Well if you had to have one or the other, I think you are actually right. Prob and stats are essential in any discussion of real-world data, which comes up pret

Well if you had to have one or the other, I think you are actually right. Prob and stats are essential in any discussion of real-world data, which comes up pretty much every day. They are also the best way into dealing with something inherently messy, which data pretty much is all the time. It's a good way to talk about what you think is true and why, critical to decision making. Calculus on the other hand is bit too beautiful. At high school level at least, you get all these smooth curves, everything has a nice closed form answer, and you start to think that maybe these beautiful shapes describe the universe. It's actually not that easy to do, it just happens that we found some examples where it works quite well, and they are just the kinds of things we teach in physics in those same years. When we then get to an economics class, it unravels. There's a whole menagerie of calculus assumptions that creep into the minds of credulous students. Lets assume Cobb-Douglass production function. Let's say the utility must be this shape. Oh look at what beautiful conclusions we get.