Body Weight and Trend Analysis
mdgivens
Posts: 206 Member
(...thanks to m23 for asking me to write this...)
I weigh myself every single day, first thing in the morning. It's a habit, so much so that I recently bought a travel scale. As we all know, weight can vary wildly from day to day. Well, my mood for the day used to follow that morning's numbers on the scale. I would be overjoyed when my weight went down that morning, and so discouraged when it went up. Up two days in a row? Sure-fire ticket to Blues-ville (not St. Louis). Now, they are just numbers that don't mean much individually other than how much water I have in my system at that time.
The reason that I can divorce myself emotionally from the day-to-day fluctuations of my weight is because I watch the big picture of how my weight is progressing. I got the idea from a book called The Hacker's Diet. *** I highly recommend reading this book. *** It's freely available on the Internet, and it was written by the guy who founded AutoDesk. In the book, he describes the systems and practices that he used to lose 70 pounds back in the 90's and keep it off to this day. It helped me to learn not to live and die by the scale and just treat every day as another data point. This was HUGE for my mental outlook on what I am doing, and this thing is 80% mental.
Here are my daily weigh-ins for the past 31 days:
If I just had the raw numbers to go by, I'd be a wreck with all of the up-and-down.
The red line is the important one, the moving average or daily trend. It smooths out the raw weight numbers and shows what's really going on. It tends to lag behind the raw numbers, but at a glance, I can see whether what I am doing is working or not, and I have time to make corrections and dial in my diet before things get out of hand. Here's how it's calculated:
Excerpt from The Hacker's Diet, p.282-283
Calculating daily trend
When you first start keeping your log, the very first day, enter your weight in the “Trend” column as well as the “Weight” column. Thereafter, calculate the number for the “Trend” column as follows:
1. Subtract yesterday’s trend from today’s weight. Write the result with a minus sign if it’s negative.
2. Shift the decimal place in the resulting number one place to the left. Round the number to one decimal place by dropping the second decimal and increasing the first decimal by one if the second decimal place is 5 or greater.
3. Add this number to yesterday’s trend number and enter in today’s trend column.
That's it. For specific examples, check out the above-referenced pages in the book. I do this in a spreadsheet because I like making numbers look pretty and I hate drawing graphs, but it's straightforward to do this with pencil-and-paper too.
I hope that this has been helpful to someone. Hit me up if you have questions.
I weigh myself every single day, first thing in the morning. It's a habit, so much so that I recently bought a travel scale. As we all know, weight can vary wildly from day to day. Well, my mood for the day used to follow that morning's numbers on the scale. I would be overjoyed when my weight went down that morning, and so discouraged when it went up. Up two days in a row? Sure-fire ticket to Blues-ville (not St. Louis). Now, they are just numbers that don't mean much individually other than how much water I have in my system at that time.
The reason that I can divorce myself emotionally from the day-to-day fluctuations of my weight is because I watch the big picture of how my weight is progressing. I got the idea from a book called The Hacker's Diet. *** I highly recommend reading this book. *** It's freely available on the Internet, and it was written by the guy who founded AutoDesk. In the book, he describes the systems and practices that he used to lose 70 pounds back in the 90's and keep it off to this day. It helped me to learn not to live and die by the scale and just treat every day as another data point. This was HUGE for my mental outlook on what I am doing, and this thing is 80% mental.
Here are my daily weigh-ins for the past 31 days:
If I just had the raw numbers to go by, I'd be a wreck with all of the up-and-down.
The red line is the important one, the moving average or daily trend. It smooths out the raw weight numbers and shows what's really going on. It tends to lag behind the raw numbers, but at a glance, I can see whether what I am doing is working or not, and I have time to make corrections and dial in my diet before things get out of hand. Here's how it's calculated:
Excerpt from The Hacker's Diet, p.282-283
Calculating daily trend
When you first start keeping your log, the very first day, enter your weight in the “Trend” column as well as the “Weight” column. Thereafter, calculate the number for the “Trend” column as follows:
1. Subtract yesterday’s trend from today’s weight. Write the result with a minus sign if it’s negative.
2. Shift the decimal place in the resulting number one place to the left. Round the number to one decimal place by dropping the second decimal and increasing the first decimal by one if the second decimal place is 5 or greater.
3. Add this number to yesterday’s trend number and enter in today’s trend column.
That's it. For specific examples, check out the above-referenced pages in the book. I do this in a spreadsheet because I like making numbers look pretty and I hate drawing graphs, but it's straightforward to do this with pencil-and-paper too.
I hope that this has been helpful to someone. Hit me up if you have questions.
0
Replies
-
Cool, Weight loss by science! I'll have to look this up when I get home and on my PC and setup a spreadsheet, although I tend to not weigh every day, just for the reasons you said you fought.
Thanks for sharing!0 -
(See dward, I TOLD you this guy was totally cool!)
"one of us. one of us. One of us ONE OF US..."
Brilliant stuff! I am a great fan of science!
Thank you so much for sharing this with us--fascinating!
(We have a strong geek streak in our little community--I think quite a few people will enjoy this.)0 -
That is super cool! I do weigh in often and get sad when it goes up! It's what I like about the ticker it shows my over all not just my day!0
-
Thanks for the kudos, y'all. I know this has helped me immensely, and I hope that it can help someone else too.
Unfortunately, the data columns are hidden in the original post. :grumble: If you right-click on the pic and download or open in a new window/tab, you can see the numbers upon which the chart is based. These are important for another number I look at called the Variance, which is the difference between the actual weight and the trend value for that day. When the variance is negative (actual weight is below the trend line), my weight is trending down. When the variance is positive (actual weight is above the trend line), my weight is trending up.
I mentioned earlier that the trend tends to lag behind the actual numbers. This is good, because it lets the Variance act as an early-warning system of sorts. If I've been losing at a good clip (variance is consistently negative), and then all of a sudden, the variance starts approaching zero on a consistent basis (i.e. a week or two), that says to me that I need to take a closer look at what I'm doing with regard to diet, exercise, sleep, etc.0 -
Love the trend analysis idea! I am a statistical nerdy mcgeek pants, so this is right up my alley! I have done this same trending using my weekly results and incorporating measurements for baseline comparison. It's very helpful to see the patterns of loss and allows me to do some predictive analysis for future weeks.
Thanks for sharing!
CC0 -
^^Love the trend analysis idea! I am a statistical nerdy mcgeek pants, so this is right up my alley! I have done this same trending using my weekly results and incorporating measurements for baseline comparison. It's very helpful to see the patterns of loss and allows me to do some predictive analysis for future weeks.
Thanks for sharing!
CC
See, now that there is what I would call a LEVEL UP. I'm still trying to figure out the predictive modeling/analysis game myself.
Care to share with the rest of the class?0 -
^^Love the trend analysis idea! I am a statistical nerdy mcgeek pants, so this is right up my alley! I have done this same trending using my weekly results and incorporating measurements for baseline comparison. It's very helpful to see the patterns of loss and allows me to do some predictive analysis for future weeks.
Thanks for sharing!
CC
See, now that there is what I would call a LEVEL UP. I'm still trying to figure out the predictive modeling/analysis game myself.
Care to share with the rest of the class?
I'm suddenly feeling diminutive, and not because of any weightloss. Putting on my nerd failure hat now.... lol0 -
I started to do all of this in Excel, but something just wasn't clicking, so I googled the Hackers Diet.
All of this is available in Excel 2003 if you want to download it, but even better it is available free online so you can access it from any device these days.
http://www.fourmilab.ch/hackdiet/online/hdo.html
I'm not sure I'm truly enough of an "engineering thought based" person to sink to the levels that I can see this going, but some more playing with it may make it an incredible tool to add into this battle. You can never have too many tools. You need to be able to find the right ones for the job.0 -
I started to do all of this in Excel, but something just wasn't clicking, so I googled the Hackers Diet.
All of this is available in Excel 2003 if you want to download it, but even better it is available free online so you can access it from any device these days.
http://www.fourmilab.ch/hackdiet/online/hdo.html
I'm not sure I'm truly enough of an "engineering thought based" person to sink to the levels that I can see this going, but some more playing with it may make it an incredible tool to add into this battle. You can never have too many tools. You need to be able to find the right ones for the job.
Thanks for filling in the info on where to find the book, Dan. One of the things that I like about the trend stuff is that you don't need a fancy spreadsheet to figure it out; it can be done with paper, pencil, and a calculator if necessary.
With that said, the analysis part of it has been an immensely helpful tool for me. Just this morning, I found something in the data which I think will help me out quite a bit with the mental aspect of this thing we do. Normally, my weight loss tends to follow a pattern where I lose steadily during the week and have a slight gain over the weekend, then return to the pattern on Monday. Lately, this pattern has changed, and gotten much more "wavy". I'm still losing from week-to-week, but I'll have slight gains intermittently during the week.
When I plugged in the last 4 weeks of data into the chart that I posted earlier, I noticed that the pattern changed around the time that school started back up. I had to change my workout schedule on strength days to fasted walking in the morning and lifting in the evening after my kids go to bed. I noticed that the gains were consistently on the days after lifting the previous night (e.g. this morning). When I lift, I lift heavy, and to failure, so this suggests to me that the gains are actually excess water and glycogen being retained by my muscles that are still healing themselves from the previous night's work. This may not be the case, but it's something to watch. If the pattern bears out, then I shouldn't have to worry about the intermittent gains so much, as they will work themselves out eventually (no pun intended).
If you're interested in adding it to your toolbox, keep messing around with it. Feel free to PM me if you have questions. I may not have all of the answers, but I'll help you find them, and we'll both learn something in the process.
Better yet, PM CC, our resident statistics and predictive modeling guru :bigsmile:0 -
My apologies for having not responded regarding the predictive analysis for trending...I will try to take some time this weekend to put together some thoughts for all our resident geeks!!
Dan - I work in the confines of Excel as well (and I don't have an engineering mind, I am more of a number-crunching accountant type)!
Givens - See what you started?! Now we probably have lurkers pouring over the Hackers Diet and analyzing weight loss trends prior years...King of Nerds!
CC0 -
Givens, I think you are probably dead on with your lifting followed by weight retention or gain. I'm pretty sure I was up so much this week because I started lifting and over did it. I think water retention while muscle healing is the primary blame, but I'm not a scientist.
My son just threatened to take the iPad away. Football has returned!0 -
My apologies for having not responded regarding the predictive analysis for trending...I will try to take some time this weekend to put together some thoughts for all our resident geeks!!
CC
Looking forward to it, CC. Thanks!0
