Let's go back to page 6 of this thread on October 31, 2012 and see what I had to say:
"323 this morning and one of today's projects is punching new holes in my belt. Always a good thing. "
Okay not exactly inspiring. But let's pretend I used the 50 data points I had collected (IE, my morning weigh-ins) from August 3 to October 31 to make my model. The graph of it would look like this:
View attachment 20800
With the red lines being the data up until that point. That was 201 days ago. How would that model have done for the next 201 days (IE not changing the model at all from there like I normally do as more data comes in)? Let's see:
View attachment 20801
I'm actually quite impressed. Way back on Halloween the model predicted that on today's date my weight would be 256.4. It is in fact 254. That ain't a half bad guess particularly since last Friday it 256.5.
But as you can see, it looks like for the two months or so I've pretty consistently been under that model's prediction. In any event, this definitely qualifies as a "proof of concept" for the model. As long as I continued to stick to the plan (which I did) my weight over time has dropped in a fairly orderly and obviously predictable fashion. I find this important to me, because it really does help put the occasional ups and downs into perspective and gives me a solid idea of where I'm really at.
Here is a link to where you can download the model:
If there's some sort of problem getting it working, let me know. Instructions are on there.