Best fit polynomial 5th degree - Trendline data points differ towards the end of my data

Hi There,

My name is Morris, and I am having trouble to get my 5th order polynomial trendline to fit observations when I back-calculate, especially towards the end of my data. I did the trendline with excel, and got the constants. See attached file. However, when I use these constants to back-calculate the trendline (Orange data points), I do not get the polynomial that I observe in the graph (Blue dotted line)?

Secondly, where can I get tabular data points of trendlines in excel? 

Regards,

Morris