BODY FAT

This post is actually about data again, but I figured if I labeled it Body Fat you'd be more likely to open the post and read it. Am I right?

I got a new scale.  It's one of those Tanita scales that supposedly measures your weight, your hydration status and your body fat.

Okay, I know. I know that body fat scales do not work. But wait! Before you decide I'm totally dumb for wasting my money on one, let me explain my thinking. I figured the scale could provide me a ball park idea as to my body fat, and then I could try to whittle that number down over the next months until IM CDA. So, if the scale said 20% body fat, I would just try to get 19%. Who cares whether the scale is accurate; what matters is that the number decrease over time, not increase. That makes sense, right? Let me also explain that I am interested in my body fat percentage because all of these data measures I'd like to employ rely on having a good estimate of my lean body mass. To determine lean body mass I need to know my body fat percentage. Capisce?

I got the scale about three weeks ago. In that time my body fat has ranged, according to the scale, from 14.5% to 24.5%. That is no small range. What I've realized is that the percent of body fat the scale believes I have is related to the hydration status it thinks I have. So, if I'm dehydrated according to the scale, my weight might be less, but my body fat is high. If I am fully hydrated my weight is higher, but the body fat reading is lower. 

The problem here is that I don't think the scale's determination of hydration status is even right. There have been times I have gotten on the scale after hydrating like crazy for 24 hours, and it reads I am dehydrated. Likewise, there have been times that I know I am dehydrated, and it reads that my hydration status is awesome. The way the scale reads hydration seems to be more correlated with whether I am retaining water because, say, I've had something super salty the night before or because I'm a few days out from getting my period. This makes me wonder: what is the relationship between hydration status and retaining water? Are they the same? If I am holding onto water before my period, does that mean I'm hydrated? That must not be right... 

Anyone?

Okay, so why do I bring this all up?

First, because I want you to save your money and abstain from buying a bod fat scale, because they read... something, but I don't know exactly what. It's not body fat, I do know that. The margin of error between 14.5% an 24.5% over a three week span is a little too great to put any stock at all in the scale's reading. I think my body fat is somewhere in between those two numbers, but who knows exactly where.  But also, I want to bring up this frustration I have, again, with data.

In my post a few days ago I brought up the fact that the accuracy of the Training Peaks' (or WKO's) performance management chart relies upon accurate threshold numbers on the bike and run. If the FTP  (functional threshold power) is overestimated (more generally the problem than under-estimation) than the chart will be skewed, and will not accurately determine fatigue levels (not that this can be perfectly charted even with a "correct" FTP estimation), and hence things like planning peak weeks and taper are harder to manage correctly. 

This problem is also true for the determination of lean body mass, which is the key piece of data in determining both fueling needs when training and what one's appropriate weight might be for executing, say, an Ironman. For example, it's thought that a BMI of about  20-21 is the "right" number for a female  who is trying to execute (well) an Ironman. That number is different for a female runner, whose ideal BMI would be slightly less than that, or a swimmer, who BMI might be greater. The problem is, of course, that BMI is determined by height and weight--but it doesn't distinguish the type of weight the athlete is carrying.  A muscular woman might have a higher BMI than a more waif-like athlete, but that might be because the "leaner" athlete has less muscle and a more body fat than the heavier athlete. This is corrected by determining weight of lean body mass rather than simply weight, and lean body mass is determined with a knowledge of one's body fat. 

Which, it seems to me, is fairly impossible to reliably determine. 

There are other methods of determining body fat, but all of them have rather large margins of error--especially the use of calipers--which is how most athletes determine their body fat. 

So we use these tools--like the performance management chart--and measures--like lean body mass--which rely on one key piece of data (FTP, BMI), which is often -- well, wrong--which renders the whole tool basically useless.  

This winter I decided I would try to be more consistent in my use of measures to determine my progress and my atheltes' progress. But the door that continually slams in my face is that these measures are based on unreliable data. You name it--determining fueling for training and racing, assigning correct pacing strategies, executing training based on power or heart rate, managing body composition--all of these things rely upon having correct data--and as far as I can tell, the data we use is nearly always flawed data.

So, I am agnostic.

When I said I want to believe in data like I want to believe in God, I truly meant it. The more I know, the more I understand there are no perfect measures--there is nothing that can be relied upon as more than a generality that may or may not help one to manage training--or anything else in life. 

When I was younger I think I believed that those who liked mathematics liked it because there is little ambiguity. In math, I reasoned, answers are right or wrong.  But the more I understand, the more I realize that those who are truly involved in math know that there are really never any precise, exact, correct answers. The mathematicians must be the most agnostic of us all! Is this true math people? If you are a math person... please speak to this. 

Conversely, we language arts type people, while believing we deal in agnosticism, read because, I think, we want to make sense of the world--we want to draw conclusions that give us peace and answers. 

I still use the data. 

I have been absolutely zealous about my assignment of TSS to my workouts and to those of my athletes (as in, I assign scores to things that are not scored, I monitor scoring, try to keep "accurate" FTP and HR data plugged in etc). And I do things like buy totally useless body fat scales to help me to track my hydration and body fat, and I try to eat the exact amount of carbs for my weight, and the right amount of salt, and the write amount of liquid per hour while training... and so on---and I try to get my athletes to do the same.

And then I think, why? 

The measures could be so off... Is this even useful? 

What is that expression....?
I looked it up: George Box-- "All models are wrong--but some are useful."

But we rely on this models to be right--or at the very least to provide us insight. And with a body fat reading between 14.5% and 24.5%--can we even have insight? With an FTP number that is too high (Can you REALLY do that wattage for an hour? I will put money on the fact that you can't...) can we even specify with any kind of certainty what is the correct TSS allotment to try to hit on the bike in an Ironman?

and so on.
I actually wrote a post on this in like 2009. Here it is. It's better written than this one, I think. Just goes to show that even though I have always claimed agnosticism I keep trying to have faith.
I really want to believe in God.