The past summer, I was at the REU in UMBC. The head of the math department guy preached about the future of mathematics is the intersection of mathematics, statistics and computing; I agree to a large extent. 
What I ended up working on seemed to be computing, with more computing, with very little mathematical analysis. The system of differential equations that described pancreatic cells was far too advanced for me to do some significant analysis. Heck, that’s how I feel about most of pure mathematics: the interesting problems are always out of reach from what I know.
Before this past winter break, I found a cool book called “Risk and Reward” in the math library. It’s all about the, surprisingly deep, game of casino blackjack, from the strategies to the mathematics behind the analysis. In the analysis portion of the book, 90% of all the results seems to be dependent on the a computer, but with relatively simple probability concepts.
If I ever do become a professor of some sorts, I want to be able to teach a class based on games like blackjack and poker. For projects of sorts, the task will be to calculate the probability of various conditions and to figure out the strategies from the math I provide in class. Maybe for a (optional) final, it’s to actually play the games that we studied in class…
As an exercise over the break, I took an existing program and added the functionality to “practice” the card counting went over the book. It’s based on this project which I butchered (honestly, I spent 2 days doing this… trying to understand someone else’s code is tough). To run the program, make sure Python is installed, and run Blackjack.py. Download