I agree with Steve. I would arrange to have the pot shaft be the hinge point for the pendulum. Since the pendulum probably does not even go all the way around a standard type of cermet element pot with 270 degree range of motion should be near perfect. Sampling with a 12 bit A/D converter would give you great resolution. (Assume a range of motion of +/- 45 degrees. Thats 90 degrees out of 270 on a pot so wiper range on a POT that is hooked to +5 & GND is going to see 1/3 Vcc range. For the simplest A/D setup with a 0->5 volt range your going to read 4096/3 = 1365 counts. This corresponds to a resolution of 15 counts per degree. Nice!!

It would even be possible to use a look up table to calibrate and linearize the pot readings to emperical pendulum position measurements.

A pot such as this one with 1/8 " shaft can be mounted in a way that a ball bearing can be used to support the end of the shaft...
http://rocky.digikey.com/WebLib/Cl...TA/381.pdf

Have Fun
Michael Karas