I'm planning to lose some fat over the next few months to get below 10%
fat. For the last week or so, I've been devising a method to do this.

For the last few months, I've been tracking my weight to see how it
behaves from day to day. To detect actual changes in body mass, I came
up with a method to measure the trend of my weight. I ended up using a
10-day exponentially smoothed moving average. This method uses
cumulative daily weigh-ins to compute a Trend Weight (this is basically
your actual weight) and is based on the Hacker's Diet formula. Anyway,
the way it works is if daily weigh-ins are above the Trend Weight (the
smoothed moving average mentioned above) much of the time, it indicates
fat gain. Conversely, if weigh-ins are below the Trend Weight most of
the time, then it indicates body fat loss. If daily weights are above,
below, and at the Trend Weight an equal number of times, then weight
and body composition is being maintained. When the Trend Weight
changes, there is always a change in body fat percentage according to
the electrical fat gauge. Basically, an increasing Trend Weight may
mean trouble and a rising fat% will catch your attention quicker than
just day-to-day fluctuating weigh-ins. A decreasing Trend Weight
usually means progress on the other hand. Here's an example of this:

Daily Weight Trend Weight Variance

7/29/05, Fri 151.6 152.01 -0.5
7/30/05, Sat 151.6 151.97 -0.4
7/31/05, Sun 150.2 151.79 -1.8
8/1/05, Mon 151.3 151.74 -0.5
8/2/05, Tue 153.2 151.89 1.5
8/3/05, Wed 152.4 151.94 0.5
8/4/05, Thu 149.2 151.66 -2.7
8/5/05, Fri 149.6 151.46 -2.1
8/6/05, Sat 148.7 151.18 -2.8
8/7/05, Sun 153.4 151.40 2.2
8/8/05, Mon 151.2 151.38 -0.2

So with this tracking system, I came up with an Appestat based on the
Variance each day. The Variance is basically the difference between the
daily weigh-in and the Trend Weight which is then divided by 10 and
used to subtract from or add to the Trend Weight to indicate change in
body fat. Positive Variance equals fat gain, and Negative Variance
equals fat loss.

Here's the Appestat ratings for my program:

Appestat
For Fat Loss
Kcal Intake
Opening Variance (Fully Open) -4.0 3,000 or 20 x Bodyweight
Closing Variance (Fully Closed) 1.5 1,500 or 10 x Bodyweight

Opening & Closing Ranges:

Variances Kcal Intake
1.5 1,200 (300x4meals)
0.5 - 1.5 1,500 (400x3meals, 300x1meal)
-0.5 - 0.5 1,800 (450x4meals)
-1.5 - -0.5 2,100 (500x3meals,600x1meal)
-2.5 - -1.5 2,400 (600x4meals)
-3.5 - -2.5 2,700 (can be distributed in any way)
-4.0 - -3.5 3,000 (can be distributed in any way)
-4.0 3,000 or Free Day

In this system, the Appestat controls the target number of Calories
going in during a fat loss program. The Appestat is closed when the
daily body weight is at a certain point above the Trend Weight (also
known as your true weight), allowing ONLY the minimum number of
Calories to get taken in, which is basically only enough to provided
necessary nutrition. This allows daily body weight to start dropping
below the Trend Weight so fat loss can occur. When the rate of fat loss
gets to a rapid steady pace, the Appestat will begin to open and allow
more Calories to be eaten. The faster the fat loss occurs, the more the
Appestat opens, allowing more and more food to be eaten. The Appestat
is basically a protocol for fat loss so you can get feedback on
progress and make adjustments as necessary. Overall, this Appestat is
imperative to my fat loss program. It works like a Thermostat in a
car's cooling system which regulates coolant flow and temperature.

For maintenance, I came up with an Adipostat. Here are the ratings for
it:

Adipostat
For Maintenance

Rating (Closing Weight): "My Goal Weight" (2 lbs over goal when fully
closed)
Opening Weight: (2 pounds below "My Goal Weight")

Kcal Intake When Closing 1,800 (BMR Level)
Kcal Intake When Fully Closed 1,500
Kcal Intake When Opened: 3,000

This also works similar to a thermostat in a car's cooling system in
that automotive cooling system thermostats have temperature ratings
like 160, 180, and 195.

I know this was a bit long, but I just wanted to share my method.