They are just numbers of calculations. And for the few fields that have duplication, I know what to do with them.

The real issue is the dates (inf.) x tickers (3300) x data items (1300) size and how to best handle that for the fastest queries.

I don't need two separate tables for Growth and Value. I just thought that might be a good way to split up the high number of data fields. But after researching it a bit more, it is not.