This is one of those situations where you can look at your past work and identify before you knew how to do something, and after. Data smoothing is one of those things. You can see the impact it has.
Smoothing data is such a valuable and useful technique to know when working in max. Sensor data is often full of noise and seems unusable at times, but add a bit of smoothing to that data and suddenly it's great! This demos shows how to smooth data for continuous streams of data and non-continuous data.
For continuous data, a one-pole low-pass filter can smooth out data in very nice ways. For non-continuous data, this demos shows how to ramp smooth between values.