Well, I'd love to be able to build a program that analyzes a set of data to determine if there's any statistically significant patterns over a period of time. Ideally, this program would be able to automatically extrapolate the relevant data from source files such as excel, and then automatically generate graphical output of the results.
Python coupled with the PANDAS library would do that for you. Even the graphs can be made almost immediately if you setup iPython, which is an interactive version where you can play in. You could probably learn in one or two weeks enough basics to see some results and decide if you want to continue or not.
You should probably also set yourself some steps in achieving your ultimate goal; this wil make it much more achievable. Same thing as an entrepreneur does : create a minimum viable product and see if it works / attracts the masses. Build on it if successful.
For example :
Highlevel :
1/ learn how to extract the data and massage them into results
2/ learn how to show those results graphically : ie in a web form or in an app ?
Those two could be done separately. You could even learn to extract things in python and show them in swift/cocoa.
Slight less high-level:
1/ select a few xlsx files that can already give you a benefit if properly analysed/graphed
2/ export them to csv for testing - you can learn later on how to actually read inside the xlsx if you deem it useful
3/ install the anaconda python environment (or swift or ...)
4/ determine how to read those csv files in python/swift/whatever you choose
5/ try to use the tools / libraries supplied to calculate and create some results
6/ display those results in a graphical way.
Python has plenty of these tools already built in. Using the PANDAS library together with iPython you could very quickly get some results that will allow you to decide to continue or not.
Even if you decide to not continue, you'll have learned something