Tuesday, November 18, 2014

Results of analysis of how I ate this summer

Between June 9 and August 22, I lost about 4 pounds, more or less steadily (lots of ups and downs, but the overall trend was definitely down). And I had been fairly strict about logging in LoseIt over that period, so I had reasonably complete datasets for nutrition as well.  So I did a little basic regression and correlation by date, to see if I could figure out what was or was not working for me in that period.

I'm sure there's a way to post charts here, but probably not via email, and that's about it for me right now. To summarize the results:
There appears to be no correlation at all between calories, fat grams, or protein grams consumed one day and the amount of weight lost or gained the next morning. None. Those charts have a flatlined linear regression trendline. Between carb % or carb grams consumed one day and the amount of weight lost or gained the next morning, there's a slight negative correlation: that is, I lost mildly more weight the day after I'd consumed higher amounts of carbs than I did the day after I'd consumed lower amounts of carbs. In a general sense, that is. I also lost weight the day after I'd consumed very low carbs, and gained after consuming high amounts.

Now, I have always noticed that my weight change seems to lag a day or two behind any spikes (positive or negative) in eating, and I may try shifting the data for intake another day against the weight changes, see if that shows anything. But I thought it was interesting that nothing I did this summer appears to have had any real effect on my weight loss success or failure - the control was, at least superficially, not in my hands at all.

Bother.

No comments:

Post a Comment