IMO & IME, the optimum placement for the controller probe depends on if you are using a fan or not. When not using a fan, the best location is near the bottom of the freezer and not touching the wall. OTOH, if you are using a fan, then the optimum location is directly in the air stream of the fan and it doesn't make a lot of difference where the fan is located so long as it can keep the air moving throughout the freezer. The larger case fans work better than the very tiny ones and running the fan continuously works better than intermittently. I think that placing the probe in a container of water or a gel pack is a bad idea. Doing so just widens the temperature swing and if the container is large enough, this could cause big time over/undershooting the temps. There are move than a few variables interacting, such as the size of the freezer, how full it is, whether or not you use a fan, whether or not your controller has an adjustable differential etc, etc. As I said at the beginning, this is JMO based on some crude trial and error testing. A Kill-A-Watt meter thingy is very handy to check the power consumption when trying various configurations. When doing so, be sure to allow plenty of time for everything to equalize when making changes. Sometimes this can be more than just waiting overnight and especially so if you have a lot of full kegs or fermenters in the freezer. YMMV...blah...blah.