Ugh, Anyone want to write some numerical code for me?

The stupid software engineer has written some crap polynomial fitting routine with 6th order fit for the shorter orbit segments  and order 16! for longer segments needed to go with ionosphere stuff.  I'm wondering how it is not making all the missions have crappy results instead of just the one that got noticed.

He did this to get a chi-square value.  Now I'm pretty sure it is not too hard to get a chi-square while using a sensible spline fit instead of a crazy high-order polynomial but I do not know how to code it up quickly.  I hate coding stuff like that.