Plateau? No.
Aaron_K123
Posts: 7,122 Member
I've seen many a thread started on this forum by people who are concerned by their "plateau" of 2-4 weeks in duration. The evidence for this "plateau" is that their weight is the same now as it was 2-4 weeks ago. Although it is certainly possible that their weight loss as stalled, meaning they are at maintenance, there is an alternative and I think more likely explanation.
To give this alternative I am going to use my own experience as an example.
On 5/26/2014 I weighted 172.6. 4 weeks later on 6/25 I weighed 172.4, pretty much the same. Plateau right? Well what does that look like on MFP's graph. I weigh myself almost every day and here is what that month looks like.
Yeah, from 5/26 until now for a month that seems like a pretty flat line, no progress it seems. At this point is where I feel people go to write a post to ask for help on what they need to change to get back on track.
May I suggest one more step before considering yourself in a "plateau". Don't look at just two data points (weight a month ago, weight now). Don't even look at all the data just arranged in a simple graph. If you really want to see what is going on you should do a moving average. Each datapoint is then an average of the previous 7 days. So what does this one month long period look like in moving average?
Now there is a clear downward trend. In fact from the trend fit I can tell I am losing a pretty steady 1 pound per week during that time which is right on schedule.
If I show this as my weight week to week during that period using the moving average data it looks like this:
5/19/14: 176.8 +/- 1
5/26/14: 174.4 +/- 1.5
6/2/14: 174.7 +/- 0.9
6/9/14: 173.2 +/- 0.9
6/16: 172.6 +/- 1.1
6/23: 171.6 +/- 1.2
Again, about 1 pound per week and pretty steadily and a significant decrease from the beginning of that period to the end.
So if that is true why did it look like a plateau before? Well on 5/26 I changed my workout routine slightly. This change caused changes in my body where I was retaining more water, going through bigger swings in glycogen and generally having a lot more fluctuation in my weight. This noise made it appear, especially to someone who is looking for problems, like I had stalled. If you use a moving average though this smooths that out, gets rid of those fluctuations, and lets you see the underlying trend. What that tells me is the 5/26 172.6 weight was a low point in the fluctuations while the 6/25 172.4 is a high point in the fluctuation. Suddenly....no plateau, no need to worry, no need to change a thing.
I feel that way to many people on here see their weight not change over a 2-4 week period, assume they are doing something wrong and start making drastic changes to their diet or exercise to compensate which just ends up exacerbating the "problem".
So what should you do? If your "plateau" has only been going on for weeks just stick with what you are doing or at least look at a moving average or some other means of measurement. Don't psych yourself out over your weight not "changing" just because on one day a month later you happen to be the same weight as a previous day a month before.
I'm not saying that for anyone who experiences a stall that this is the reason, but it is usually my first guess if that persons stall is on the order of a few weeks. Sure, maybe you are suddenly at maintenance unexpectedly despite not changing much of anything in your approach and having previously been losing 1 to 2 pounds per week....but is that really likely?
To give this alternative I am going to use my own experience as an example.
On 5/26/2014 I weighted 172.6. 4 weeks later on 6/25 I weighed 172.4, pretty much the same. Plateau right? Well what does that look like on MFP's graph. I weigh myself almost every day and here is what that month looks like.
Yeah, from 5/26 until now for a month that seems like a pretty flat line, no progress it seems. At this point is where I feel people go to write a post to ask for help on what they need to change to get back on track.
May I suggest one more step before considering yourself in a "plateau". Don't look at just two data points (weight a month ago, weight now). Don't even look at all the data just arranged in a simple graph. If you really want to see what is going on you should do a moving average. Each datapoint is then an average of the previous 7 days. So what does this one month long period look like in moving average?
Now there is a clear downward trend. In fact from the trend fit I can tell I am losing a pretty steady 1 pound per week during that time which is right on schedule.
If I show this as my weight week to week during that period using the moving average data it looks like this:
5/19/14: 176.8 +/- 1
5/26/14: 174.4 +/- 1.5
6/2/14: 174.7 +/- 0.9
6/9/14: 173.2 +/- 0.9
6/16: 172.6 +/- 1.1
6/23: 171.6 +/- 1.2
Again, about 1 pound per week and pretty steadily and a significant decrease from the beginning of that period to the end.
So if that is true why did it look like a plateau before? Well on 5/26 I changed my workout routine slightly. This change caused changes in my body where I was retaining more water, going through bigger swings in glycogen and generally having a lot more fluctuation in my weight. This noise made it appear, especially to someone who is looking for problems, like I had stalled. If you use a moving average though this smooths that out, gets rid of those fluctuations, and lets you see the underlying trend. What that tells me is the 5/26 172.6 weight was a low point in the fluctuations while the 6/25 172.4 is a high point in the fluctuation. Suddenly....no plateau, no need to worry, no need to change a thing.
I feel that way to many people on here see their weight not change over a 2-4 week period, assume they are doing something wrong and start making drastic changes to their diet or exercise to compensate which just ends up exacerbating the "problem".
So what should you do? If your "plateau" has only been going on for weeks just stick with what you are doing or at least look at a moving average or some other means of measurement. Don't psych yourself out over your weight not "changing" just because on one day a month later you happen to be the same weight as a previous day a month before.
I'm not saying that for anyone who experiences a stall that this is the reason, but it is usually my first guess if that persons stall is on the order of a few weeks. Sure, maybe you are suddenly at maintenance unexpectedly despite not changing much of anything in your approach and having previously been losing 1 to 2 pounds per week....but is that really likely?
0
Replies
-
Ahhhh dang, Aaron! There ya go with your "high brow" science and math again!
Seriously, yes. Yes! YES!
I track daily weigh-ins in Excel and trendline and all that good stuff, too. If I didn't know better, my June so far looks like a "plateau." Alas, it is not. Still losing, albeit slowly and surely and healthily and stuff.
Also, Libra app for Android (sorry iOS peeps, no Libra for you) does the same trend and forecasting, without the hassle of Excel spreadsheets.
:drinker: 0 -
Ahhhh dang, Aaron! There ya go with your "high brow" science and math again!
Seriously, yes. Yes! YES!
I track daily weigh-ins in Excel and trendline and all that good stuff, too. If I didn't know better, my June so far looks like a "plateau." Alas, it is not. Still losing, albeit slowly and surely and healthily and stuff.
Also, Libra app for Android (sorry iOS peeps, no Libra for you) does the same trend and forecasting, without the hassle of Excel spreadsheets. :drinker:
I do not know what you mean by "hassle" with regards to excel. Everyone loves excel. Right?0 -
Ahhhh dang, Aaron! There ya go with your "high brow" science and math again!
Seriously, yes. Yes! YES!
I track daily weigh-ins in Excel and trendline and all that good stuff, too. If I didn't know better, my June so far looks like a "plateau." Alas, it is not. Still losing, albeit slowly and surely and healthily and stuff.
Also, Libra app for Android (sorry iOS peeps, no Libra for you) does the same trend and forecasting, without the hassle of Excel spreadsheets. :drinker:
I do not know what you mean by "hassle" with regards to excel. Everyone loves excel. Right?
For regression analysis? No. For general "piddling" around with number fun, and making weight loss as nerdy as possible, without traveling down to the university, then yes. Excel will do. In a pinch.
0 -
What...you want me to bust out Graphpad PRISM or R on the MFP forums? Think that is a bit overkill. Plus excel is the one spreadsheet style program people are most likely to have if any.
0 -
:drinker: Looks like another post that needs to be pinned to this board.
I use an app on my phone called Weight Trend + that does a pretty good job of tracking overall trends both with a graph and a chart. Being able to look at the big picture instead of just a short moment in time does put things into perspective when the scale starts coming to a halt.0 -
Great post!
I'm in maintenance so mine's a little different but I also track daily in a spreadsheet. In the past month I've had weigh-ins between 125 and 129. I was seeing a lot more 128s and 129s so I was starting to think I was creeping up, but I average my weekly numbers and here are those numbers:
126.4
126.9
126.3
126.6
126.2
Even though it looks like I went up in week 2 and week 4, by week 5 I was a little lower than week 1.0 -
I've been doing this with Excel and a couple of iOS apps since I started at the end of April. It really is kind of amazing how much clarity a trend line brings. I mean, I've seen it before, but not on something that has the potential to mess with my mind quite as much as my weight jumping all over the place.0
-
To give even more context here is a moving average graph of my weight loss since I started on my exercise and fitness kick about 4 months ago.
There is a little blip around 5/26 where I changed my exercise routine but it quickly returns to the trend. In this context basically nothing has changed and yet from all the posts I've seen here of "Why aren't I losing?" I get the sense that 2-4 weeks of no clear change by individual numbers psychs people out.
Also just wanted to add that with a moving average that "weight loss isn't linear" comment sort of goes away. R-square of 0.99 pretty darn linear.0 -
Absolutely on point.
For any that might benefit, I made a "Weight Fluctuation" entry to note these changes rather than have my ticker fluctuate (and trigger any crash dieting or over-exercising). That tracking has saved me a few near meltdowns from hopping on the scale and has made me far more aware of how my body reacts to certain foods, workouts and hormones.0 -
This content has been removed.
-
Sure, maybe you are suddenly at maintenance unexpectedly despite not changing much of anything in your approach and having previously been losing 1 to 2 pounds per week....but is that really likely?
When you get into the science of it... Yes, it's actually likely... But to add to it that's also because the majority of people are dieting like the majority of people and not what it takes to get the job done.
So first things first - reasons for temporary changes in weight. The biggest reason for rapid changes in weight is changes in glycogen storage. Glycogen is our muscles primary fuel and at full capacity we store enough for 72 hours. This is why ketosis normally takes at least 3 - 4 days to set in. Now as a person reduces caloric intake glycogen stores will also decrease which consequently causes a decrease in water. The body doesn't store more glycogen because glycogen requires water to utilize hence the preference to store fat for long term energy. So when on caloric restriction glycogen stores drop therefore dropping water retention therefore causing a rapid loss in weight.
Depending on how calorie restrictive you are will dictate how quickly your glycogen levels drop. So if you fast, for example, it will completely deplete in around 3 days. If you're on a calorie restrictive diet it can take 2 - 4 weeks easily, hence the first major reason why people see a perceived "plateau" around this point. The trouble is it could actually be a real plateau because even though they might have just lost 5 - 10 lbs in 2 - 4 weeks there is no telling if they've actually lost any fat! Fat isn't greatly impacted by water retention because it only stores around 15% water in adipose tissue but lean mass is around 80% water weight. So most of these short term changes are actually considered to be lean mass losses.
The next thing you have to do is account for BMR adaptation which can actually occur in as little as 36 hours, but even with dieting methods like alternate day fasting can fully adapt within weeks. Studies show over and over again that in caloric restriction BMR can drop as much as 25%. What this means is that if you are only using a 200 - 300 calorie deficit within a few weeks it's very possible your BMR has dropped by that much.
Additionally for obese people they have to deal with BMR reduction from reductions in leptin. Being obese permanently alters the body physiologically. When an obese person loses weight regardless of how fast/slow or whatever diet protocols leptin levels drop causing drops in BMR. Even when returning to eating normal leptin will remain dropped until weight is restored to the starting point. It is clear scientific evidence for both plateaus and why obese people struggle with weight loss more than non-obese people.
There are a lot of other factors and support for this too, but that'd be a thread on it's own. The plateau has been scientifically established as real; however, are you really on a weight loss diet if you aren't losing weight? That's like a having a broke car, a mechanic saying he fixed it just by doing something, and the car still is broke. You have to be losing weight to be on a weight loss diet otherwise you're implementing it incorrectly. So again plateaus are real, but you have to understand the different mechanisms so you can fix it and start losing weight again.
References:
http://ajcn.nutrition.org/content/68/3/599.full.pdf
http://www.ncbi.nlm.nih.gov/pubmed/23555620
http://www.sportsci.org/encyc/adipose/adipose.html
http://ajcn.nutrition.org/content/48/2/240.full.pdf
Oh I believe that transitioning into maintenance unintentionally (referred to as plateauing) is real, I just feel it gets misdiagnosed quite often where a perceived stall due to fluctuations is read as a plateau. If I had done my moving average analysis and seen a flat line there too I probably would have considered re-addressing my plan and changing my intake or exercise.
The scenario in which I think unlikely that it is a true plateau is the scenario where you have a person who has been steadily losing weight for months at a significant rate (1+ pounds a week) who for the last 2-4 weeks doesn't seem to have lost weight by scale. Sure your BMR can reduce but it happens over time not all at once in one plummeting fall. Going from 1 pound to no loss at all is not going to happen overnight due to a change in your BMR. Of course it is possible that the person who thinks they have made no change to their diet is actually eating more all of a sudden or exercising less but thats on them and their accountability.
If your deficit is small, like the 200 cal deficit you suggested, I think its much more believable especially with error in calorie counting, that you would accidentally just transition into maintenance for a while. I just feel its less and less likely the larger the deficit you are running is.
I'm running a 600 cal deficit, have been losing a pretty steady 1-1.2 pound per week for about 4 months. If that suddenly stopped for several weeks I would consider the scenario I described above as more plausible than the one you described here. That is all, not claiming that plateauing never happens.0 -
Hacker's Diet Online has a similar function (it's free), and he has a waaaay in depth article about it to go along with it.
I weigh myself every day as well, because I love numbers, but it actually helped me not stress over the weeks where I didn't "lose", because I saw the daily fluctuations and realized that it might have looked like I didn't lose on a certain week if you took Friday-Friday weights, but if you had taken Thursday-Thursday, there would have been a loss.
The link is: https://www.fourmilab.ch/cgi-bin/HackDiet/
Tried to screen shot my graph, but it looks very similar to yours - average of 1.18 pounds per week lost over the last 6 mos, with an average calorie deficit of 590cal/day.
I tell people all the time that weight loss is all about math.0 -
This content has been removed.
-
Great post! Thank you! I just went through an 8 day "plateau" - but I knew I was excersising and counting calories correctly - just had to add in a bit of patience and trust in the "process".0
-
I appreciate the information!0
-
To give even more context here is a moving average graph of my weight loss since I started on my exercise and fitness kick about 4 months ago.
I'm in excel, selected moving average. What should my period say?0 -
Okay, I'm an English major, so anytime graphs are broken out my eyes begin to glaze over; however I do appreciate how you've broken it down for us. I wonder how set point weight plays into all of this or is that up just a myth?0
-
Bumping for later. 0
-
Good info, thanks to both the OP and btente for posting.0
-
As always, love it. I try tell people this all the time, but of course you've actually taken the time to break it down and explain it in a much clearer and more patient manner than I ever manage. As corroborating anecdotal evidence, I added marathon training to my schedule two weeks ago and haven't seen my usually 2 lb weight loss yet this month. Do I think I'm off schedule and not losing weight despite considerably increasing my calorie burn for the week? Nope. It'll start working it's way off again. Now, I'm in that cautionary range because at this point I have very little left to lose and am on a very small deficit, so if I hit next month and no loss, yea, then it's time to decide if I just want to move in to a very slow recomp, or drop another couple hundred calories off my intake to get the last couple pounds off.0
-
:flowerforyou:0
-
I love this thread so much. All the science and smart talk, it made my morning.
0 -
To give even more context here is a moving average graph of my weight loss since I started on my exercise and fitness kick about 4 months ago.
I'm in excel, selected moving average. What should my period say?
I used a 7 day period/interval for mine. The longer the interval the less fluctuations but the longer it takes for new measurements to change the overall value so tue more sluggish to respond to change. I found 7 days was enough to flatten the fluctuations for me but feel free to play with it, no wrong answer really (well okay an interval of 1 would be pointless).0 -
Isn't math great?! Great post. And Excel is not really that bad for a simple trend line like you've done here. Excellent post and a great way to look at the big picture. Thank you.0
-
Bookmarking. Thanks Aaron 0
Categories
- All Categories
- 1.4M Health, Wellness and Goals
- 393.3K Introduce Yourself
- 43.8K Getting Started
- 260.2K Health and Weight Loss
- 175.9K Food and Nutrition
- 47.4K Recipes
- 232.5K Fitness and Exercise
- 422 Sleep, Mindfulness and Overall Wellness
- 6.5K Goal: Maintaining Weight
- 8.5K Goal: Gaining Weight and Body Building
- 153K Motivation and Support
- 8K Challenges
- 1.3K Debate Club
- 96.3K Chit-Chat
- 2.5K Fun and Games
- 3.7K MyFitnessPal Information
- 23 News and Announcements
- 1.1K Feature Suggestions and Ideas
- 2.6K MyFitnessPal Tech Support Questions