Correlation between steps taken and weight loss

[AlexLockard]

I've been an active Fitbit user since they first came out, ordering my first one in early 2010. So I've got 7-8 years of data to mine, and I would love to know if anyone out there has ever done a correlation between steps taken and floors gained (daily, weekly, monthly, etc.) and weight loss. I experienced great weight loss in my first 6-8 months (dropping 48 pounds), but then it slowly decreased for the next couple of years, and I hit my lowest weight in summer 2013. Over the last four years, it has gone steadily up, and I would like to know what activity level is needed for maintenance and actual loss. I am a very active hiker and mountaineer, so most people assume that I should be able to maintain and even lose weight easily, but that has not been the case for the last four years. Any ideas? I would like an actual statistical analysis, if anyone has the expertise. Not looking for generic "exercise more and eat less" advice. 

[Daves_Not_Here]

Hi @AlexLockard -- I followed your path almost exactly -- dropped 50 pounds in 2010, hit my minimum 2 years later, and then gained weight in spite of increased activity level.

The answer, which I realize you may not want to hear, is that I gained the weight back because I focused on activity instead of eating.  I ate more than I burned.  I assumed that because I was active, I could eat "healthy foods" to satiety, like I did when I initially lost the weight.

<insert here all the reasons why this doesn't work>

Weight control is 90% eating and 10% activity.   You are asking for a statistical analysis of the 10% factor in the absence of the 90% factor.  Here's the math: Weight change equals calories burned minus calories eaten divided by 3,500.  There is no way to develop a mathematical model that omits the key variable (eating), but statistically speaking, there is a 100% probability that you will gain weight if you eat more than you burn. 🙂

So the advice is not to "exercise more and eat less".  You are plenty active and can gain weight at any activity level.  The advice is to eat less.

[AlexLockard]

What's been your experience since then?

[Mukluk4]

First:

Exercise isn't step based, it's intensity based.  

Secondly:

It's not necessarily simple as eating less.  Depending on activity and intensity, you may need to eat more to get proper recovery.  A few ideas/questions that will determine what track you are on...

1. If this change occurred after turning 35 or 40, I would get some blood work done to get your T and Estrogen levels validated.  

2. Weight gain is not always a bad thing if it's done correctly.  Are you tracking your BF%, measurements, clothes sizes?  If so, how has that changed during this time.  If not, get on it as it's critical.  

3. Mountaineering is some intense s#!t, During this time of weight gaining, how has your performance been?  Are you better or worse than when you were at your lowest weight point.  

[Daves_Not_Here]

---

@AlexLockard wrote:
What's been your experience since then?

---

@AlexLockard -- my experience was that I remained active, and fairly healthy as evidenced by my endurance and blood panels.  But my quality of life deteriorated as I gained weight (sleep apnea, no wardrobe, embarrassing physical appearance, huffing and puffing up hills, etc.). 

Three months ago, I had the "what the hell are you doing" inner dialog, and decided to get to my ideal weight.  I'm halfway there. 

The mistake I made previously was to assume that because I had adopted an active lifestyle, I could freely eat "healthy foods" to satiety.  This time, I know I need to continue to monitor my weight and eating even after I have reached my target weight.

[AlexLockard]

Right, but I guess I was hoping to see if a certain threshold of steps was
needed for weight maintenance and loss, vs. gain. My motivation to get out
and make sure I hit my step goal has waned definitely over the last couple
of years. The feedback I get from my fitbit doesn't provide anywhere near
the level of motivation now that it did in the first year or two. I need to
see that if I only worked a little harder, I would catapult myself back
into weight loss territory. For quite a while, i was very consistent on
getting 12k - 15k (and many days, 20k+) steps daily, but now it's more like
5k to 10k probably. Over an extended period of time, I would think that has
to be a determining factor. But with such a huge data set (7-8 years of
step and floor data), I was hoping a statistician could tease out some
inferences.

[Daves_Not_Here]

@AlexLockard - my belief, and I'm just a guy on the internet with no formal training in sports medicine or related fields, is that if you "only worked a little harder", you will NOT catapult yourself back into weight loss territory.

I base this on my own very similar experience, and dozens of posts here from people who are extremely frustrated because they are working out for literally hours every day and cannot lose weight after an initial period of success.

It makes sense when you think about it:  the more you work out, the more fit you become, and the less calories you burn for a given level of activity.  And, at lower weight, your Base Metabolic Rate is lowered.  So you may need to walk 20,000 steps to burn what 12,000 steps used to take.  Meanwhile, as you increase your activity to compensate for all this, your appetite and eating increases.  I don't think it is statistically possible to tease out how much your appetite and eating will increase or how energy efficient you will become.

I believe it is unreasonable to expect to have to work out 2 or 3 hours a day to maintain a desired weight when you can simply reduce your eating by 500 calories per day to lose a pound a week.  But, the power is in your hands - try an experiment:  walk 15,000 steps a day for the next month, and see where it takes you.  It will probably motivate you to log your eating, at which point the weight will start coming off.

[Julia_G]

Hi, @AlexLockard, a 2016 study on weight loss, published in the Journal of American Medicine, showed that among the study participants, people issued with a tracking device (such as Fitbit) actually lost LESS weight than those who weren't given the device.  (Both groups were given a low calorie diet to follow.)

It seems that many people overestimate (and my experience from reading these forums vastly overestimate) the number of calories burned from exercise.  For most people, the majority of calories they burn are BMR calories (calories just from being alive).

Apparently people in the study looked at all their steps and thought they could cheat on their diet, just a littlle...

The moral of this story is not that Fitbits make you fat -- or that you shouldn't bother meeting your step goal.  I believe there are huge benefits from a significant walk every day -- physical and mental -- but I think unless you are undertaking a big challenge most people will need to focus on what they are eating as well as the steps they are taking.

To answer your question directly, how many steps it takes to maintain, lose or gain (given the same caloric intake) very much depends on your gender, how old you are and what you currently weigh.

[Daves_Not_Here]

@Julia_G - agreeing with everything you wrote.  Reason I feel so strongly about this is that I fell into the exact same trap.  I assumed my high activity level would allow me to eat mindlessly, as long as I ate "healthy" foods.

I also continue to see posts of the form:

• "I don't track my eating and work out 2 hours a day, but I can't lose weight.  How much harder do I need to work?"
• "I don't track my eating and work out 20 hours a day, don't sleep, smash my head against the wall for 4 hours a day, run double marathons on Fridays, and perform belly-flops off the high dive, but I can't lose weight.  How much harder do I need to work?"
• "I don't track my eating and I'll do absolutely anything except tracking my eating, but I can't lose weight.  How much harder do I need to work?"

I make fun of these neurotic statements because that was me!-- what the heck was I thinking?

I know it has to be frustrating for @AlexLockard not to be able to get a straight answer to a seemingly simple question.  But for me, the question is based on an impossibility (e.g., how hard do I push the gas pedal in my Camry to go 500 mph?) 

[Mukluk4]

---

@AlexLockard wrote:
Right, but I guess I was hoping to see if a certain threshold of steps was
needed for weight maintenance and loss, vs. gain. My motivation to get out
and make sure I hit my step goal has waned definitely over the last couple
of years. The feedback I get from my fitbit doesn't provide anywhere near
the level of motivation now that it did in the first year or two. I need to
see that if I only worked a little harder, I would catapult myself back
into weight loss territory. For quite a while, i was very consistent on
getting 12k - 15k (and many days, 20k+) steps daily, but now it's more like
5k to 10k probably. Over an extended period of time, I would think that has
to be a determining factor. But with such a huge data set (7-8 years of
step and floor data), I was hoping a statistician could tease out some
inferences.

---

Stop counting steps and count calories burned and gained (as an example, my goal is to burn 3.8K calories a day).  It's not so much the quantity of steps but the quality of steps.  There are also tons of great calorie burners that aren't step based.

[jxd]

There is no correlation without controlling your caloric intake.  Sorry, looks like you have to "exercise more and eat less"

[GettingfitinSC]

 "For quite a while, I was very consistent on
getting 12k - 15k (and many days, 20k+) steps daily, but now it's more like
5k to 10k probably. "

Hi, I'm not a statistician, but I think you answered your own question. During the time that you were consistent with getting 12K-15K, was that when you were losing weight? If so, that would be the level I would start with to start losing weight again. If you don't want to change your eating, just start getting more steps like you did when you first lost the weight. I hope you get the answer you're looking for. 🙂 

[AlexLockard]

That would seem obvious, I suppose. I have lost the motivation I once had
to get out there super consistently, and was hoping that I could find the
minimum number needed to maintain, and then how many more to lose weight.
After a car accident in 2013, my lower back has given me quite a bit of
pain, so I can't walk/hike/climb like I used to. I still get after it, but
not with quite the same level of intensity for the same duration. The last
four years have been a steady increase.

[jxd]

You can lose fat even if you stayed in bed all day as long as you don't eat more calories than you burn.  There's no secret, it's all down to CICO.  Quit thinking you can out exercise a bad diet.

[Julia_G]

@jxd, to SOME extent it is true that you need to think broadly in terms of calories in / calories out.  But much recent evidence confirms that all calories are not created equal.

See, just for example, this information from the Harvard Health Letter.

[AlexLockard]

I never said I had a "bad diet," but a lot of people here seemed to assume that. What has been more or less fixed over all the years is my eating habits, while the major variable has been the number of steps taken, whether measured daily, weekly, monthly, etc. I guess I'll just dive into the dataset and see if I can find anything meaningful in the correlations. I was hoping someone might have already done this. I know there's no easy answer, but was hoping that my (very large) data would be of statistical interest to someone with the expertise to analyze it.

[GershonSurge]

@AlexLockard

A couple times while losing weight, I plotted the total calorie deficit on the x-axis and total weight loss on the y-axis. After about 25 days, the R=squared was about .98. This means 98% of the weight loss was predicted by CICO. 

The next step is calibrating the Fitbit. Divide the total calorie deficit by the total weight loss and you will find the total calories needed to lose a pound. The difference between this number and 3500 is the Fitbit error or the difference in your body. You can think of it either way. 

It doesn't work well for maintenance or for trying to gain weight. It works well regardless of the number of steps a day. 

To translate this into an easily usable method for losing weight, start with a 750 calorie deficit (stay at or above BMR) for a few weeks. If you are not losing weight, increase it to a 1,000 calorie deficit. If you are losing more than about 1.5 pounds a week, decrease it to a 500 calorie deficit. 

If you don't want to count calories and are trying to lose weight, you can wait until you feel real hunger pangs before you eat. Just eat enough to stop the hunger pangs. If you are trying to gain weight or maintain, eat as soon as you feel you can. Don't stuff yourself in either case. 

I haven't even found a weak association between step count and weight loss. 

For people who don't like to count calories, there is another way to succeed, but that's beyond the scope of this topic.

[Chiholly]

I am a very soon to be 64 year old woman.  I try to be very active.  I have no health problems.  I've found that, as painful as it sounds, estimation of activity and calories of food eaten is simply a large no-no.  I lose weight consistently when I set my goals and stick to them.  I am at an age where I have to restrict my caloric intake to no more than 1,600 calories a day - in order to lose weight, I have to stay around 1,200 a day.  I have a step goal of 10,000, but usually walk 12,000.  I have an activity goal of 45 minutes a day, plus 10 flights of stairs.  If I do not meet my goals, I will not lose weight.  I also force myself to track my calories on my Fitbit log.  It is a terrible pain, but I have to do it.  I've developed a plan over time.  If I stick to it, all is well.  If not, the weight goes on.  Good luck!

[AlexLockard]

Thanks!