Advanced Layer: Calculated fields.

**Is your feature request related to a problem? Please describe.**
I'd like to be able to see how different calculations impact the visualizations.

**Describe the solution you'd like**
Kepler is awesome and allows me to visualize my data based on any pre-processed field, but what if I want to base my visualization on the result of a calculation on two fields? As far as I can tell, I could only do this by doing the calculation before and adding it to my dataset as a static value. This works in some cases but prevents on-the-fly experimentation. Pre-calculating is fine if you know you need a particular value, but if you're experimenting with lots of possibilities it means you have to have very bloated data with every possibility.

**Describe alternatives you've considered**
Pre-calculating every conceivable value and adding it to the dataset.

**Additional context**
The data set I care about is sailboat races. This is esoteric but I think a good example of a feature that would be useful for a lot of esoteric situations in a variety of industries. In sailing, to make races fair, boats that are inherently faster by design typically get time added to their race elapsed time, and slow boats typically get time removed from their race time. There are countless ways to calculate how the race times are normalized for fairness. The result of the race is based on this calculated time, not the actual elapsed time because if we just used elapsed time races would be decided on who owned the fastest boat not who was the best sailor. If I could create on-the-fly values based on some custom calculation, I could make maps and animations without having to pre-process my data.