I had the same problem. I have not found a solution either. Everything does appear fine up to that moment. Then somehow the Flip function makes the second point go in a polarizing direction from the first point.

Could you post a screenshot of the polarizing direction you are talking about so I can see whats going on and help. Have you taken a look to the exercise definition to see if everything is plugged in correctly?

Sure. In the First image, the two are exactly as you point out--in the same coordinate position and moving along in the same direction. In the Second image, and once the Flip component is applied, the point on the left moves immediately to an opposing position and moves counter to the point on the right (See blue circles).

It seems there is a problem where the seam of the curve(Start/End of the circle) is, are you sure you did the previous steps of setting up properly the seam of the circle so when you evaluate the curve starts from the same point.

I will go ahead and take a look at it, but not today. Hopefully I will have an update soon. Stay tuned for a proper answer...