On the first graph I think if you could expand the scale on the vertical axis which should make it more readable. I could barely find my data in the mix of several others. Whatever the scale was which you used on the web page www.sethroberts.net
seems to discriminate between the individual's data much better.
On the graph with the "normalized" data, I'm not sure why you plotted the average as opposed to plotting regression lines. (Perhpas averaging was the easiest to do for a first cut.) Actually, since the rate of weight loss is generally proportial the weight, an exponential regression would perhaps give a better fit of the data. I'm not sure how difficult this is to do with your software, but in some later excel systems this capability is present.
I'm not sure how far you intend to push the analysis of this data. Or perhpas your intent is merely for display purposes for others to see. A statistician could analize this data until the proverbial cows come home.
P.S. This is my 100th post. I think something special happens at this point. Maybe I'll turn into a skinny prince rather than a chubby frog.