Since the beginning of January I've been on a diet. I'm not new at this, I've dieted before, although I failed to hold the weight down over a long period of time. At heart, however, I'm a bit of a closet statistician. I always want to know how far I am, how far I have to go, and what happens next. I'm currently developing a method for tracking almost everything: this is my first guide.
Most diets say "Don't weight yourself more than once a week". There are good reasons for this: most people lose weight slowly, maybe a pound, maybetwo pounds every week at most. I disagree with these instructions. Partially this is because my weight loss if averaging around 3.5 pounds a week. I tend to think that the best way of encouraging yourself when you're on a diet is to see progress. On average I'm seeing half a pound of progress every day. Since my scales are accurate to the half pound, this is a good reason to make sure I see how I'm doing every morning. However the real reason is that weight fluctuates, and there isn't much you can do about it. Sometimes food sticks about in your body for a few days. Water weighs about 2 pounds a litre, that's about a pound a pint. If you drink a pint of water before you weigh yourself, that's a pound you're going to hate yourself for putting on. If you're only losing a pound a week, its quite possible you'll find yourself heavier one week than you were the week earlier, even if you've been sticking to the plan.
I have a different solution:
I weigh myself every day. And as I showed above I see progress everyday. Except I don't. Just like people who weigh themselves weekly, Is ee my weight stay the same from day to day, bob up or fall by several pounds at a time, even though my daily intake of food stays very similar, it certainly doesn't vary by more than 3500 calories each way in a given day. The solution isn't hard. Each day I take an average of my weight over the past week. 7 measurements should allow me to cancel out the worst of the errors. It also lets me cancel out the fact that my food consumption is different over the weekend than over the week.
However the weight this gives me is slightly wrong. Its out of date. This is the average of what I weighed 3 days ago. I don't know how I'm doing, but how I was doing 3 days ago. How do I know how I'm doing now? Well, what I can do is look at the weight I weighed this morning. If I'm sticking to the average, and I'm losing the 3 and a half pounds a week I expect, then I should be 1 and a half pounds lighter. But actually, my distance from the moving average varies. What I've noticed is that some times I'm a long way below the moving average, and some days I'm close to it. When I'm a long way below the moving average, I tend not to lose much weight over the next few days: looking at my daily weight measurements, you would think I was gaining weight, or plateauing. But I'm not. My moving average is still dropping, and as I get back to around 1 and a half pounds below, my weight begins to drop too.
I am interested in how much weight I lose over a week. But while looking at the difference from one week to the next makes more sense when you're losing weight at the rate I am, it doesn't make great sense. If I have a lot to drink (just water, naturally) on the night before the weigh-in, I may look like I've only lost a little. If some food just won't leave my body, again, my diet isn't doing as well as it was. So now, I'm taking the average of the last week and subtracting that from the average of the week before: the result, a far more consistent picture of day on day, week on week, weight loss.
Will this work for you? Well, lets assume you are only losing a little each week. In week one you weigh yourself and see on subsequent days
15.1, 15.1, 15.0, 15.1, 15.1, 15.1, 15.0
the next week you see
15.0, 15.1, 15.1, 15.0, 15.0, 15.1, 15.1
you begin to get disheartened. In the final week you see
15.0, 15.0, 15.0, 15.0, 15.0: a good start, but then disaster, 15.1, 15.2!
The diet isn't working. You give up and decide a career in sumo wresting is your only choice.
But using the moving average system we get the following measurements
15.07, 15.06, 15.06, 15.07, 15.06, 15.04, 15.04, 15.06, 15.06, 15.04, 15.03, 15.03, 15.03, 15.03, 15.04
Sure, you're wobbling, but the downward trend is clear: in fact you can see you're losing weight, when the numbers you see every morning just don't make it obvious at all.
Other top tricks
I measure lots of other things too. At first I was interested in measuring how much of my total body weight I've lost so far. Simple maths, not too interesting. Other measurements have occurred to me, so far the most useful is a measure of what percentage of my target weight loss I've achieved. My target weight is really quite a long way off, and after only less than a month, the figures I'm getting already look comforting.
I'm also measuring what percentage of the difference between my moving average weight and my target weight I am losing each week. As the gap between my target weight and my actual weight decreases, this should cause the number to rise, but as my weight decreases I expect to lose less each week. In theory, I guess these two issues will balance themselves out, more or less. So long as this percentage stays in the same ballpark (actually, I would expect a slight upward drift, otherwise I could never reach my target weight), I'm probably doing okay.
I'm playing around with other figures, figuring out statistics like my basal metabolic rate (BMR) every day. So far I'm not finding it that useful, and I have some tweaks to make (it assumes my age is constant, rather than getting higher every day). I also work out my weight in kilos, not because I care, but because my Fiancee is European, and thus incapable of using sensible units of measurement (my car gets 40 rods to the hogshead and that's the way I like it). At the end of the month, I plan on publishing my spreadsheet, so you can see how I'm doing. I'll also put together a blank in case you don't know how to produce one myself.
Most diets say "Don't weight yourself more than once a week". There are good reasons for this: most people lose weight slowly, maybe a pound, maybetwo pounds every week at most. I disagree with these instructions. Partially this is because my weight loss if averaging around 3.5 pounds a week. I tend to think that the best way of encouraging yourself when you're on a diet is to see progress. On average I'm seeing half a pound of progress every day. Since my scales are accurate to the half pound, this is a good reason to make sure I see how I'm doing every morning. However the real reason is that weight fluctuates, and there isn't much you can do about it. Sometimes food sticks about in your body for a few days. Water weighs about 2 pounds a litre, that's about a pound a pint. If you drink a pint of water before you weigh yourself, that's a pound you're going to hate yourself for putting on. If you're only losing a pound a week, its quite possible you'll find yourself heavier one week than you were the week earlier, even if you've been sticking to the plan.
I have a different solution:
I weigh myself every day. And as I showed above I see progress everyday. Except I don't. Just like people who weigh themselves weekly, Is ee my weight stay the same from day to day, bob up or fall by several pounds at a time, even though my daily intake of food stays very similar, it certainly doesn't vary by more than 3500 calories each way in a given day. The solution isn't hard. Each day I take an average of my weight over the past week. 7 measurements should allow me to cancel out the worst of the errors. It also lets me cancel out the fact that my food consumption is different over the weekend than over the week.
However the weight this gives me is slightly wrong. Its out of date. This is the average of what I weighed 3 days ago. I don't know how I'm doing, but how I was doing 3 days ago. How do I know how I'm doing now? Well, what I can do is look at the weight I weighed this morning. If I'm sticking to the average, and I'm losing the 3 and a half pounds a week I expect, then I should be 1 and a half pounds lighter. But actually, my distance from the moving average varies. What I've noticed is that some times I'm a long way below the moving average, and some days I'm close to it. When I'm a long way below the moving average, I tend not to lose much weight over the next few days: looking at my daily weight measurements, you would think I was gaining weight, or plateauing. But I'm not. My moving average is still dropping, and as I get back to around 1 and a half pounds below, my weight begins to drop too.
I am interested in how much weight I lose over a week. But while looking at the difference from one week to the next makes more sense when you're losing weight at the rate I am, it doesn't make great sense. If I have a lot to drink (just water, naturally) on the night before the weigh-in, I may look like I've only lost a little. If some food just won't leave my body, again, my diet isn't doing as well as it was. So now, I'm taking the average of the last week and subtracting that from the average of the week before: the result, a far more consistent picture of day on day, week on week, weight loss.
Will this work for you? Well, lets assume you are only losing a little each week. In week one you weigh yourself and see on subsequent days
15.1, 15.1, 15.0, 15.1, 15.1, 15.1, 15.0
the next week you see
15.0, 15.1, 15.1, 15.0, 15.0, 15.1, 15.1
you begin to get disheartened. In the final week you see
15.0, 15.0, 15.0, 15.0, 15.0: a good start, but then disaster, 15.1, 15.2!
The diet isn't working. You give up and decide a career in sumo wresting is your only choice.
But using the moving average system we get the following measurements
15.07, 15.06, 15.06, 15.07, 15.06, 15.04, 15.04, 15.06, 15.06, 15.04, 15.03, 15.03, 15.03, 15.03, 15.04
Sure, you're wobbling, but the downward trend is clear: in fact you can see you're losing weight, when the numbers you see every morning just don't make it obvious at all.
Other top tricks
I measure lots of other things too. At first I was interested in measuring how much of my total body weight I've lost so far. Simple maths, not too interesting. Other measurements have occurred to me, so far the most useful is a measure of what percentage of my target weight loss I've achieved. My target weight is really quite a long way off, and after only less than a month, the figures I'm getting already look comforting.
I'm also measuring what percentage of the difference between my moving average weight and my target weight I am losing each week. As the gap between my target weight and my actual weight decreases, this should cause the number to rise, but as my weight decreases I expect to lose less each week. In theory, I guess these two issues will balance themselves out, more or less. So long as this percentage stays in the same ballpark (actually, I would expect a slight upward drift, otherwise I could never reach my target weight), I'm probably doing okay.
I'm playing around with other figures, figuring out statistics like my basal metabolic rate (BMR) every day. So far I'm not finding it that useful, and I have some tweaks to make (it assumes my age is constant, rather than getting higher every day). I also work out my weight in kilos, not because I care, but because my Fiancee is European, and thus incapable of using sensible units of measurement (my car gets 40 rods to the hogshead and that's the way I like it). At the end of the month, I plan on publishing my spreadsheet, so you can see how I'm doing. I'll also put together a blank in case you don't know how to produce one myself.
No comments:
Post a Comment