I'm going to respectfully disagree with Hale here. .... Point is, if you do it correctly and are honest with yourself in terms of what you are eating and your activity level, calorie counting is one way that will always work.
I haven't said that counting calories is a bad thing.
Given you are a bit of an empiricist, let's consider. If you look at the screen shot at the top of this page, the MFP app reports that a 3/4 cup of cooked, steamed white Mahatma brand rice contains 4 grams of salt and 2 1/2 grams of potassium. Add the macros: 246 grams fat, 796 grams carbs and 76 grams of protein and you get 1118 grams of food. Now I've done a quick Google of how much water is in rice generally: " 1 cup - 6.1 oz. dry weighed 17.9 oz. cooked". That goes along with the concept of 2 cups of water to 1 cup of rice, provided we are discussing volumes and rice's specific gravity isn't far above one, and further assuming that an ounce is an ounce, though there is a 4% difference in size between imperial and US ounces and we are assuming US. How on earth does 6.1 oz. ~173 grams of rice contain 1118 grams of macros? That means a pretty full coffee cup of steamed rice would have very close to 2 1/2 lbs of food in it. Before you add the weight of the water!!! Last time I looked the water portion has zero calories. It is the ludicrousness of these numbers and the claims of Mahatma producers as to calories and salt in total contradiction to MFP that led to my comments.
Cory, you know what you need to solve an equation with 'n' variables, each of which is subject to error ranges compounded through the consequent system of equations. Doing it based on an averaged weight result gives you a statistical confidence r squared on essentially a single variable which coincides with what your tracking and can provide accurate data points.
I agree there is enormous value in understanding the calorie content of various foods to help guide choices and knowing that fat is around 9 calories per gram and protein and carbs are near enough 4.2 calories per gram. I'd be very impressed if any of those who've tracked for say, 6 weeks, would be able to take their calculated net calories over those 6 weeks, with averaged start week and end week weights, and have that equate to pounds times 3500, loss or gain. This is an untestable hypothesis as I don't track, but as many of you do, I'd like to know.
Lastly, I don't think it's particularly beneficial to eat to a particular count. I try to gauge my meals by what satisfies my appetite rather than 'x' scoops of 'n' size. While I'm not the best at reading how full I am, I have felt hunger and thirst and I've encountered neither in my 16 days posting here. Making a few changes and measuring the results seems to me, and apparently few others, a great way of evaluating whether your choices are making a difference. It's great that people do what works for them and I guess I'm asking whether it really does. Thermodynamics.